Digital Digest -> Software -> x265 -> History & Old Downloads

x265 -> Version History

This is the version history page for x265. You can return to the main page for this software by clicking here.

Index:


Version 2.5
  • Release Date: Jul 14, 2017
  • Encoder enhancements
    • Improved grain handling with --tune grain option by throttling VBV operations to limit QP jumps.
    • Frame threads are now decided based on number of threads specified in the --pools, as opposed to the number of hardware threads available. The mapping was also adjusted to improve quality of the encodes with minimal impact to performance.
    • CSV logging feature (enabled by --csv) is now part of the library; it was previously part of the x265 application. Applications that integrate libx265 can now extract frame level statistics for their encodes by exercising this option in the library.
    • Globals that track min and max CU sizes, number of slices, and other parameters have now been moved into instance-specific variables. Consequently, applications that invoke multiple instances of x265 library are no longer restricted to use the same settings for these parameter options across the multiple instances.
    • x265 can now generate a seprate library that exports the HDR10+ parsing API. Other libraries that wish to use this API may do so by linking against this library. Enable ENABLE_HDR10_PLUS in CMake options and build to generate this library.
    • SEA motion search receives a 10% performance boost from AVX2 optimization of its kernels.
    • The CSV log is now more elaborate with additional fields such as PU statistics, average-min-max luma and chroma values, etc. Refer to documentation of --csv for details of all fields.
    • x86inc.asm cleaned-up for improved instruction handling.
    API changes
    • New API x265_encoder_ctu_info() introduced to specify suggested partition sizes for various CTUs in a frame. To be used in conjunction with --ctu-info to react to the specified partitions appropriately.
    • Rate-control statistics passed through the x265_picture object for an incoming frame are now used by the encoder.
    • Options to scale, reuse, and refine analysis for incoming analysis shared through the x265_analysis_data field in x265_picture for runs that use --analysis-reuse-mode load; use options --scale, --refine-mv, --refine-inter, and --refine-intra to explore.
    • VBV now has a deterministic mode. Use --const-vbv to exercise.
    Bug fixes
    • Several fixes for HDR10+ parsing code including incompatibility with user-specific SEI, removal of warnings, linking issues in linux, etc.
    • SEI messages for HDR10 repeated every keyint when HDR options (--hdr-opt, --master-display) specified.
Version 2.4+98 Nightly
  • Release Date: Jul 8, 2017
Version 2.4+97 Nightly
  • Release Date: Jul 7, 2017
Version 2.4+96 Nightly
  • Release Date: Jul 1, 2017
Version 2.4+90 Nightly
  • Release Date: Jun 30, 2017
Version 2.4+89 Nightly
  • Release Date: Jun 29, 2017
Version 2.4+87 Nightly
  • Release Date: Jun 24, 2017
Version 2.4+74 Nightly
  • Release Date: Jun 22, 2017
Version 2.4+73 Nightly
  • Release Date: Jun 20, 2017
Version 2.4+68 Nightly
  • Release Date: Jun 16, 2017
Version 2.4+61 Nightly
  • Release Date: Jun 14, 2017
Version 2.4+41 Nightly
  • Release Date: Jun 13, 2017
Version 2.4+37 Nightly
  • Release Date: Jun 7, 2017
Version 2.4+36 Nightly
  • Release Date: Jun 6, 2017

Version 2.4+28 Nightly
  • Release Date: May 27, 2017
Version 2.4+27 Nightly
  • Release Date: May 26, 2017
Version 2.4+25 Nightly
  • Release Date: May 20, 2017
Version 2.4+22 Nightly
  • Release Date: May 16, 2017
Version 2.4+13
  • Release Date: May 12, 2017

Version 2.4+6 Nightly
  • Release Date: May 4, 2017
    • Ma0 cmake: set '-std=gnu++11' for GCC if ENABLE_DYNAMIC_HDR10 is on 2017-04-20
    • Ma0 silence MSVC warning C4334 and GCC warnings for '-std=gnu++11' 2017-04-20
    • Ma0 CLI: small fixes for dhdr10 2017-04-20
    • Bhavna Hariharan dhdr10:fix linux build error 2017-04-24
Version 2.4
  • Release Date: Apr 23, 2017
  • Encoder enhancements
    • HDR10+ supported. Dynamic metadata may be either supplied as a bitstream via the userSEI field of x265_picture, or as a json jile that can be parsed by x265 and inserted into the bitstream; use --dhdr10-info to specify json file name, and --dhdr10-opt to enable optimization of inserting tone-map information only at IDR frames, or when the tone map information changes.
    • Lambda tables for 8, 10, and 12-bit encoding revised, resulting in significant enhancement to subjective visual quality.
    • Enhanced HDR10 encoding with HDR-specific QP optimzations for chroma, and luma planes of WCG content enabled; use --hdr-opt to activate.
    • Ability to accept analysis information from other previous encodes (that may or may not be x265), and selectively reuse and refine analysis for encoding subsequent passes enabled with the --refine-level option.
    • Slow and veryslow presets receive a 20% speed boost at iso-quality by enabling the --limit-tu option.
    • The bitrate target for x265 can now be dynamically reconfigured via the reconfigure API.
    • Performance optimized SAO algorithm introduced via the --limit-sao option; seeing 10% speed benefits at faster presets.
    API changes
    • x265_reconfigure API now also accepts rate-control parameters for dynamic reconfiguration.
    • Several additions to data fields in x265_analysis to support --refine-level: see x265.h for more details.
    Bug fixes
    • Avoid negative offsets in x265 lambda2 table with SAO enabled.
    • Fix mingw32 build error.
    • Seek now enabled for pipe input, in addition to file-based input
    • Fix issue of statically linking core-utils not working in linux.
    • Fix visual artifacts with --multi-pass-opt-distortion with VBV.
    • Fix bufferFill stats reported in csv.
Version 2.3+40 Nightly
  • Release Date: Apr 21, 2017
Version 2.3+33 Nightly
  • Release Date: Apr 15, 2017
Version 2.3+32 Nightly
  • Release Date: Apr 13, 2017
    • Ma0 CLI: informs if '--ssim-rd' is used 11 hours ago
    • Gopi Satykrishna Akisetty rcStats: add more fields to rcStats 2017-03-28
    • Ashok Kumar Mishra Improved sao implementation by limiting sao types 8 hours ago
    • Aruna Matheswaran Add dynamic rate-control reconfiguration 2017-02-22
    • Aruna Matheswaran Fix artifacts issue in multi-pass-opt-distortion with VBV 2017-03-27
    • Pradeep Ramachandran cmake: Fix static-link-crt not working in linux 5 hours ago
    • Bhavna Hariharan test: disable limitTU with refine-level 10 2017-03-29
    • Pradeep Ramachandran doc: Update copyright information 13 hours ago
    • Bhavna Hariharan param: add limit-tu to slower, veryslow presets - upto 20% increase in FPS
    • 2017-03-10
    • Pradeep Ramachandran comment: Fix equation for new 8-bit lambda2 table 15 hours ago
    • Pradeep Ramachandran lambda: New 8-bit lambda2 table. Considerably improves subjective visual quality. 2017-03-13
    • Ma0 fix maxSlices limit checking 2017-03-10
    • Ma0 CLI: fix '--seek' option for pipe input 2017-03-10
    • Pradeep Ramachandran Merge with stable 6 hours ago
    • Radhakrishnan Fix : mingw32 build error stable 2017-03-09
    • Kavitha Sampath multi-level refinement: fix typo and missing space 2017-03-08
    • Kavitha Sampath regression: include test for refine levels 2017-02-27
    • Gopi Satykrishna Akisetty CLI: Remove redundant cli option 'capture-csp' 23 hours ago
    • Ma0 SAO: avoid negative indexes in 'x265_lambda2_tab' table 2017-02-16
    • Gopi Satykrishna Akisetty fix the chroma offset computation for HDR/WCG content 23 hours ago
    • Gopi Satykrishna Akisetty Add Luma and Chroma offsets for HDR/WCG content 2017-02-01
    • Ma0 silence GCC 7 warnings 2017-02-11
    • Ma0 update the year in output stream header 2017-02-11
    • Ma0 CLI: allow Unicode filenames (Windows) for 'scaling-list', 'lambda-file' and 'analysis-file' 2017-02-11
Version 2.3+29 Nightly
  • Release Date: Apr 8, 2017
    • Bhavna Hariharan test: disable limitTU with refine-level 10 2017-03-29
Version 2.3+28 Nightly
  • Release Date: Apr 2, 2017
    • Aruna Matheswaran Fix artifacts issue in multi-pass-opt-distortion with VBV 2017-03-27
    • Pradeep Ramachandran cmake: Fix static-link-crt not working in linux 5 hours ago
    • Bhavna Hariharan test: disable limitTU with refine-level 10 2017-03-29
    • Pradeep Ramachandran doc: Update copyright information 13 hours ago
    • Bhavna Hariharan param: add limit-tu to slower, veryslow presets - upto 20% increase in FPS
Version 2.3+27 Nightly
  • Release Date: Mar 31, 2017
    • Pradeep Ramachandran cmake: Fix static-link-crt not working in linux 5 hours ago
    • Bhavna Hariharan test: disable limitTU with refine-level 10 2017-03-29
    • Pradeep Ramachandran doc: Update copyright information 13 hours ago
    • Bhavna Hariharan param: add limit-tu to slower, veryslow presets - upto 20% increase in FPS
Version 2.3+25 Nightly
  • Release Date: Mar 28, 2017
    • Pradeep Ramachandran doc: Update copyright information 13 hours ago
Version 2.3+24 Nightly
  • Release Date: Mar 21, 2017
    • Bhavna Hariharan param: add limit-tu to slower, veryslow presets - upto 20% increase in FPS
    • 2017-03-10
    • Pradeep Ramachandran comment: Fix equation for new 8-bit lambda2 table 15 hours ago
    • Pradeep Ramachandran lambda: New 8-bit lambda2 table. Considerably improves subjective visual quality. 2017-03-13

Version 2.3+23 Nightly
  • Release Date: Mar 18, 2017
    • Pradeep Ramachandran comment: Fix equation for new 8-bit lambda2 table 15 hours ago
    • Pradeep Ramachandran lambda: New 8-bit lambda2 table. Considerably improves subjective visual quality. 2017-03-13
Version 2.3+22 Nightly
  • Release Date: Mar 17, 2017
    • Ma0 fix maxSlices limit checking 2017-03-10
    • Ma0 CLI: fix '--seek' option for pipe input 2017-03-10
    • Pradeep Ramachandran Merge with stable 6 hours ago
    • Radhakrishnan Fix : mingw32 build error stable 2017-03-09
    • Kavitha Sampath multi-level refinement: fix typo and missing space 2017-03-08
    • Kavitha Sampath regression: include test for refine levels 2017-02-27
Version 2.3+18 Nightly
  • Release Date: Mar 11, 2017
    • Kavitha Sampath multi-level refinement: fix typo and missing space 2017-03-08
    • Kavitha Sampath regression: include test for refine levels 2017-02-27
Version 2.3+8 Nightly
  • Release Date: Feb 21, 2017
    • Ma0 SAO: avoid negative indexes in 'x265_lambda2_tab' table 2017-02-16
    • Gopi Satykrishna Akisetty fix the chroma offset computation for HDR/WCG content 23 hours ago
    • Gopi Satykrishna Akisetty Add Luma and Chroma offsets for HDR/WCG content 2017-02-01
    • Ma0 silence GCC 7 warnings 2017-02-11
    • Ma0 update the year in output stream header 2017-02-11
    • Ma0 CLI: allow Unicode filenames (Windows) for 'scaling-list', 'lambda-file' and 'analysis-file' 2017-02-11
Version 2.3+7 Nightly
  • Release Date: Feb 18, 2017
    • Gopi Satykrishna Akisetty fix the chroma offset computation for HDR/WCG content 23 hours ago
    • Gopi Satykrishna Akisetty Add Luma and Chroma offsets for HDR/WCG content 2017-02-01
    • Ma0 silence GCC 7 warnings 2017-02-11
    • Ma0 update the year in output stream header 2017-02-11
    • Ma0 CLI: allow Unicode filenames (Windows) for 'scaling-list', 'lambda-file' and 'analysis-file' 2017-02-11
Version 2.3
  • Release Date: Feb 15, 2017
  • Encoder enhancements
    • New SSIM-based RD-cost computation for improved visual quality, and efficiency; use --ssim-rd to exercise.
    • Multi-pass encoding can now share analysis information from prior passes (in addition to rate-control information) to improve performance and quality of subsequent passes; to your multi-pass command-lines that use the --pass option, add --multi-pass-opt-distortion to share distortion information, and --multi-pass-opt-analysis to share other analysis information.
    • A dedicated thread pool for lookahead can now be specified with --lookahead-threads.
    • option:–dynamic-rd dynamically increase analysis in areas where the bitrate is being capped by VBV; works for both CRF and ABR encodes with VBV settings.
    • The number of bits used to signal the delta-QP can be optimized with the --opt-cu-delta-qp option; found to be useful in some scenarios for lower bitrate targets.
    • Experimental feature option:–aq-motion adds new QP offsets based on relative motion of a block with respect to the movement of the frame.
    API changes
    • Reconfigure API now supports signalling new scaling lists.
    • x265 application’s csv functionality now reports time (in milliseconds) taken to encode each frame.
    • --strict-cbr enables stricter bitrate adherence by adding filler bits when achieved bitrate is lower than the target; earlier, it was only reacting when the achieved rate was higher.
    • --hdr can be used to ensure that max-cll and max-fall values are always signaled (even if 0,0).
    Bug fixes
    • Fixed incorrect HW thread counting on MacOS platform.
    • Fixed scaling lists support for 4:4:4 videos.
    • Inconsistent output fix for --opt-qp-pss by removing last slice’s QP from cost calculation.
    • VTune profiling (enabled using ENABLE_VTUNE CMake option) now also works with 2017 VTune builds.
Version 2.2+35
  • Release Date: Feb 11, 2017
    • Pradeep Ramachandran Merge with default; pref for 2.3 stable 51 minutes ago
    • Pradeep Ramachandran doc: Label aq-motion as experimental feature 55 minutes ago
    • Pradeep Ramachandran cli: Enable new --hdr option to force signaling HDR parameters Useful when there is a desire to signal --max-cll 0,0 in the bitstream. 6 hours ago
    • Praveen Tiwari Threadpool-fix: correctly detect MacOS 4 hours ago
    • Pradeep Ramachandran cmake: Enable detecting vtune-2017 3 hours ago
    • Bhavna Hariharan complexAnalysis: clean up 9 hours ago
    • Bhavna Hariharan complexAnalysis: add test command lines 2017-01-25
    • Bhavna Hariharan complexAnalysis: increase analysis for areas capped by VBV 2017-01-24
    • Bhavna Hariharan cli: add options to support complex-analysis 2017-01-24
Version 2.2+33
  • Release Date: Feb 10, 2017
    • Pradeep Ramachandran cli: Enable new --hdr option to force signaling HDR parameters Useful when there is a desire to signal --max-cll 0,0 in the bitstream. 6 hours ago
    • Praveen Tiwari Threadpool-fix: correctly detect MacOS 4 hours ago
    • Pradeep Ramachandran cmake: Enable detecting vtune-2017 3 hours ago
    • Bhavna Hariharan complexAnalysis: clean up 9 hours ago
    • Bhavna Hariharan complexAnalysis: add test command lines 2017-01-25
    • Bhavna Hariharan complexAnalysis: increase analysis for areas capped by VBV 2017-01-24
    • Bhavna Hariharan cli: add options to support complex-analysis 2017-01-24
Version 2.2+32
  • Release Date: Feb 8, 2017
    • Praveen Tiwari Threadpool-fix: correctly detect MacOS 4 hours ago
    • Pradeep Ramachandran cmake: Enable detecting vtune-2017 3 hours ago
    • Bhavna Hariharan complexAnalysis: clean up 9 hours ago
    • Bhavna Hariharan complexAnalysis: add test command lines 2017-01-25
    • Bhavna Hariharan complexAnalysis: increase analysis for areas capped by VBV 2017-01-24
    • Bhavna Hariharan cli: add options to support complex-analysis 2017-01-24
Version 2.2+31
  • Release Date: Feb 4, 2017
    • Pradeep Ramachandran cmake: Enable detecting vtune-2017 3 hours ago
    • Bhavna Hariharan complexAnalysis: clean up 9 hours ago
    • Bhavna Hariharan complexAnalysis: add test command lines 2017-01-25
    • Bhavna Hariharan complexAnalysis: increase analysis for areas capped by VBV 2017-01-24
    • Bhavna Hariharan cli: add options to support complex-analysis 2017-01-24

Version 2.2+30
  • Release Date: Jan 31, 2017
    • Bhavna Hariharan complexAnalysis: clean up 9 hours ago
    • Bhavna Hariharan complexAnalysis: add test command lines 2017-01-25
    • Bhavna Hariharan complexAnalysis: increase analysis for areas capped by VBV 2017-01-24
    • Bhavna Hariharan cli: add options to support complex-analysis 2017-01-24
Version 2.2+29
  • Release Date: Jan 28, 2017
    • Bhavna Hariharan complexAnalysis: add test command lines 2017-01-25
    • Bhavna Hariharan complexAnalysis: increase analysis for areas capped by VBV 2017-01-24
    • Bhavna Hariharan cli: add options to support complex-analysis 2017-01-24

Version 2.2+26
  • Release Date: Jan 27, 2017
    • Aruna Matheswaran Add filler bits when frame bits < vbv target in strict-cbr 2017-01-20
    • Gopi Satykrishna Akisetty add support for aq-motion even when aq-mode is disabled 2017-01-20
    • Divya Manivannan pps: remove last slice qp from pps qp calculation18 hours ago
    • Vignesh Include elapsed time for encoding each frame in CSV 2017-01-23
Version 2.2+25
  • Release Date: Jan 25, 2017
    • Gopi Satykrishna Akisetty add support for aq-motion even when aq-mode is disabled 2017-01-20
    • Divya Manivannan pps: remove last slice qp from pps qp calculation18 hours ago
    • Vignesh Include elapsed time for encoding each frame in CSV 2017-01-23
Version 2.2+23
  • Release Date: Jan 25, 2017
    • Vignesh Include elapsed time for encoding each frame in CSV 2017-01-23
Version 2.2+22
  • Release Date: Jan 6, 2017
    • Ashok Kumar Mishra doc: updated rst doc about param ssim-rd 21 hours ago
    • Ashok Kumar Mishra ssim-rd: fix for high bit depth 23 hours ago
    • Ashok Kumar Mishra ssim-rd: fix for transform skip 2017-01-03
    • Vignesh Scalinglists support for 4:4:4 videos 2016-12-28
    • Pradeep Ramachandran Merge with stable 5 hours ago
    • Vignesh SEA motion search: Set m_reconRowFlag only after completing integral calculation Branches stable 2017-01-03

Version 2.2+17
  • Release Date: Dec 31, 2016
    • Pradeep Ramachandran warn: Fix warning from x265cli.h 4 hours ago
    • Pradeep Ramachandran warn: Fix build warning from rdcost.h 4 hours ago
    • Aruna Matheswaran Fix memory leak in analysis-multi-pass stats file renaming 22 hours ago
    • Santhoshini S correct slicetype passed to writeAnalysis2PassFile. Fixes memory leak 2016-12-29
Version 2.2
  • Release Date: Dec 27, 2016
  • Encoder enhancements
    • Enhancements to TU selection algorithm with early-outs for improved speed; use --limit-tu to exercise.
    • New motion search method SEA (Successive Elimination Algorithm) supported now as :option: –me 4
    • Bit-stream optimizations to improve fields in PPS and SPS for bit-rate savings through --[no-]opt-qp-pps, --[no-]opt-ref-list-length-pps, and --[no-]multi-pass-opt-rps.
    • Enabled using VBV constraints when encoding without WPP.
    • All param options dumped in SEI packet in bitstream when info selected.
    • x265 now supports POWERPC-based systems. Several key functions also have optimized ALTIVEC kernels.
    API changes
    • Options to disable SEI and optional-VUI messages from bitstream made more descriptive.
    • New option --scenecut-bias to enable controlling bias to mark scene-cuts via cli.
    • Support mono and mono16 color spaces for y4m input.
    • --min-cu-size of 64 no-longer supported for reasons of visual quality (was crashing earlier anyways.)
    • API for CSV now expects version string for better integration of x265 into other applications.
    Bug fixes
    • Several fixes to slice-based encoding.
    • --log2-max-poc-lsb‘s range limited according to HEVC spec.
    • Restrict MVs to within legal boundaries when encoding.
Version 2.1+74
  • Release Date: Dec 24, 2016
    • Santhoshini S cudata: init mvmemblock with zero stable 2016-12-22
    • Aarthi Thirumalai Set up separate threadpool for lookahead The user can allocate specific number of threads for lookahead by specifying --lookahead-threads <val>, creating a separate lookahead pool for these threads. This will improve performance when lookahead becomes a bottle neck. THe threads for lookahead must be ideally less than half the total number of available worker threads. 2016-12-09
    • Divya Manivannan param: remove filenames from info stable 2016-12-16
    • Pradeep Ramachandran Merge with default; prep for 2.2 stable 2016-12-15

Version 2.1+71
  • Release Date: Dec 21, 2016
    • Aarthi Thirumalai Set up separate threadpool for lookahead The user can allocate specific number of threads for lookahead by specifying --lookahead-threads <val>, creating a separate lookahead pool for these threads. This will improve performance when lookahead becomes a bottle neck. THe threads for lookahead must be ideally less than half the total number of available worker threads. 2016-12-09
    • Divya Manivannan param: remove filenames from info stable 2016-12-16
    • Pradeep Ramachandran Merge with default; prep for 2.2 stable 2016-12-15
Version 2.1+70
  • Release Date: Dec 13, 2016
    • Bhavna Hariharan fix analysis save/load bug in limitTU 2016-12-09
Version 2.1+69
  • Release Date: Dec 9, 2016
    • Divya Manivannan rc: frameSizePlanned is calculated from the predictors for vbv case 2016-05-26
    • Vignesh SEA motion search: Copy integral planes for ANALYSIS LOAD 2016-12-07
    • mahesh pittala csv: modify csv api 2016-12-07
Version 2.1+65
  • Release Date: Dec 5, 2016
    • Vignesh Warn no support of pme and pmode in SEA motion search 2016-12-02

Version 2.1+64
  • Release Date: Dec 2, 2016
    • Vignesh Fix memory leak in fpelMvCosts 2016-11-30
Version 2.1+63
  • Release Date: Dec 1, 2016
    • Vignesh SEA motion search Implementation 2016-11-28
Version 2.1+62
  • Release Date: Nov 30, 2016
    • Divya Manivannan Fix source width and height in info 2016-11-25
    • Zheng Wang doc: Improve rst doc and help doc about param multi-pass-opt-rps 2016-11-25
Version 2.1+60
  • Release Date: Nov 25, 2016
    • M CHEN fix logic timing bug 2016-11-23
Version 2.1+59
  • Release Date: Nov 24, 2016
    • Bhavna Hariharan tests: update command lines to cover limitTU 3 and 4 2016-11-16
    • Bhavna Hariharan limitTU : use neighbouring CUs' TU depth to limit 1st subTU's depth 2016-11-18
    • Bhavna Hariharan limitTU : use spatial and temporal CUs' TU depth to limit recursion 2016-11-18
    • Bhavna Hariharan limitTU : modify condition for limitTU 1 and 2 2016-11-18
Version 2.1+55
  • Release Date: Nov 20, 2016
    • Praveen Tiwari encoder.cpp: print reconfigure params for debug purpose 2016-11-17
    • Chen cleanup debug code 2016-11-16 
    • Divya Manivannan 6737486 rc: change stat read options 2016-11-16
    • Divya Manivannan Add all the param options to bitstream when info is enabled 2016-11-16

Version 2.1+49
  • Release Date: Nov 16, 2016
    • Chen fix logic bug in weight module (the latest row was not weighted) 13 hours ago
    • Divya Manivannan Add param options to bitstream when info is enabled 2016-11-11
Version 2.1+47
  • Release Date: Nov 10, 2016
    • Ma0 update the year in output stream header 2016-11-06
Version 2.1+46
  • Release Date: Nov 5, 2016
    • mahesh pittala regression: remove pme from smoke tests 2016-11-03
Version 2.1+41
  • Release Date: Nov 3, 2016
    • Chen [slices] fix multi-slices output non-determination bug 2016-10-31
    • Pradeep Ramachandran test: Fix rate-control test cli for 2-pass test with --multi-pass-opt-rps 16 hours ago
    • Pradeep Ramachandran doc: Improve rst docs 20 hours ago
    • Pradeep Ramachandran doc: Improve doc in help 21 hours ago
Version 2.1+36
  • Release Date: Oct 29, 2016
    • Pradeep Ramachandran test: Update 2-pass rc test with fixed GOP structure 2016-10-27
    • Pradeep Ramachandran param: Limit range of log2-max-poc-lsb according to HEVC spec. 2016-10-27
    • Aruna Matheswaran Add VBV support for No-WPP 2016-10-13
    • Divya Manivannan Add scenecut bias to info 2016-10-27
    • Ma0 CLI: allow 'mono' & 'mono16' color space for y4m input 2016-10-16
    • Chen [PATCH] [PPC] GPL v2 copyright header [x265,PPC] GPL v2 copyright header >From 1bea85513646e4d9d992bbe326a9cb3275ec313a Mon Sep 17 00:00:00 2001 2016-10-24
    • Chen [PATCH] [PPC] support option --no-asm to disable Altivec [x265,PPC] support option --no-asm to disable Altivec >From d23527c6204921b782ef8bc2f1a69de88163202a Mon Sep 17 00:00:00 2001 2016-10-24
    • Roger Moussalli [PPC] Altivec support on IBM Power8 CPU Patch modify and integrate by Min Chen 2016-10-25
Version 2.1+25
  • Release Date: Oct 20, 2016
    • Bhavna Hariharan limitTU: fix energy calculation used in limiting TU recursion This commit changes the output of limit TU 2016-10-12
    • Pradeep Ramachandran nit: whitespace fixes 19 hours ago
    • Pradeep Ramachandran api: Introduce param options to toggle PPS bitstream optimizations 22 hours ago
Version 2.1+24
  • Release Date: Oct 19, 2016
    • Pradeep Ramachandran nit: whitespace fixes 19 hours ago
    • Pradeep Ramachandran api: Introduce param options to toggle PPS bitstream optimizations 22 hours ago
Version 2.1+22
  • Release Date: Oct 15, 2016
    • weixuan Store most frequently used list in ref list in PPS 2016-10-13
Version 2.1+21
  • Release Date: Oct 13, 2016
    • xuefeng NA Fix bug that over-writing PPS QP causing inconsisent output and optimise this method. 22 hours ago

Version 2.1+20
  • Release Date: Oct 8, 2016
    • Kavitha Sampath tests: update command lines to cover test for limit-tu 22 hours ago
    • Kavitha Sampath limitTU: cleanup 2016-10-05
    • Kavitha Sampath limitTU: compare full TU's cost with split cost to limit recursion 2016-09-30
    • Kavitha Sampath limitTU: use max TU depth of first subTU to limit recursion of others in split 2016-09-30
    • Bhavna Hariharan limit TU : use cbf and quantization coefficients to limit recursion 2016-09-23
    • Kavitha Sampath cli: add option to support limit-TU 2016-10-04

Version 2.1+14
  • Release Date: Oct 7, 2016
    • Min Chen [x265,slice] fix calculate on slice address bits >From df3f1ed60baa2b5baab0fe9f98f6f2d0f4a6265d Mon Sep 17 00:00:00 2001 From: Min Chen <min.chen@multicorewareinc.com> Date: Tue, 4 Oct 2016 13:43:40 -0500 Subject: [PATCH] [slice] fix calculate on slice address bits 2016-10-05
Version 2.1+13
  • Release Date: Oct 6, 2016
    • Pradeep Ramachandran bitstream: Modify API to discard optional VUI params, remove discard SEI API Cleaner API to remove optional timing & HRD info from VUI. Remove discard SEI API as it can be done through a combo of other API options 2016-10-03
Version 2.1+12
  • Release Date: Oct 2, 2016
    • mahesh pittala regression: add tests for slices 2016-09-28
    • xuefeng NA Set PPS delta QP with a best value to save bits 23 hours ago
    • Pradeep Ramachandran test: Add fixed GOP with scenecut and repeat headers to rate-ctrl tests This test is more representative of a broadcast like setting with fixed GOP, and repeat headers enabled 22 hours ago
    • Divya Manivannan csv: add buffer fill per frame in csv 2016-09-24
    • vignesh Change the scope of MatrixType and MatrixType_DC 2016-06-22
    • vignesh Bug fix in reading 32x32 custom scalinglists 2016-09-20
    • Pradeep Ramachandran Merge with stable an hour ago
    • Pradeep Ramachandran doc: Formating fix for release notes 3 hours ago
    • Pradeep Ramachandran Merge with stable 4 hours ago
    • Pradeep Ramachandran doc: Typo in index file 4 hours ago
    • Pradeep Ramachandran Merge with stable 4 hours ago
    • Pradeep Ramachandran doc: Add release notes rst file 4 hours ago
    • Ashok Kumar Mishra frameFilter: check for reconRowFlag 20 hours ago
    • Pradeep Ramachandran Merge with stable 22 hours ago
    • Pradeep Ramachandran Added tag 2.1 for changeset 981e3bfef16a 22 hours ago
Version 2.1+11
  • Release Date: Sep 29, 2016
    • xuefeng NA Set PPS delta QP with a best value to save bits 23 hours ago
    • Pradeep Ramachandran test: Add fixed GOP with scenecut and repeat headers to rate-ctrl tests This test is more representative of a broadcast like setting with fixed GOP, and repeat headers enabled 22 hours ago
    • Divya Manivannan csv: add buffer fill per frame in csv 2016-09-24
    • vignesh Change the scope of MatrixType and MatrixType_DC 2016-06-22
    • vignesh Bug fix in reading 32x32 custom scalinglists 2016-09-20
    • Pradeep Ramachandran Merge with stable an hour ago
    • Pradeep Ramachandran doc: Formating fix for release notes 3 hours ago
    • Pradeep Ramachandran Merge with stable 4 hours ago
    • Pradeep Ramachandran doc: Typo in index file 4 hours ago
    • Pradeep Ramachandran Merge with stable 4 hours ago
    • Pradeep Ramachandran doc: Add release notes rst file 4 hours ago
    • Ashok Kumar Mishra frameFilter: check for reconRowFlag 20 hours ago
    • Pradeep Ramachandran Merge with stable 22 hours ago
    • Pradeep Ramachandran Added tag 2.1 for changeset 981e3bfef16a 22 hours ago
Version 2.1+5
  • Release Date: Sep 29, 2016
  • Version 2.1
    Release date - 27th September, 2016

    Encoder enhancements
    Support for qg-size of 8
    Support for inserting non-IDR I-frames at scenecuts and when running with settings for fixed-GOP (min-keyint = max-keyint)
    Experimental support for slice-parallelism.
    API changes
    Encode user-define SEI messages passed in through x265_picture object.
    Disable SEI and VUI messages from the bitstream
    Specify qpmin and qpmax
    Control number of bits to encode POC.
    Bug fixes
    QP fluctuation fix for first B-frame in mini-GOP for 2-pass encoding with tune-grain.
    Assembly fix for crashes in 32-bit from dct_sse4.
    Threadpool creation fix in windows platform.

    x265 2.1+5 changelog
    Divya Manivannan csv: add buffer fill per frame in csv 2016-09-24
    vignesh Change the scope of MatrixType and MatrixType_DC 2016-06-22
    vignesh Bug fix in reading 32x32 custom scalinglists 2016-09-20
    Pradeep Ramachandran Merge with stable an hour ago
    Pradeep Ramachandran doc: Formating fix for release notes 3 hours ago
    Pradeep Ramachandran Merge with stable 4 hours ago
    Pradeep Ramachandran doc: Typo in index file 4 hours ago
    Pradeep Ramachandran Merge with stable 4 hours ago
    Pradeep Ramachandran doc: Add release notes rst file 4 hours ago
    Ashok Kumar Mishra frameFilter: check for reconRowFlag 20 hours ago
    Pradeep Ramachandran Merge with stable 22 hours ago
    Pradeep Ramachandran Added tag 2.1 for changeset 981e3bfef16a 22 hours ago

    See https://bitbucket.org/multicoreware/x265/commits/all
Version 2.0+65
  • Release Date: Sep 24, 2016
    • Pradeep Ramachandran Merge with stable 20 hours ago
    • Gopi Satykrishna Akisetty modify the minimum keyint value This patch enables fixed gop size even when scenecut option is enabled 20 hours ago
    • Pradeep Ramachandran bdf098e Merge with stable 14 hours ago
    • Divya Manivannan rc: fix average P-frame qp in 2-pass 2016-09-21
    • Pradeep Ramachandran Merge with default; prep for 2.1
    • Pradeep Ramachandran vui: Fix discard-vui to only remove optional VUI information 2016-09-20

Version 2.0+64
  • Release Date: Sep 23, 2016
    • Pradeep Ramachandran bdf098e Merge with stable 14 hours ago
    • Divya Manivannan rc: fix average P-frame qp in 2-pass 2016-09-21
    • Pradeep Ramachandran Merge with default; prep for 2.1
    • Pradeep Ramachandran vui: Fix discard-vui to only remove optional VUI information 2016-09-20
Version 2.0+61
  • Release Date: Sep 21, 2016
    • Pradeep Ramachandran vui: Fix discard-vui to only remove optional VUI information 2016-09-20
Version 2.0+60
  • Release Date: Sep 21, 2016
    • Gopi Satykrishna Akisetty rc: fix inconsistent output with qg-size 817 hours ago
Version 2.0+59
  • Release Date: Sep 18, 2016
    • Divya Manivannan rc: fix non-IDR slicetype in multi-pass 2016-09-14
Version 2.0+58
  • Release Date: Sep 15, 2016
    • Divya Manivannan csv: fix reference list for non-IDR frames 22 hours ago
    • Divya Manivannan csv: Change I-slice into i-slice for non-IDR frames 2016-09-13
    • Chen [PATCH] [slice] fix help information defaule value mistake [x265,slice] fix help information defaule value mistake 
Version 2.0+55
  • Release Date: Sep 14, 2016
    • Min Chen [x265,slice] verify untest path and enable it 12 hours ago
Version 2.0+54
  • Release Date: Sep 10, 2016
    • Pradeep Ramachandran Backed out changeset: 16359da0653e 13 hours ago
    • Divya Manivannan rc: fix slicetype in 2-pass 2016-09-08
    • Pradeep Ramachandran Merge with stable 2016-09-08
Version 2.0+22
  • Release Date: Sep 2, 2016
    • Tom Vaughan doc: cleanup cli.rst for profile listing 10 hours ago
Version 2.0+21
  • Release Date: Sep 1, 2016
    • Gopi Satykrishna Akisetty nits: Whitespace, remove a comment 2016-08-29
    • liujun add --[no-]analysis-src-pics parameter to enable/disable use source frame pixels to do motion estimation. 23 hours ago
Version 2.0+12
  • Release Date: Aug 11, 2016
    • Ma0 cmake: fix unaligned stack in 32-bit Windows for GCC 6 2016-07-27
    • Min Chen asm: fix typo mistake 2016-07-27
    • Santhoshini S bugfix: encode only once the final best mode when picturecsp=400 and not merge2Nx2N 2016-07-22
    • Ma0 silence VS 2013 warning about g_scan4x4 2016-07-19
    • Deepthi Nandakumar doc: clarify aq-mode/aq-strength effects an hour ago
    • Ramya Sriraman arm: Implement psyCost_pp_8x8_neon 2016-07-05
    • Satoshi Nakagawa constants: g_zscanToRaster, g_rasterToZscan as const table 2016-07-05
    • Gopi Satykrishna Akisetty rc: add qpmin and qpmax options 2016-07-15
Version 2.0+11
  • Release Date: Jul 30, 2016
    • in Chen asm: fix typo mistake 2016-07-27
    • Santhoshini S bugfix: encode only once the final best mode when picturecsp=400 and not merge2Nx2N 2016-07-22
    • Ma0 silence VS 2013 warning about g_scan4x4 2016-07-19
    • Deepthi Nandakumar doc: clarify aq-mode/aq-strength effects an hour ago
    • Ramya Sriraman arm: Implement psyCost_pp_8x8_neon 2016-07-05
    • Satoshi Nakagawa constants: g_zscanToRaster, g_rasterToZscan as const table 2016-07-05
    • Gopi Satykrishna Akisetty rc: add qpmin and qpmax options 2016-07-15
Version 2.0+9
  • Release Date: Jul 23, 2016
    • Ma0 silence VS 2013 warning about g_scan4x4 2016-07-19
    • Deepthi Nandakumar doc: clarify aq-mode/aq-strength effects an hour ago
    • Ramya Sriraman arm: Implement psyCost_pp_8x8_neon 2016-07-05
    • Satoshi Nakagawa constants: g_zscanToRaster, g_rasterToZscan as const table 2016-07-05
    • Gopi Satykrishna Akisetty rc: add qpmin and qpmax options 2016-07-15

Version 2.0+8
  • Release Date: Jul 20, 2016
    • Deepthi Nandakumar doc: clarify aq-mode/aq-strength effects an hour ago
    • Ramya Sriraman arm: Implement psyCost_pp_8x8_neon 2016-07-05
    • Satoshi Nakagawa constants: g_zscanToRaster, g_rasterToZscan as const table 2016-07-05
    • Gopi Satykrishna Akisetty rc: add qpmin and qpmax options 2016-07-15
Version 2.0+5
  • Release Date: Jul 20, 2016
    • Deepthi Nandakumar doc: clarify aq-mode/aq-strength effects an hour ago
    • Ramya Sriraman arm: Implement psyCost_pp_8x8_neon 2016-07-05
    • Satoshi Nakagawa constants: g_zscanToRaster, g_rasterToZscan as const table 2016-07-05
    • Gopi Satykrishna Akisetty rc: add qpmin and qpmax options 2016-07-15
Version 2.0+4
  • Release Date: Jul 17, 2016
  • x265 2.0+4
    Deepthi Nandakumar sei: initialise payload type for infoSEI 2016-07-15
    Deepthi Nandakumar sei: add arbitrary user SEI 2016-01-07

    x265 version 2.0 has been released. This release supports many new features as well as support for ARM assembly optimizations for most basic pixel and ME operations, as well as SAO cleanups and a fully tested reconfigure functionality.
    Full documentation is available at http://x265.readthedocs.org/en/stable/
    ==================== New Features====================
    • uhd-bd: Enforce Ultra-HD Blu-ray Disc parameters (overrides any other settings)
    • rskip: Enables skipping recursion to analyze lower CU sizes using heuristics at different rd-levels. Provides good visual quality gains at the highest quality presets.
    • rc-grain: Enables a new rate control mode specifically for grainy content. Strictly prevents QP oscillations within and between frames to avoid grain fluctuations.
    • tune grain: A fully refactored and improved option to encode film grain content including QP control as well as analysis options.
    • asm: ARM assembly is now enabled by default, native or cross compiled builds supported on armv6 and later systems.
    ============ API and Key Behavior Changes ==============
    • x265_rc_stats added to x265_picture, containing all RC decision points for that frame
    • PTL: high tier is now allowed by default, chosen only if necessary
    • multi-pass: First pass now uses slow-firstpass by default, enabling better RC decisions in future passes
    • pools: fix behaviour on multi-socketed Windows systems, provide more flexibility in determining thread and pool counts
    • ABR: improve bits allocation in the first few frames, abr reset, vbv and cutree improved
    ====================== Misc =========================
    • An SSIM calculation bug was corrected

    See https://bitbucket.org/multicoreware/x265/commits/all
Version 1.9+230
  • Release Date: Jul 14, 2016
    • Aarthi Thirumalai rcStats: add more fields to rcStats 2016-07-11
Version 1.9+229
  • Release Date: Jul 5, 2016
    • Deepthi Nandakumar Merge with stable 15 hours ago
    • Min Chen Reduce operators on row address compare 2016-07-01
    • Santhoshini S analysis: compute Inter2Nx2N after merge in analysis mode=load if reuse mode chosen is skip 23 hours ago

Version 1.9+227
  • Release Date: Jul 2, 2016
    • Deepthi Nandakumar Merge with stable 52 minutes ago
    • Pradeep Ramachandran warning: Remove shadow declaration of refMasks[] # HG changeset patch # User Pradeep Ramachandran <pradeep@multicorewareinc.com> # Date 1467351960 -19800 # Fri Jul 01 11:16:00 2016 +0530 # Branch stable # Node ID e9da1432113fc20b865760fe12bf5ec29c876cae # Parent 1b154ad27f61c93dc073563d6a95f9e26f12e700 warning: Remove shadow declaration of refMasks[] an hour ago
    • Divya Manivannan rc: change max value of ABR and disable the frame parallelism in first frame 2016-06-28
    • Deepthi Nandakumar Merge with default; prep for 2.0 17 hours ago
    • Pradeep Ramachandran test: Adding test for veryslow without recursion skip # HG changeset patch # User Pradeep Ramachandran <pradeep@multicorewareinc.com> # Date 1467259354 -19800 # Thu Jun 30 09:32:34 2016 +0530 # Node ID 7f5aa417f09763f84a0bfd17612ca38aabe37d1a # Parent 626fcbac7ffba723dabd3a9f0507c4c80f3e7bc9 test: Adding test for veryslow without recursion skip 2016-06-30
    • Santhoshini S introduce multi-level recursion skip 2016-06-23
    • Divya Manivannan rc: enable row diagonal vbv for crf multi-pass 2016-06-22
    • Aarthi Thirumalai fix warnings in sao 2016-06-24
    • Ramya Sriraman arm: Fix build.Does not fix output change caused by <a5362b9> 2016-06-30
    • Min Chen asm: fix output change bug in pixel_sa8d_32x32, the reason is intermedia result overflow 
Version 1.9+204
  • Release Date: Jun 5, 2016
    • mahesh pittala uhdbd: fix sourceHeight check when sourceHeight is 1080, it is padded by some value to become multiple of the minimum CU size(for ex- 1088) so just moved down those calculations after uhd-bd. 2016-04-25
    • nandaku2 uhd-bd: Clarifications to level and tier determination 2016-06-03
Version 1.9+200
  • Release Date: Jun 3, 2016
    • Ma0 sao: silence warning 2 hours ago
    • nandaku2 level: fix bug in level and tier determination (#refs 278) This patch changes the behaviour of tier determination (and sometimes level as well), and hence bumping up build number to warn users. High-tier is "allowed" by default, and --no-high-tier tells the encoder to never choose high tier. 17 hours ago
    • Ricardo Constantino Make git X265_VERSION output the same as mercurial (except X265_REVISION_ID) 2016-05-31
    • Ricardo Constantino Fix git X265_VERSION and X265_LATEST_TAG 2016-05-31
    • Ricardo Constantino Fix HG_REVISION_ID in Windows cmake 2016-05-31
    • Pradeep Ramachandran threadpool: Fix counting of cpusPerNode for linux 2016-06-01
    • Santhoshini S Allow different Csp for picture 2016-05-25
    • Ashok Kumar Mishra [OUTPUT CHANGED]SAO: convert sao rdo cost calculation from float to int 2016-05-30
Version 1.9+192
  • Release Date: May 29, 2016
    • mahesh pittala correct macro to find first bit set position from LSB to MSB 2016-05-26
    • Min Chen asm_arm: rewrite filterPixelToShort_4x4, improve by remove unnecessary and slow instructions 2016-05-25
    • Min Chen asm_arm: NEON version of psyCost_4x4 2016-05-25
    • Divya Manivannan sei: fix initial_cpb_removal_delay in HRD SEI 2016-05-25
    • nandaku2 threadpool: prevent multilib build failures 2016-05-27
    • nandaku2 sao: cleanup function names, processSaoCu is used too often to be meaningful 2016-05-25
    • Praveen Tiwari CLI: new logic for '--pools <int>' option [Credit to Mateusz <mateuszb@poczta.onet.pl>] For '--pools N' option we create exactly N threads. For old logic please add a comma, for example '--pools 24,' 2016-05-23
    • Praveen Tiwari threadpool-fix: utilize all available NUMA nodes for threadpool distribution for windows system, linux threadpool configuration info, match with windows -> clean logic 2016-05-23
    • Ashok Kumar Mishra SAO: consider chroma distortion when chroma sao is disabled It is a valid modification which I missed in my last patch. Note that output may change where chroma sao is disabled. 2016-04-25
Version 1.9+183
  • Release Date: May 21, 2016
    • Divya Manivannan rc: clip the qp for 2-pass with grain 2016-05-13
    • nandaku2 multi-pass: enable slow-firstpass by default. The better the match across multi-pass encode settings, the better the efficiency improvement from each pass. 20 hours ago
    • Radhakrishnan arm: Implement ssim_4x4x2_core_neon ARM NEON 2016-05-10
    • Radhakrishnan arm: Implement addAvg chroma ARM NEON 2016-05-06
    • Radhakrishnan arm: Implement addAvg ARM NEON 2016-05-06
    • Min Chen asm_arm: redesign algorithm and rewrite interp_8tap_vert_pp_4xN Origin: luma_vpp[ 4x4]1.87x 45.23 84.41 luma_vpp[ 4x8]2.10x 70.36 147.78 luma_vpp[ 4x16]2.25x 121.24 272.18 New: luma_vpp[ 4x4]3.10x 27.47 85.05 luma_vpp[ 4x8]4.59x 32.21 147.76 luma_vpp[ 4x16]6.38x 42.73 272.48 2016-05-18
    • Min Chen asm_arm: improve interp_8tap_vert_pp_4xN By: 1. remove unnecessary cache prefetch instructions pld 2. replace register r6 by lr Origin: luma_vpp[ 4x4]1.87x 45.23 84.41 luma_vpp[ 4x8]2.10x 70.36 147.78 luma_vpp[ 4x16]2.25x 121.24 272.18 Optimized: luma_vpp[ 4x4]1.98x 42.42 84.02 luma_vpp[ 4x8]2.32x 63.70 147.49 luma_vpp[ 4x16]2.51x 108.39 272.18 2016-05-18
    • Ma0 CLI: allow Unicode filenames (Windows) for '--qpfile' and '--csv' 2016-05-14
    • Min Chen asm: ARM NEON version of DCT[8x8] 2016-05-12
    • Min Chen asm: new macro TRANSPOSE4x4x2_16 to avoid pipeline conflict in combo matrix transpose 2016-05-12
    • Satoshi Nakagawa predict: remove m_immedVals 2016-05-12
    • Ximing Cheng weightsAnalyse: use CLZ to get a smaller denominator 2016-05-07
    • Tom Vaughan [x265] docs: fix presets.rst to fit on one page without scrolling 2016-05-18
    • Tom Vaughan [x265] docs: edited presets.rst to enable full table to display without horizontal scrolling on readthedocs.org 2016-05-18

Version 1.9+167
  • Release Date: May 14, 2016
    • Sagar Kotecha remove redudant check 2016-05-11
    • Ma0 cmake: apply -march=i686 for 32bit GCC only if there is no -march in CXXFLAGS (fixes #270) 2016-05-05
    • nandaku2 cmake: update build (fixes #271) 4 hours ago
    • nandaku2 grain: improve grain handling settings Turn off SAO, increase psychovisual settings to better retain high frequency 21 hours ago
    • nandaku2 analysis: introduce recursion skip Earlier, analysis would exit recursion if a skip was found. This option provides the ability to control when early exit from recursion needs to be turned on 19 hours ago
    • nandaku2 analysis: use only skip costs for recursionDepthCheck Also cleanup variable names 2016-05-06
Version 1.9+162
  • Release Date: May 11, 2016
    • Pradeep Ramachandran arm: Enable asm by default, allow gcc to auto-detect cpu - Enabled ASM by default, and fixed compilation problem without asm - GCC now auto-detects CPU instead of forcing armv6; significant speed boost - Convert ARM compile to be native by default - cross compile requires special work now 2016-05-04

Version 1.9+161
  • Release Date: May 9, 2016
    • Ramya sriraman arm: Implement nquant (revised) 2016-04-22
    • Ramya sriraman arm: Implement quant (revised) 2016-04-20
    • Radhakrishnan arm: Implement add_ps chroma ARM NEON 2016-04-20
    • Radhakrishnan arm: Implement sub_ps chroma ARM NEON 2016-04-20
    • Radhakrishnan arm: Implement blockcopy_sp, ps, ss chroma ARM NEON 2016-04-19
    • Radhakrishnan arm: Implement blockcopy_pp chroma ARM NEON 2016-04-19
    • Radhakrishnan arm: Implement dequant_normal ARM NEON 2016-04-15
    • Radhakrishnan arm: Implement dequant_scaling ARM NEON 2016-04-15
    • Divya Manivannan rc: check scenecut flag for resetting abr 2016-05-03
    • Ma0 CLI: free memory allocated for utf8 command line in Windows 2016-05-03
    • Ma0 CLI: fix Unicode output in Windows for old mingw-w64 2016-04-29
Version 1.9+150
  • Release Date: May 2, 2016
    • Ma0 Allows for Unicode filenames in Windows (output and stat files). Copy of x264 code for processing Unicode filenames in Windows. Output file and stat file(s) are very important for GUI makers. 2016-04-28
    • Min Chen asm: ARM NEON version of DCT[4x4] 2016-04-26
    • Divya Manivannan rc: fix the average qp calculation2016-04-28
Version 1.9+144
  • Release Date: Apr 27, 2016
    • Divya Manivannan rc: clip the qp in ratecontrolEnd 2016-04-22
    • Divya Manivannan rc: add option for encoding next gops 2016-04-07
    • Radhakrishnan arm: Implement pixel_satd ARM NEON 2016-03-30
Version 1.9+141
  • Release Date: Apr 18, 2016
    • Tom Vaughan doc: Improve description of --tune grain 10 hours ago
Version 1.9+140
  • Release Date: Apr 15, 2016
    • Ashok Kumar Mishra SAO: fix for output mismatch in windows 2016-04-12
Version 1.9+139
  • Release Date: Apr 13, 2016
    • nandaku2 uhd-bd: turn off open GOP for UHD Blu-ray 9 hours ago
    • Ma0 console: change '--limit-refs' info from 1/0 to on/off Change displayed info about '--limit-refs' to better understanding by user. Example: change from x265 [info]: References / ref-limit cu / depth : 5 / 0 / 1 to x265 [info]: References / ref-limit cu / depth : 5 / off / on 12 hours ago
    • Min Chen asm: rewrite interpolate hps width of [32,48,64], improve ~20% OLD: luma_hps[32x32] 6.32x 16429.69 103771.02 luma_hps[32x16] 6.04x 10121.56 61140.21 luma_hps[32x64] 6.47x 30813.70 199438.95 luma_hps[32x24] 6.23x 13277.26 82747.75 luma_hps[48x64] 6.13x 46002.25 282176.44 luma_hps[64x64] 6.15x 61393.88 377670.03 luma_hps[64x32] 6.79x 33001.77 224096.58 luma_hps[64x48] 6.21x 47242.66 293529.16 luma_hps[64x16] 6.51x 19207.61 125016.56 NEW: luma_hps[32x32] 7.66x 13404.22 102730.96 luma_hps[32x16] 7.32x 8355.57 61133.25 luma_hps[32x64] 7.68x 24496.17 188086.11 luma_hps[32x24] 8.00x 10879.09 87077.93 luma_hps[48x64] 7.62x 37094.37 282758.94 luma_hps[64x64] 7.82x 48535.86 379390.78 luma_hps[64x32] 7.91x… 14 hours ago
    • Min Chen asm: AVX2 version of sa8d[32x32] AVX: sa8d[32x32] 5.47x 7403.68 40490.18 AVX2: sa8d[32x32] 10.57x 3783.80 40001.89 14 hours ago
    • nandaku2 deblock: print 0:0 offsets as well in info SEI (fixes #258) 12 hours ago
    • Ma0 [x265] change int literals to char literals to avoid VS 2015 warnings Fix VS 2015 warning: x265\source\common\cpu.cpp(277): warning C4838: conversion from 'int' to 'const char' requires a narrowing conversion 14 hours ago
    • Radhakrishnan arm : Implement interp_4tap_horiz_pp,ps ARM NEON 2016-03-29
    • Radhakrishnan arm :Implement interp_8tap_horiz_ps ARM NEON 2016-03-29
    • Radhakrishnan arm :Implement interp_8tap_horiz_pp ARM NEON 2016-03-15
    • Ramya sriraman arm: Implement sa8d for all luma and chroma partitions 2016-04-05
    • Ramya sriraman arm: Implement interp_4tap_vert_pp,ps &sp for NxN NEON 2016-03-24
    • Ramya sriraman arm: Implement interp_8tap_vert_ps_NxN NEON 2016-03-22
    • Ramya sriraman arm: Implement interp_8tap_vert_sp_NXN NEON 2016-03-22
    • Ramya sriraman arm: Implement interp_8tap_vert_pp_NxN NEON 2016-03-16
Version 1.9+125
  • Release Date: Apr 10, 2016
    • nandaku2 doc: update tune grain documentation 17 hours ago
    • nandaku2 rd and rdoq: clarify documentation, remove obsolete comments Also emphasize that rd-level 0 is not supported. 17 hours ago
Version 1.9+123
  • Release Date: Apr 9, 2016
    • nandaku2 dither: fix dithering to 8-bit 14 hours ago
    • nandaku2 psy-rdoq: default values explained better, film grain details removed Our recent tests and analysis showed that the best way to get a consistent encode of film grain was to prevent QP fluctuations throughout the sequence as much as possible. See rc-grain. 16 hours ago
    • nandaku2 dither: return if encoder and picture depth are the same 16 hours ago
    • Ma0 [x265] '--dither' patch for 10/12-bit and for not default bit depth fix '--dither' option for 10/12-bit and for not default bit depth (fixes #255) Mateusz 16 hours ago
    • Min Chen fix threading conflict in low resolution video (Issue #260) The threading conflict because video resolution too low, it made threading approach finish in same time. The root cause in our sync logic, we relase all of filter sync-lock in latest column processed, I give more details in below. Time 0: Row0 - request assign work threading Row1 - request assign work threading Time 1: Row0 - assign threading failure, and will continue in FrameFilter::proc… 2016-04-07
    • Ashok Kumar Mishra SAO: removed redundant distortion calculation 2016-03-29
    • Divya Manivannan Fix poc value written in csv file for open-gop 2016-04-05
    • Divya Manivannan Fix the order of fields in x265_picture 2016-03-31
    • Min Chen asm: new AVX2 version of sa8d[8x8, 16x16] AVX: sa8d[8x8] 4.82x 517.79 2493.20 sa8d[16x16] 5.65x 1952.40 11039.93 AVX2: sa8d[8x8] 5.13x 489.15 2507.44 sa8d[16x16] 10.27x 1006.08 11206.09 2016-03-30
    • Min Chen asm: new AVX2 version of satd_8x8 (509c -> 307c) 2016-03-30
    • Ramya sriraman arm: Port pixel_sa8d_8x8_neon and pixel_sa8d_16x16_neon 2016-03-28
    • Satoshi Nakagawa move tables from .h to .cpp 2016-03-28
    • Satoshi Nakagawa remove broadcast of non-leaf CBF 2016-03-25
    • Sagar Kotecha analysis: skip rect/amp in analysis load mode Avoid doing rect/amp analysis in load mode if the save mode has not chosen it as the best partition 2016-03-24
    • nandaku2 x265: cleanup unnecessary header files 2016-03-29
Version 1.9+108
  • Release Date: Apr 3, 2016
    • Mat Booth Fix compilation errors when using C++11 Since the introduction of user-defined literals in C++11, it is now mandatory to have whitespace between the operands of string concatenation operations. 13 hours ago
Version 1.9+107
  • Release Date: Mar 29, 2016
    • nandaku2 uhd-bd: 1088 needs to be cropped to conformance window of 1080, which x265 doesnt allow 23 hours ago
Version 1.9+106
  • Release Date: Mar 25, 2016
    • mahesh pittala add UHD bluray compatibility support 2016-02-19
    • nandaku2 hdr: maxCLL/maxFALL convert YUV to RGB domain, and then to linear luminance levels miscellaneous cleanup, move PIXEL_MAX to common 2016-03-22
    • nandaku2 level: correct logging of warning and error messages 23 hours ago
    • nandaku2 doc: correct master-display example (fixes #249) 2016-03-21
    • Vignesh fix bug in scaling list 2016-03-15
    • Pradeep Ramachandran cmake: Export PLATFORM_LIBS only if x265 has a parent scope # HG changeset patch # User Pradeep Ramachandran <pradeep@multicorewareinc.com> # Date 1458185054 -19800 # Thu Mar 17 08:54:14 2016 +0530 # Node ID cb9a50f36ac6fb5ae60d08f51f4ca27eb9f02eb2 # Parent b09998b1256ed8e652c4bf2d688cbfab3a84d5cd cmake: Export PLATFORM_LIBS only if x265 has a parent scope 2016-03-17
    • nandaku2 analysis: initialise chroma sa8d for slave pmode threads as well 2016-03-18
    • nandaku2 param: prevent race conditions in initialising global tables (refs #247) Reset param if different encoder instances have different CU sizes. 2016-03-18
    • nandaku2 encoder: the first frame encoder resets reconfigure for itself and all other FEs Reconfigure also now prints the next frame ID to be queued (and not the next one to be encoded) 2016-03-17
    • nandaku2 level: correct reference buffer initialisations 2016-03-17
    • Divya Manivannan rc: fix zone in multi-pass encodes (fixes #172) 2016-03-16
    • Radhakrishnan arm: Implement count_nonzero ARM NEON 2016-03-15
    • Pradeep Ramachandran cmake: Use PARENT_SCOPE instead of CACHE to export PLATFORM_LIBS # HG changeset patch # User Pradeep Ramachandran <pradeep@multicorewareinc.com> # Date 1458041280 -19800 # Tue Mar 15 16:58:00 2016 +0530 # Node ID e166d9690442a21eac9d0672c1b1729bff7f5996 # Parent 4a2f94a592511afabd434fc6cf02a469b6d65091 cmake: Use PARENT_SCOPE instead of CACHE to export PLATFORM_LIBS 2016-03-15
Version 1.9+104
  • Release Date: Mar 22, 2016
    • nandaku2 level: correct logging of warning and error messages 23 hours ago
    • nandaku2 doc: correct master-display example (fixes #249) 2016-03-21
    • Vignesh fix bug in scaling list 2016-03-15
    • Pradeep Ramachandran cmake: Export PLATFORM_LIBS only if x265 has a parent scope # HG changeset patch # User Pradeep Ramachandran <pradeep@multicorewareinc.com> # Date 1458185054 -19800 # Thu Mar 17 08:54:14 2016 +0530 # Node ID cb9a50f36ac6fb5ae60d08f51f4ca27eb9f02eb2 # Parent b09998b1256ed8e652c4bf2d688cbfab3a84d5cd cmake: Export PLATFORM_LIBS only if x265 has a parent scope 2016-03-17
    • nandaku2 analysis: initialise chroma sa8d for slave pmode threads as well 2016-03-18
    • nandaku2 param: prevent race conditions in initialising global tables (refs #247) Reset param if different encoder instances have different CU sizes. 2016-03-18
    • nandaku2 encoder: the first frame encoder resets reconfigure for itself and all other FEs Reconfigure also now prints the next frame ID to be queued (and not the next one to be encoded) 2016-03-17
    • nandaku2 level: correct reference buffer initialisations 2016-03-17
    • Divya Manivannan rc: fix zone in multi-pass encodes (fixes #172) 2016-03-16
    • Radhakrishnan arm: Implement count_nonzero ARM NEON 2016-03-15
    • Pradeep Ramachandran cmake: Use PARENT_SCOPE instead of CACHE to export PLATFORM_LIBS # HG changeset patch # User Pradeep Ramachandran <pradeep@multicorewareinc.com> # Date 1458041280 -19800 # Tue Mar 15 16:58:00 2016 +0530 # Node ID e166d9690442a21eac9d0672c1b1729bff7f5996 # Parent 4a2f94a592511afabd434fc6cf02a469b6d65091 cmake: Use PARENT_SCOPE instead of CACHE to export PLATFORM_LIBS 2016-03-15
Version 1.9+88
  • Release Date: Mar 13, 2016
    • Divya Manivannan rc: change reencode position for cappedvbr 2016-03-10
    • nandaku2 ratecontrol: fix Darwin warning 2016-03-11
    • nandaku2 x265cli: fix format type 16 hours ago
    • nandaku2 rc-grain: clarify doc 21 hours ago
    • nandaku2 param: fix double assignment 2016-03-09
    • Aarthi Thirumalai rc: fix tune grain to adapt to scene changes and preserve quality 2016-02-05
    • Aarthi Thirumalai rc: fix Rate Control for grainy content optimize params for tune grain, reduce frequent qp fluctions to prevent grain loss 2016-03-04
    • Sagar Kotecha frame: add encodeOrder 21 hours ago
    • Ashok Kumar Mishra sao: Use qp of encoded CU, instead of slice qp When dqp is enabled, qp value applied on CU for encoding is different from slice qp. So in this case the slice qp value should not be used for SAO offset distortion calculation. 2016-03-08
    • Ramya sriraman arm: Implement planecopy_cp NEON 22 hours ago
    • Sagar Kotecha analysis: init depth and mode analysis buffer Encoder nits. 2016-03-08
    • Sagar Kotecha param: return gracefully on invalid analysis mode option 2016-03-09
    • Sagar Kotecha analysis: Improve and fix indexing issue for analysis-load/save 2016-03-08
    • Radhakrishnan arm: Implement copy_count_neon ARM NEON 2016-03-08
    • nandaku2 reconfigure: modify return values and docs 2016-03-10
Version 1.9+86
  • Release Date: Mar 11, 2016
    • nandaku2 x265cli: fix format type 16 hours ago
    • nandaku2 rc-grain: clarify doc 21 hours ago
    • nandaku2 param: fix double assignment 2016-03-09
    • Aarthi Thirumalai rc: fix tune grain to adapt to scene changes and preserve quality 2016-02-05
    • Aarthi Thirumalai rc: fix Rate Control for grainy content optimize params for tune grain, reduce frequent qp fluctions to prevent grain loss 2016-03-04
    • Sagar Kotecha frame: add encodeOrder 21 hours ago
    • Ashok Kumar Mishra sao: Use qp of encoded CU, instead of slice qp When dqp is enabled, qp value applied on CU for encoding is different from slice qp. So in this case the slice qp value should not be used for SAO offset distortion calculation. 2016-03-08
    • Ramya sriraman arm: Implement planecopy_cp NEON 22 hours ago
    • Sagar Kotecha analysis: init depth and mode analysis buffer Encoder nits. 2016-03-08
    • Sagar Kotecha param: return gracefully on invalid analysis mode option 2016-03-09
    • Sagar Kotecha analysis: Improve and fix indexing issue for analysis-load/save 2016-03-08
    • Radhakrishnan arm: Implement copy_count_neon ARM NEON 2016-03-08
    • nandaku2 reconfigure: modify return values and docs 2016-03-10
Version 1.9+73
  • Release Date: Mar 10, 2016
    • nandaku2 reconfigure: revise reconfigurable parameters2016-03-02
    • Sagar Kotecha analysis: Do not save/load MV's if analysis mode is enabled2016-03-07
    • Pradeep Ramachandran ratecontrol: use fabs() instead of abs() to remove clang warning in darwin 2016-03-08
    • mahesh pittala regression: add crf multi pass commandlines 2016-03-04
Version 1.9+68
  • Release Date: Mar 5, 2016
    • nandaku2 encoder: handle m_latestParam and m_param correctly During a reconfigure, m_latestParam and m_param will be different. Once all FEs have been reconfigured, the Encoder considers its reconfigure complete. 21 hours ago
    • nandaku2 framefilter: initialise param from FrameEncoder::m_param 21 hours ago
    • Dnyaneshwar Gorade arm: Implement filterPixelToShort ARM NEON asm 2016-03-01
    • Pradeep Ramachandran cmake: Export PLATFORM_LIBS for other integrators to re-use # HG changeset patch # User Pradeep Ramachandran <pradeep@multicorewareinc.com> # Date 1456997684 -19800 # Thu Mar 03 15:04:44 2016 +0530 # Node ID d037c0f685b195bafd2946dfbe86f1c47c5359bc # Parent 9cc9920bf82be1b43efd2a3628e28a3a78ab3b2f cmake: Export PLATFORM_LIBS for other integrators to re-use 2016-03-03
    • Radhakrishnan arm: Implement blockfill_s_neon ARM NEON 2016-03-01
    • nandaku2 lowresMV: enable only if analysisMode load/save is disabled 2016-03-04
    • nandaku2 encoder: nits 2016-03-03
    • nandaku2 Merge with stable 2016-03-03
    • nandaku2 calcHDRStats: cleanup, remove bug related to min/max luma 2016-03-03
Version 1.9+59
  • Release Date: Mar 4, 2016
    • nandaku2 dpb: cleanup, use param->maxReferences 21 hours ago
    • nandaku2 motion: set search method/subpel refine for each PU 23 hours ago
    • nandaku2 quant: set rdoq level in setQPforQuant, and not init 23 hours ago
    • nandaku2 search: remove RDOQ level initialisations in initSearch 23 hours ago
    • nandaku2 Backed out changeset: 7ede72109f16 Breaks C compatibility 22 hours ago
Version 1.9+54
  • Release Date: Mar 2, 2016
    • Radhakrishnan arm: Implement blockcopy_ss_neon ARM NEON 2016-02-26
    • Radhakrishnan arm: Implement blockcopy_ps_neon ARM NEON 2016-02-26
    • Radhakrishnan arm: Implement blockcopy_sp_neon ARM NEON 2016-02-26
    • Radhakrishnan arm: Implement pixel_add_ps_neon ARM NEON2016-02-25
    • Ramya sriraman arm: Implement sad for all versions 2016-02-24
    • Dnyaneshwar Gorade asm: asm code for pelFilterLumaStrong_V/H & pelFilterChroma_V/H for main10 & main12 2016-02-26
    • Dnyaneshwar Gorade asm: asm code for pelFilterChroma_V/H 2016-02-26
    • Dnyaneshwar Gorade asm: separated pelFilterChroma function into horizontal & vertical primitives for asm 2016-02-26
    • Dnyaneshwar Gorade arm: Implement cpy2Dto1D_shr ARM NEON asm 2016-02-25
    • Dnyaneshwar Gorade arm: Implement pixel_ssd_s ARM NEON asm 2016-02-22
    • Dnyaneshwar Gorade arm: Implement pixel_sse_ss ARM NEON asm 2016-02-25
    • Divya Manivannan rc: fix ABR 2-pass 2016-02-26
    • nandaku2 primitives: change planeClipAndMax to calcStats_HDR, add YUV to RGB conversions 2016-02-27
    • nandaku2 picyuv: luma max/min only for HDR-relevant encodes, fix param initialisations 2016-02-28
    • Pradeep Ramachandran stats: add clear() function to frame-level stats struct 2016-02-23
Version 1.9+27
  • Release Date: Feb 18, 2016
    • Divya Manivannan rc: fix GOP reencode logic for cappedvbr 2016-02-11
    • Divya Manivannan Add x265_rc_stats structure to collect rc data for each frame 2016-02-05
    • nandaku2 encoder: replace with colour planes macro 22 hours ago
    • chenm003 asm: high_bit_depth sse4 version of saoCuStatsE0 & saoCuStatsE1 2016-02-17
    • Kavitha Sampath cleanup: remove explicit passing of zOrder to compressInter of rd 5,6 Also cleanup earlySkip used in compressInter functions 2016-02-08
    • nandaku2 Merge with stable 21 hours ago
    • Dnyaneshwar Gorade arm: Implement sad_x3 and sad_x4 ARM NEON asm 2016-02-16
    • radh...@multicorewareinc.com arm: Implement blockcopy_pp_NxN_neon 2016-02-11
    • chenm003 fix VBV hash mistake on re-encode mode Branches stable 2016-02-17
    • Ramya sriraman testbench: Change error range in check_ssim_end4 from 0.0001 to 0.001 2016-02-11
    • Kavitha Sampath analysis: share skip/merge data for rd levels 0 to 4 2016-02-05
    • Kavitha Sampath analysis: do not save best merge candidate index anymore 2016-02-08
    • Kavitha Sampath analysis: fix output mismatch in load/save 2016-02-09
    • nandaku2 Merge with stable 2016-02-11
    • Divya Manivannan rc: modify the condition for enabling 2pass in CRF mode 2016-02-11
    • Divya Manivannan rc: modify the condition for enabling 2pass in CRF mode 2016-02-09
    • Dnyaneshwar Gorade threadpool: utilize all processors on embedded ARM platforms 2016-02-09
    • Dnyaneshwar Gorade arm: added support for armv7l arch 2016-02-05
    • Ramya sriraman arm: Port sad_4x4 and sad_4x8 from x264 2016-02-04
    • Dnyaneshwar Gorade arm: Implement blockcopy_pp_16x16_neon. Modified include guards with ARM suffix. 2016-02-02
    • nandaku2 Merge with stable 2016-02-05
    • nandaku2 threadpool: stop all pools before destroying Branches stable 2016-02-05
    • nandaku2 nit: whitespace 2016-02-02
    • Sagar Kotecha analysis: Reduce data dump in analysis save mode for inter picture 2016-01-25
    • Sagar Kotecha analysis: Reduce data dump in analysis save mode for intra picture 2016-01-25
    • Sagar Kotecha clean analysis save/load read/write 2016-01-25
    • Dnyaneshwar Gorade testbench: port x264 stack & register check code for ARM arch 2016-02-01
Version 1.9+15
  • Release Date: Feb 15, 2016
    • nandaku2 Merge with stable 2016-02-11
    • Divya Manivannan rc: modify the condition for enabling 2pass in CRF mode 2016-02-11
    • Divya Manivannan rc: modify the condition for enabling 2pass in CRF mode 2016-02-09
    • Dnyaneshwar Gorade threadpool: utilize all processors on embedded ARM platforms 2016-02-09
    • Dnyaneshwar Gorade arm: added support for armv7l arch 2016-02-05
    • Ramya sriraman arm: Port sad_4x4 and sad_4x8 from x264 2016-02-04
    • Dnyaneshwar Gorade arm: Implement blockcopy_pp_16x16_neon. Modified include guards with ARM suffix. 2016-02-02
    • nandaku2 Merge with stable 2016-02-05
    • nandaku2 threadpool: stop all pools before destroying Branches stable 2016-02-05
    • nandaku2 nit: whitespace 2016-02-02
    • Sagar Kotecha analysis: Reduce data dump in analysis save mode for inter picture 2016-01-25
    • Sagar Kotecha analysis: Reduce data dump in analysis save mode for intra picture 2016-01-25
    • Sagar Kotecha clean analysis save/load read/write 2016-01-25
    • Dnyaneshwar Gorade testbench: port x264 stack & register check code for ARM arch 2016-02-01
Version 1.9+14
  • Release Date: Feb 12, 2016
    • Divya Manivannan rc: modify the condition for enabling 2pass in CRF mode 2016-02-09
    • Dnyaneshwar Gorade threadpool: utilize all processors on embedded ARM platforms 2016-02-09
    • Dnyaneshwar Gorade arm: added support for armv7l arch 2016-02-05
    • Ramya sriraman arm: Port sad_4x4 and sad_4x8 from x264 2016-02-04
    • Dnyaneshwar Gorade arm: Implement blockcopy_pp_16x16_neon. Modified include guards with ARM suffix. 2016-02-02
Version 1.9+9
  • Release Date: Feb 9, 2016
  • 1.9+9
    nandaku2 Merge with stable 2016-02-05
    nandaku2 threadpool: stop all pools before destroying Branches stable 2016-02-05
    nandaku2 nit: whitespace 2016-02-02

    1.9+7
    Sagar Kotecha analysis: Reduce data dump in analysis save mode for inter picture 2016-01-25
    Sagar Kotecha analysis: Reduce data dump in analysis save mode for intra picture 2016-01-25
    Sagar Kotecha clean analysis save/load read/write 2016-01-25
    Dnyaneshwar Gorade testbench: port x264 stack & register check code for ARM arch 2016-02-01

    x265 version 1.9 has now been released. This release supports many new features as well as additional assembly optimizations for Main12, intra prediction and SAO. Recently added features lookahead-slices, limit-refs and limit-modes have been enabled by default in the supported presets.

    =================== New Features ==============================================
    Quant offsets: This feature allows block level quantization offsets to be specified for every frame. An API-only feature.
    --intra-refresh: Keyframes can be replaced by a moving column of intra blocks in non-keyframes.
    --limit-modes: Intelligently restricts mode analysis. 
    --max-luma and --min-luma for luma clipping, optional for HDR use-cases
    Emergency denoising is now enabled by default in very low bitrate, VBV encodes
    ====================== API Changes ==============================================
    x265_frame_stats returns many additional fields: maxCLL, maxFALL, residual energy, scenecut and latency logging
    --qpfile now supports frametype 'K"
    x265 now allows CRF ratecontrol in pass N (N greater than or equal to 2)
    Chroma subsampling format YUV 4:0:0 is now fully supported and tested
    ================= Presets and Performance ==========================================
    Recently added features lookahead-slices, limit-modes, limit-refs have been enabled by default for applicable presets.
    The default psy-rd strength has been increased to 2.0
    Multi-socket machines now use a single pool of threads that can work cross-socket.
Version 1.9.7
  • Release Date: Feb 2, 2016
    • Sagar Kotecha analysis: Reduce data dump in analysis save mode for inter picture 2016-01-25
    • Sagar Kotecha analysis: Reduce data dump in analysis save mode for intra picture 2016-01-25
    • Sagar Kotecha clean analysis save/load read/write 2016-01-25
    • Dnyaneshwar Gorade testbench: port x264 stack & register check code for ARM arch 2016-02-01
Version 1.9.3
  • Release Date: Jan 30, 2016
  • x265 version 1.9 has now been released. This release supports many new features as well as additional assembly optimizations for Main12, intra prediction and SAO. Recently added features lookahead-slices, limit-refs and limit-modes have been enabled by default in the supported presets.

    =================== New Features ==============================================
    Quant offsets: This feature allows block level quantization offsets to be specified for every frame. An API-only feature.
    --intra-refresh: Keyframes can be replaced by a moving column of intra blocks in non-keyframes.
    --limit-modes: Intelligently restricts mode analysis. 
    --max-luma and --min-luma for luma clipping, optional for HDR use-cases
    Emergency denoising is now enabled by default in very low bitrate, VBV encodes
    ====================== API Changes ==============================================
    x265_frame_stats returns many additional fields: maxCLL, maxFALL, residual energy, scenecut and latency logging
    --qpfile now supports frametype 'K"
    x265 now allows CRF ratecontrol in pass N (N greater than or equal to 2)
    Chroma subsampling format YUV 4:0:0 is now fully supported and tested
    ================= Presets and Performance ==========================================
    Recently added features lookahead-slices, limit-modes, limit-refs have been enabled by default for applicable presets.
    The default psy-rd strength has been increased to 2.0
    Multi-socket machines now use a single pool of threads that can work cross-socket.
Version 1.8.221
  • Release Date: Jan 27, 2016
Version 1.8.219
  • Release Date: Jan 21, 2016
Version 1.8.212
  • Release Date: Jan 15, 2016
Version 1.8.211
  • Release Date: Jan 13, 2016
Version 1.8.210
  • Release Date: Jan 13, 2016
Version 1.8.205
  • Release Date: Jan 10, 2016
Version 1.8.203
  • Release Date: Jan 9, 2016
Version 1.8.201
  • Release Date: Jan 7, 2016
Version 1.8.198
  • Release Date: Jan 7, 2016
Version 1.8.191
  • Release Date: Jan 1, 2016
Version 1.8.190
  • Release Date: Dec 30, 2015
Version 1.8.188
  • Release Date: Dec 24, 2015
Version 1.8.186
  • Release Date: Dec 22, 2015
Version 1.8.179
  • Release Date: Dec 20, 2015
Version 1.8.169
  • Release Date: Dec 18, 2015
Version 1.8.167
  • Release Date: Dec 12, 2015
Version 1.8.133
  • Release Date: Dec 10, 2015
Version 1.8.132
  • Release Date: Dec 9, 2015
Version 1.8.130
  • Release Date: Dec 8, 2015
Version 1.8.129
  • Release Date: Dec 1, 2015
Version 1.8.127
  • Release Date: Nov 27, 2015
Version 1.8.124
  • Release Date: Nov 26, 2015
Version 1.8.118
  • Release Date: Nov 25, 2015
Version 1.8.111
  • Release Date: Nov 23, 2015
Version 1.8.109
  • Release Date: Nov 19, 2015
Version 1.8.106
  • Release Date: Nov 18, 2015
Version 1.8.93
  • Release Date: Nov 7, 2015
Version 1.8.89
  • Release Date: Nov 6, 2015
Version 1.8.78
  • Release Date: Nov 6, 2015
Version 1.8.72
  • Release Date: Nov 4, 2015
Version 1.8.60
  • Release Date: Oct 27, 2015
Version 1.8.56
  • Release Date: Oct 24, 2015
Version 1.8.52
  • Release Date: Oct 23, 2015
Version 1.8.49
  • Release Date: Oct 21, 2015
Version 1.8.46
  • Release Date: Oct 21, 2015
Version 1.8.38
  • Release Date: Oct 16, 2015
    • 10 days ago Dnyaneshwar G cleanup: align NR buffer for asm, rearrange member variables to avoid padding
    • 8 days ago Sagar Kotecha remove unnecessary function argument
    • 8 days ago Deepthi Nandakumar Merge with stable
    • 9 days ago Sagar Kotecha Fix: Provide width and height of input file to dither, param may have padded that. stable
    • 9 days ago Ashok Kumar Mishra analysis: avoid motion references not used by split blocks in pme mode stable

Version 1.7.512
  • Release Date: Oct 6, 2015
    • 22 hours ago Deepthi Nandakumar Merge with stable
    • 2 weeks ago Ramya Sriraman asm: fix sse_pp[32x64] sse2 asm for 12 bit stable
    • 23 hours ago Min Chen rdcost: fix checked build warnings, nits stable
    • 23 hours ago Steve Borho rdcost: fix 12bit checked build warnings, nits stable
    • 23 hours ago Min Chen fix Main12 satd overflow bug up to SSE4, (fixes #180) stable
    • 23 hours ago Min Chen fix SSE_PP intermediate result overflow in Main12, (fixes #180) stable
    • 24 hours ago Deepthi Nandakumar Merge with stable
    • 25 hours ago Dnyaneshwar G main12: fix intra_pred_planar8x8 overflow stable
    • 2 weeks ago Aasaipriya Chandran asm: Re-enabling asm primitives stable
    • 4 days ago Min Chen reduce operators on get DeQuant in rdoQuant()
    • 4 days ago Min Chen replace c1c2Idx by c1c2Rate
    • 4 days ago Min Chen asm: modify nquant() output to reduce abs operator in rdoQuant()
    • 13 days ago Dnyaneshwar G asm: reduced 'vpermq' instructions, improved 1040c->800c, 22%
    • 4 days ago Dnyaneshwar G asm: avx2 code for sad_x3_64xN, improved over 40% than SSE
    • 4 days ago Dnyaneshwar G asm: avx2 code for sad_x3_32xN, improved over 40% than SSE
    • 2 weeks ago Dnyaneshwar G asm: AVX2 code for pixel_var primitive, improved over 40% than SSE
    • 4 days ago Min Chen testbench: fix bug on seed 0x55EFB6F4, the numNonZero can't be Zero
    • 3 days ago Deepthi Nandakumar Merge with stable
    • 3 days ago Deepthi Nandakumar Backed out changeset: a54edf106023 stable
    • 4 days ago Deepthi Nandakumar Merge with stable
    • 4 days ago Pradeep Ramachandran nit: typo in comment stable
    • 4 days ago Min Chen fix PSYVALUE shift overflow (fixes #180) [OUTPUT CHANGE on 12bpp] stable
    • 4 days ago Min Chen testbench: fix bug on seed 0x55EFB6F4, the numNonZero can't be Zero stable
    • 5 days ago Pradeep Ramachandran nit: white space
    • 8 days ago Pradeep Ramachandran bug: Fixes with numa thread settings on windows/linux
    • 5 days ago Deepthi Nandakumar Merge with stable
    • 6 days ago Aarthi Thirumalai slicetype: identify and mark scenecuts as P slices when scenecutThreshold is 0 stable
    • 2 weeks ago Rajesh Paulraj asm: pixelavg_pp[48x64] avx2 8bpp code
    • 2 weeks ago Min Chen reduce operators on rdoQuant() compute sigHide cost
    • 2 weeks ago Min Chen reduce conditional operators in rdoQuant
    • 2 weeks ago Min Chen reduce stack space since log2TrSize is constant
    • 2 weeks ago Min Chen improve rdoQuant 10%+- by C+-+- template <log2TrSize>
    • 5 days ago Deepthi Nandakumar Merge with stable
    • 2 weeks ago Min Chen fix build error, define INT16_MAX stable
    • 13 days ago Rajesh Paulraj asm: avx2 8bpp code for chroma_p2s[16xN] for i420, i422, improved over 25% than SSE
    • 13 days ago Rajesh Paulraj asm: avx2 8bpp code for convert_p2s[16xN],improved over 25% than SSE
    • 2 weeks ago Rajesh Paulraj asm: avx2 code for sad_x4_48x64, improved over 25% than SSE
    • 2 weeks ago Rajesh Paulraj asm: avx2 code for sad_x4_64xN, improved over 40% than SSE
    • 2 weeks ago Ramya Sriraman asm: fix sse_pp[32x64] sse2 asm for 12 bit

Version 1.7.511
  • Release Date: Oct 1, 2015
    • 22 hours ago Deepthi Nandakumar Merge with stable
    • 2 weeks ago Ramya Sriraman asm: fix sse_pp[32x64] sse2 asm for 12 bit stable
    • 23 hours ago Min Chen rdcost: fix checked build warnings, nits stable
    • 23 hours ago Steve Borho rdcost: fix 12bit checked build warnings, nits stable
    • 23 hours ago Min Chen fix Main12 satd overflow bug up to SSE4, (fixes #180) stable
    • 23 hours ago Min Chen fix SSE_PP intermediate result overflow in Main12, (fixes #180) stable
    • 24 hours ago Deepthi Nandakumar Merge with stable
    • 25 hours ago Dnyaneshwar G main12: fix intra_pred_planar8x8 overflow stable
    • 2 weeks ago Aasaipriya Chandran asm: Re-enabling asm primitives stable
    • 4 days ago Min Chen reduce operators on get DeQuant in rdoQuant()
    • 4 days ago Min Chen replace c1c2Idx by c1c2Rate
    • 4 days ago Min Chen asm: modify nquant() output to reduce abs operator in rdoQuant()
    • 13 days ago Dnyaneshwar G asm: reduced 'vpermq' instructions, improved 1040c->800c, 22%
    • 4 days ago Dnyaneshwar G asm: avx2 code for sad_x3_64xN, improved over 40% than SSE
    • 4 days ago Dnyaneshwar G asm: avx2 code for sad_x3_32xN, improved over 40% than SSE
    • 2 weeks ago Dnyaneshwar G asm: AVX2 code for pixel_var primitive, improved over 40% than SSE
    • 4 days ago Min Chen testbench: fix bug on seed 0x55EFB6F4, the numNonZero can't be Zero
    • 3 days ago Deepthi Nandakumar Merge with stable
    • 3 days ago Deepthi Nandakumar Backed out changeset: a54edf106023 stable
    • 4 days ago Deepthi Nandakumar Merge with stable
    • 4 days ago Pradeep Ramachandran nit: typo in comment stable
    • 4 days ago Min Chen fix PSYVALUE shift overflow (fixes #180) [OUTPUT CHANGE on 12bpp] stable
    • 4 days ago Min Chen testbench: fix bug on seed 0x55EFB6F4, the numNonZero can't be Zero stable
    • 5 days ago Pradeep Ramachandran nit: white space
    • 8 days ago Pradeep Ramachandran bug: Fixes with numa thread settings on windows/linux
    • 5 days ago Deepthi Nandakumar Merge with stable
    • 6 days ago Aarthi Thirumalai slicetype: identify and mark scenecuts as P slices when scenecutThreshold is 0 stable
    • 2 weeks ago Rajesh Paulraj asm: pixelavg_pp[48x64] avx2 8bpp code
    • 2 weeks ago Min Chen reduce operators on rdoQuant() compute sigHide cost
    • 2 weeks ago Min Chen reduce conditional operators in rdoQuant
    • 2 weeks ago Min Chen reduce stack space since log2TrSize is constant
    • 2 weeks ago Min Chen improve rdoQuant 10%+- by C+-+- template <log2TrSize>
    • 5 days ago Deepthi Nandakumar Merge with stable
    • 2 weeks ago Min Chen fix build error, define INT16_MAX stable
    • 13 days ago Rajesh Paulraj asm: avx2 8bpp code for chroma_p2s[16xN] for i420, i422, improved over 25% than SSE
    • 13 days ago Rajesh Paulraj asm: avx2 8bpp code for convert_p2s[16xN],improved over 25% than SSE
    • 2 weeks ago Rajesh Paulraj asm: avx2 code for sad_x4_48x64, improved over 25% than SSE
    • 2 weeks ago Rajesh Paulraj asm: avx2 code for sad_x4_64xN, improved over 40% than SSE
    • 2 weeks ago Ramya Sriraman asm: fix sse_pp[32x64] sse2 asm for 12 bit
Version 1.7.509
  • Release Date: Sep 30, 2015
    • 22 hours ago Deepthi Nandakumar Merge with stable
    • 2 weeks ago Ramya Sriraman asm: fix sse_pp[32x64] sse2 asm for 12 bit stable
    • 23 hours ago Min Chen rdcost: fix checked build warnings, nits stable
    • 23 hours ago Steve Borho rdcost: fix 12bit checked build warnings, nits stable
    • 23 hours ago Min Chen fix Main12 satd overflow bug up to SSE4, (fixes #180) stable
    • 23 hours ago Min Chen fix SSE_PP intermediate result overflow in Main12, (fixes #180) stable
    • 24 hours ago Deepthi Nandakumar Merge with stable
    • 25 hours ago Dnyaneshwar G main12: fix intra_pred_planar8x8 overflow stable
    • 2 weeks ago Aasaipriya Chandran asm: Re-enabling asm primitives stable
    • 4 days ago Min Chen reduce operators on get DeQuant in rdoQuant()
    • 4 days ago Min Chen replace c1c2Idx by c1c2Rate
    • 4 days ago Min Chen asm: modify nquant() output to reduce abs operator in rdoQuant()
    • 13 days ago Dnyaneshwar G asm: reduced 'vpermq' instructions, improved 1040c->800c, 22%
    • 4 days ago Dnyaneshwar G asm: avx2 code for sad_x3_64xN, improved over 40% than SSE
    • 4 days ago Dnyaneshwar G asm: avx2 code for sad_x3_32xN, improved over 40% than SSE
    • 2 weeks ago Dnyaneshwar G asm: AVX2 code for pixel_var primitive, improved over 40% than SSE
    • 4 days ago Min Chen testbench: fix bug on seed 0x55EFB6F4, the numNonZero can't be Zero
    • 3 days ago Deepthi Nandakumar Merge with stable
    • 3 days ago Deepthi Nandakumar Backed out changeset: a54edf106023 stable
    • 4 days ago Deepthi Nandakumar Merge with stable
    • 4 days ago Pradeep Ramachandran nit: typo in comment stable
    • 4 days ago Min Chen fix PSYVALUE shift overflow (fixes #180) [OUTPUT CHANGE on 12bpp] stable
    • 4 days ago Min Chen testbench: fix bug on seed 0x55EFB6F4, the numNonZero can't be Zero stable
    • 5 days ago Pradeep Ramachandran nit: white space
    • 8 days ago Pradeep Ramachandran bug: Fixes with numa thread settings on windows/linux
    • 5 days ago Deepthi Nandakumar Merge with stable
    • 6 days ago Aarthi Thirumalai slicetype: identify and mark scenecuts as P slices when scenecutThreshold is 0 stable
    • 2 weeks ago Rajesh Paulraj asm: pixelavg_pp[48x64] avx2 8bpp code
    • 2 weeks ago Min Chen reduce operators on rdoQuant() compute sigHide cost
    • 2 weeks ago Min Chen reduce conditional operators in rdoQuant
    • 2 weeks ago Min Chen reduce stack space since log2TrSize is constant
    • 2 weeks ago Min Chen improve rdoQuant 10%+- by C+-+- template <log2TrSize>
    • 5 days ago Deepthi Nandakumar Merge with stable
    • 2 weeks ago Min Chen fix build error, define INT16_MAX stable
    • 13 days ago Rajesh Paulraj asm: avx2 8bpp code for chroma_p2s[16xN] for i420, i422, improved over 25% than SSE
    • 13 days ago Rajesh Paulraj asm: avx2 8bpp code for convert_p2s[16xN],improved over 25% than SSE
    • 2 weeks ago Rajesh Paulraj asm: avx2 code for sad_x4_48x64, improved over 25% than SSE
    • 2 weeks ago Rajesh Paulraj asm: avx2 code for sad_x4_64xN, improved over 40% than SSE
    • 2 weeks ago Ramya Sriraman asm: fix sse_pp[32x64] sse2 asm for 12 bit
Version 1.7.500
  • Release Date: Sep 26, 2015
Version 1.7.497
  • Release Date: Sep 24, 2015
Version 1.7.488
  • Release Date: Sep 23, 2015
    • 31 hours ag Min Chen update copyright header
    • 30 hours ago Min Chen rdcost: fix checked build warnings, nits
Version 1.7.482
  • Release Date: Sep 20, 2015
    • 30 hours ago Min Chen rdcost: fix checked build warnings, nits
Version 1.7.481
  • Release Date: Sep 20, 2015
    • 15 hours ago Steve Borho rdcost: fix 12bit checked build warnings, nits
    • 2 days ago Min Chen fix Main12 satd overflow bug up to SSE4, (fixes #180)
    • 2 days ago Min Chen fix SSE_PP intermediate result overflow in Main12, (fixes #180)
    • 10 days ago Pradeep Ramachandran bug: build warnings removal for linux, correct threading masks for windows.
    • 2 weeks ago Dnyaneshwar G asm: avx2 asm for intra_ang32 mode 17 & 19
    • 2 weeks ago Dnyaneshwar G asm: avx2 asm for intra_ang32 mode 16 & 20
    • 3 weeks ago Ramya Sriraman asm: fix sse_pp[16x32] sse2 asm for 12-bit
Version 1.7.478
  • Release Date: Sep 11, 2015
Version 1.7.474
  • Release Date: Sep 8, 2015
    • 4 days ago Pradeep Ramachandran Performance: Don't split threads into per-NUMA pools unless specified in cli
    • 24 hours ago Kavitha Sampath param: replace contentLightLevelInfo with seperate integers maxCLL and maxFALL
    • 24 hours ago Dnyaneshwar G Merge with stable
    • 24 hours ago Dnyaneshwar G asm: fix alignment of NR buffers, assembly code expects alignment stable
    • 4 days ago Deepthi Nandakumar Merge with stable
    • 5 days ago Deepthi Nandakumar SEI: remove recovery point SEI messages inserted every keyframe stable
    • 5 days ago Ashok Kumar Mishra Quant: fix for decoding hash mismatch and non-deterministic output in multi-socket m/c stable
    • 6 days ago Ashok Kumar Mishra Merge with stable
    • 6 days ago Ashok Kumar Mishra Quant: fix for non-deterministic ouput in lossless mode - issue #173 stable
    • 6 days ago Deepthi Nandakumar api: remove bool to prevent portability issues
    • 6 days ago Deepthi Nandakumar encoder: DCT coefficients are always 16-bit
    • 7 days ago Pradeep Ramachandran bug: Clip # threads only if # threads is > MAX_POOL_THREADS, no check earlier
    • 10 days ago Min Chen asm: new AVX2 algorithm on interp_4tap_horiz_ps_32x32, 3664c -> 3362c
    • 8 days ago Deepthi Nandakumar Merge with stable
    • 11 days ago Dnyaneshwar G asm: fix dynamic range of input to quant primitive stable
    • 11 days ago Divya Manivannan stats: add scenecut value for each frame
    • 12 days ago Dnyaneshwar G asm: avx2 asm for intra_ang32 mode 15, 5700c->1600c
    • 12 days ago Min Chen Testbench for costC1C2Flag and enable SSE2 primivite
    • 12 days ago Min Chen testbench: fix bug in check_costCoeffRemain, use different data in every iteration
    • 12 days ago Dnyaneshwar G asm: avx2 asm for intra_ang32 mode 14, 5600c->1400c
    • 12 days ago Dnyaneshwar G asm: avx2 asm for intra_ang32 mode 13, 4800c->1400c
    • 12 days ago Dnyaneshwar G asm: avx2 asm for intra_ang32 mode 12, 4758c->1474c
    • 2 weeks ago Dnyaneshwar G asm: optimized intra_ang16 mode 11 avx2 asm, 520c->370c
    • 2 weeks ago Dnyaneshwar G asm: avx2 asm for intra_ang32 mode 11, 4550c->1326c
    • 2 weeks ago Dnyaneshwar G asm: AVX2 asm for intra_ang_32 mode 10, 816c->452c
    • 2 weeks ago Dnyaneshwar G asm: AVX2 asm for intra_ang_32 mode 9, 4552c->1350c
    • 13 days ago Ashok Kumar Mishra analysis: avoid motion references not used by split blocks in pme mode
    • 12 days ago Pradeep Ramachandran Performance: Prevent small thread-pools if NUMA disabled and # CPUs > MAX_POOL_THREADS
    • 2 weeks ago Deepthi Nandakumar doc: update with VBV emergency denoising specifics
    • 13 days ago Sagar Kotecha Add emergency denoising when frame qp > QP_MAX_SPEC
    • 12 days ago Kavitha Sampath param: update calculated maxCLL, maxFALL values to contentLightLevelInfo
    • 13 days ago Min Chen faster ctxSig by replace array table_cnt by 64-bits constant
    • 2 weeks ago Rajesh Paulraj asm: avx2 code for intra_ang_16 mode 17, improved over 65% than SSE asm
    • 2 weeks ago Rajesh Paulraj asm: replace movu+-vinserti128 by vbroadcasti128 instruction
    • 2 weeks ago Min Chen reorder on intra_pred_8x8 function pointer
    • 2 weeks ago Min Chen asm: re-design AVX2 algorithm for intra_pred_8x8[20], 246c -> 194c
    • 2 weeks ago Min Chen asm: re-design AVX2 algorithm for intra_pred_8x8[16], 334c -> 200c
    • 2 weeks ago Divya Manivannan stats: average residual energy after motion prediction per frame
    • 2 weeks ago Pradeep Ramachandran Enabling the option to compile without libNUMA support in Linux.
    • 2 weeks ago Rajesh Paulraj asm: avx2 code for intra_ang_16 mode 16, improved over 63% than SSE asm
    • 2 weeks ago Rajesh Paulraj asm: avx2 code for intra_ang_16 mode 15, improved over 60% than SSE asm
    • 2 weeks ago Rajesh Paulraj asm: avx2 code for intra_ang_16 mode 14, improved over 59% than SSE asm
    • 2 weeks ago Rajesh Paulraj asm: avx2 code for intra_ang_16 mode 13, improved over 58% than SSE asm
    • 2 weeks ago Rajesh Paulraj asm: avx2 code for intra_ang_16 mode 12, improved over 43% than previous avx2 asm
    • 2 weeks ago Rajesh Paulraj asm: avx2 code for intra_ang_16 mode 9, improved over 53% than previous avx2 asm
    • 2 weeks ago Deepthi Nandakumar Merge with stable
    • 2 weeks ago Henrik Gramner checkasm: Properly save rdx/edx in checkasm_call() on x86 stable
    • 2 weeks ago Steve Borho cmake: show multilib details in build info string stable
    • 2 weeks ago Deepthi Nandakumar Merge with stable
    • 2 weeks ago Steve Borho api: move recursion check to just before generic libx265 load (fixes #171) stable
    • 2 weeks ago Deepthi Nandakumar Merge with stable
    • 2 weeks ago Aasaipriya Chandran asm: disabled 8bpp AVX & AVX2 primitives having less than 10% speed up over SSE stable
    • 2 weeks ago Deepthi Nandakumar doc: update csv description
    • 2 weeks ago Min Chen asm: Testbench for planeClipAndMax and enable it
    • 2 weeks ago Min Chen asm: new primivite planeClipAndMax for Clip Luma to custom range and statistics MaxLumaLeve
    • 2 weeks ago Divya Manivannan stats: write performance statistics per frame when --csv-log-level=2
    • 2 weeks ago Deepthi Nandakumar Merge with stable
    • 2 weeks ago Aasaipriya Chandran asm: disabled 10bpp AVX & AVX2 primitives having less than 10% speed up over SSE stable
    • 2 weeks ago Steve Borho doc: param.interlaceMode is set by option name "--interlace" stable
    • 2 weeks ago Steve Borho main12: remove scary experimental warnings
    • 2 weeks ago Steve Borho slicetype: fix compiler warning and misc nits stable
    • 2 weeks ago Steve Borho cli: derive output bit depth from --profile if --output-depth is not specified stable
    • 2 weeks ago Steve Borho rc: move stats-file write into if (outframe) conditional stable
    • 2 weeks ago Deepthi Nandakumar Merge with stable
    • 3 weeks ago Steve Borho threading: add a slow-path for platforms with no system atomics (closes #85) stable
    • 3 weeks ago Steve Borho api: prevent recursion in libx265.dll fallback (refs #156) stable
    • 3 weeks ago Min Chen asm: remove macro vpbroadcastd and fix all of invalid case on vpbroadcastd stable
    • 3 weeks ago Mahesh Pittala encoder: fix typo stable
    • 3 weeks ago Divya Manivannan rdcost: fix warning in format specifier of X265_CHECK in calcRdCost stable
    • 3 weeks ago Aarthi Thirumalai rc: fix corrupted stat file issue in multipass encodes (ref #147) stable
    • 3 weeks ago Aarthi Thirumalai rc: adjust qp for scene transitions and fade ins to avoid quality loss with vbv stable
    • 6 weeks ago Aarthi Thirumalai slicetype: Modify Scenecut algorithm to detect scene transition points stable
    • 3 weeks ago Steve Borho cmake: default ENABLE_ASSEMBLY=OFF for non-x86 builds (refs #167) stable
    • 3 weeks ago Deepthi Nandakumar Merge with stable
    • 3 weeks ago Kavitha Sampath stats: log maxCLL and maxFALL values calculated from input pictures
    • 3 weeks ago Mahesh Pittala regression: reset qg-size stable
    • 3 weeks ago Ashok Kumar Mishra analysis: skip intra if split was analyzed and no split CUs used intra in pmode
    • 3 weeks ago Ashok Kumar Mishra analysis: avoid motion references if not used by split blocks in pmode
    • 3 weeks ago Ashok Kumar Mishra analysis: re-order analysis to do splits before ME or intra for pmode
    • 3 weeks ago Deepthi Nandakumar Merge with stable
    • 3 weeks ago Steve Borho api: fallback to multilib library if profile-named library not found (refs #156) stable
    • 3 weeks ago Steve Borho api: nit stable
    • 3 weeks ago Dnyaneshwar G asm: AVX2 asm for intra_ang_32 mode 8, improved over 40% than SSE asm
    • 3 weeks ago Dnyaneshwar G asm: AVX2 asm for intra_ang_32 mode 7, improved over 40% than SSE asm
    • 3 weeks ago Dnyaneshwar G asm: AVX2 asm for intra_ang_32 mode 6, improved over 48% than SSE asm
    • 3 weeks ago Dnyaneshwar G asm: AVX2 asm for intra_ang_32 mode 5, improved over 48% than SSE asm
    • 3 weeks ago Deepthi Nandakumar picyuv: cleanup, remove always on padx, pady conditionals
    • 3 weeks ago Deepthi Nandakumar Merge with stable
    • 3 weeks ago Deepthi Nandakumar encoder: initialise cuCount
    • 3 weeks ago Kavitha Sampath stats: fix use of uninitialized pixels in maximum and average luma level calculation
    • 3 weeks ago Min Chen remove reduce operators outside loop in FrameEncoder::processRowEncoder()
    • 3 weeks ago Deepthi Nandakumar quantOffsets: move allocation to frame::create
    • 3 weeks ago Divya Manivannan stats: fix error in average calculation stable
    • 4 weeks ago Divya Manivannan Main12: fix distortion data type stable
Version 1.7.471
  • Release Date: Sep 4, 2015
Version 1.7.470
  • Release Date: Sep 3, 2015
Version 1.7.448
  • Release Date: Aug 28, 2015
Version 1.7.442
  • Release Date: Aug 25, 2015
Version 1.7.431
  • Release Date: Aug 21, 2015
    • 25 hours ago Steve Borho Merge with stable
    • 26 hours ago Steve Borho doc: fix typos and improve readability of threading chapterstable
    • 27 hours ago Steve Borho getopt: remove double underscores from prototype arguments (fixes #162)stable
    • 20 hours ago Pradeep Ramachandran cleanup: simplify slice ref lists; moving shared offset buffers into SPS
    • 27 hours ago Deepthi Nandakumar doc: edit min-luma and max-luma descriptions
    • 41 hours ago Deepthi Nandakumar Merge with stable
    • 7 days ago Santhoshini Sekar add API and implementation for Region of Interest(ROI)
    • 2 days ago Min Chen asm: new avx2 algorithm on intra_pred_ang[11], 730c -> 481c
    • 2 days ago Min Chen asm: replace LEA by ADD in macro TRANSPOSE_STORE_8x32
    • 2 days ago Dnyaneshwar G asm: AVX2 asm for intra_ang_32 mode 4, improved over 45% than SSE asm
    • 41 hours ago Kavitha Sampath api: add options max-luma and min-luma to set luma range of source pictures
    • 45 hours ago Ashok Kumar Mishra Quant: fix for decoding hash mismatch in lossless mode
Version 1.7.427
  • Release Date: Aug 21, 2015
    • 25 hours ago Steve Borho Merge with stable
    • 26 hours ago Steve Borho doc: fix typos and improve readability of threading chapterstable
    • 27 hours ago Steve Borho getopt: remove double underscores from prototype arguments (fixes #162)stable
    • 20 hours ago Pradeep Ramachandran cleanup: simplify slice ref lists; moving shared offset buffers into SPS
    • 27 hours ago Deepthi Nandakumar doc: edit min-luma and max-luma descriptions
    • 41 hours ago Deepthi Nandakumar Merge with stable
    • 7 days ago Santhoshini Sekar add API and implementation for Region of Interest(ROI)
    • 2 days ago Min Chen asm: new avx2 algorithm on intra_pred_ang[11], 730c -> 481c
    • 2 days ago Min Chen asm: replace LEA by ADD in macro TRANSPOSE_STORE_8x32
    • 2 days ago Dnyaneshwar G asm: AVX2 asm for intra_ang_32 mode 4, improved over 45% than SSE asm
    • 41 hours ago Kavitha Sampath api: add options max-luma and min-luma to set luma range of source pictures
    • 45 hours ago Ashok Kumar Mishra Quant: fix for decoding hash mismatch in lossless mode

Version 1.7.425
  • Release Date: Aug 19, 2015
    • 25 hours ago Steve Borho Merge with stable
    • 26 hours ago Steve Borho doc: fix typos and improve readability of threading chapterstable
    • 27 hours ago Steve Borho getopt: remove double underscores from prototype arguments (fixes #162)stable
    • 20 hours ago Pradeep Ramachandran cleanup: simplify slice ref lists; moving shared offset buffers into SPS
    • 27 hours ago Deepthi Nandakumar doc: edit min-luma and max-luma descriptions
    • 41 hours ago Deepthi Nandakumar Merge with stable
    • 7 days ago Santhoshini Sekar add API and implementation for Region of Interest(ROI)
    • 2 days ago Min Chen asm: new avx2 algorithm on intra_pred_ang[11], 730c -> 481c
    • 2 days ago Min Chen asm: replace LEA by ADD in macro TRANSPOSE_STORE_8x32
    • 2 days ago Dnyaneshwar G asm: AVX2 asm for intra_ang_32 mode 4, improved over 45% than SSE asm
    • 41 hours ago Kavitha Sampath api: add options max-luma and min-luma to set luma range of source pictures
    • 45 hours ago Ashok Kumar Mishra Quant: fix for decoding hash mismatch in lossless mode
Version 1.7.407
  • Release Date: Aug 14, 2015
    • 20 hours ago Pradeep Ramachandran cleanup: simplify slice ref lists; moving shared offset buffers into SPS
    • 27 hours ago Deepthi Nandakumar doc: edit min-luma and max-luma descriptions
    • 41 hours ago Deepthi Nandakumar Merge with stable
    • 7 days ago Santhoshini Sekar add API and implementation for Region of Interest(ROI)
    • 2 days ago Min Chen asm: new avx2 algorithm on intra_pred_ang[11], 730c -> 481c
    • 2 days ago Min Chen asm: replace LEA by ADD in macro TRANSPOSE_STORE_8x32
    • 2 days ago Dnyaneshwar G asm: AVX2 asm for intra_ang_32 mode 4, improved over 45% than SSE asm
    • 41 hours ago Kavitha Sampath api: add options max-luma and min-luma to set luma range of source pictures
    • 45 hours ago Ashok Kumar Mishra Quant: fix for decoding hash mismatch in lossless mode
Version 1.7.399
  • Release Date: Aug 9, 2015
    • 35 hours ago Steve Borho vui: tweak the string name of ARIB STD-B67, to avoid possible future conflicts
    • 24 hours ago Dnyaneshwar G asm: disabled AVX primitives having less than 3% speed up over SSE
    • 45 hours ago Dnyaneshwar G asm: avx2 code for pixelavg_pp 32xN & 64xN, improved over 40% than SSE
Version 1.7.393
  • Release Date: Aug 8, 2015
    • 35 hours ago Steve Borho vui: tweak the string name of ARIB STD-B67, to avoid possible future conflicts
    • 24 hours ago Dnyaneshwar G asm: disabled AVX primitives having less than 3% speed up over SSE
    • 45 hours ago Dnyaneshwar G asm: avx2 code for pixelavg_pp 32xN & 64xN, improved over 40% than SSE

Version 1.7.390
  • Release Date: Aug 6, 2015
    • 35 hours ago Steve Borho vui: tweak the string name of ARIB STD-B67, to avoid possible future conflicts
    • 24 hours ago Dnyaneshwar G asm: disabled AVX primitives having less than 3% speed up over SSE
    • 45 hours ago Dnyaneshwar G asm: avx2 code for pixelavg_pp 32xN & 64xN, improved over 40% than SSE
Version 1.7.387
  • Release Date: Aug 5, 2015
    • 15 hours ago Steve Borho threadpool: nit
    • 19 hours ago Deepthi Nandakumar vui: update help
    • 26 hours ago Deepthi Nandakumar vui: add support for transfer characteristic std-b67
    • 26 hours ago Deepthi Nandakumar param: set qgsize to default 32 for medium and all slower presets
    • 26 hours ago Deepthi Nandakumar info: add qg-size to info SEI
    • 6 days ago Divya Manivannan stats: fix loss of precision in average luma level per frame
    • 6 days ago Ashok Kumar Mishra analysis: fix for rd-0 non-deterministic output
    • 7 days ago Divya Manivannan stats: average and maximum luma level per frame
    • 7 days ago Min Chen asm: fix Main12 fault on AVX2 weight_pp
    • 7 days ago Min Chen asm: fix Main12 fault on AVX2 dequant
    • 7 days ago Min Chen update Main12 asm value range check on dequant_normal_c
    • 7 days ago Deepthi Nandakumar threadpool: fix calculation of JobProviders
    • 7 days ago Deepthi Nandakumar doc: example is for a D65, P3 color space
    • 8 days ago Deepthi Nandakumar asm: add missing prefix, remove TODO comments
    • 11 days ago Divya Manivannan rc: fix rate factor calculation after updating m_avgQpRc in rateControlEnd
    • 11 days ago Sumalatha Polureddy asm: testbench code for costCoeffRemain()
Version 1.7.382
  • Release Date: Jul 31, 2015
    • 44 hours ago Mahesh Pittala regression: add recon-y4m-exec commads
    • 44 hours ago Mahesh Pittala regression: add analysis-mode save, load commands
    • 2 days ago Divya Manivannan stats: average psyEnergy per frame
    • 22 hours ago Dnyaneshwar G main12: added lambda tables based based on qp values
Version 1.7.380
  • Release Date: Jul 29, 2015
    • 44 hours ago Mahesh Pittala regression: add recon-y4m-exec commads
    • 44 hours ago Mahesh Pittala regression: add analysis-mode save, load commands
    • 2 days ago Divya Manivannan stats: average psyEnergy per frame
    • 22 hours ago Dnyaneshwar G main12: added lambda tables based based on qp values
Version 1.7.374
  • Release Date: Jul 28, 2015
    • 44 hours ago Mahesh Pittala regression: add recon-y4m-exec commads
    • 44 hours ago Mahesh Pittala regression: add analysis-mode save, load commands
    • 2 days ago Divya Manivannan stats: average psyEnergy per frame
    • 22 hours ago Dnyaneshwar G main12: added lambda tables based based on qp values
Version 1.7.373
  • Release Date: Jul 26, 2015
    • 44 hours ago Mahesh Pittala regression: add recon-y4m-exec commads
    • 44 hours ago Mahesh Pittala regression: add analysis-mode save, load commands
    • 2 days ago Divya Manivannan stats: average psyEnergy per frame
    • 22 hours ago Dnyaneshwar G main12: added lambda tables based based on qp values

Version 1.7.367
  • Release Date: Jul 23, 2015
    • 13 hours ago Steve Borho slicetype: nit
    • 33 hours ago Min Chen asm: fix Main12 fault on saoCuOrgB0_avx2
    • 33 hours ago Min Chen asm: fix Main12 fault on intra_dc_avx2
    • 33 hours ago Min Chen asm: disable Main12 fault functions on AVX2
    • 33 hours ago Min Chen asm: fix Main12 Assembly error and disable fault functions, now we are work with assembly up to AVX
    • 5 weeks ago Santhoshini Sekar aq: new auto variance mode with biasing to dark scenes
    • 2 days ago Min Chen asm: apply new algorithm on upShift_8_sse4
    • 2 weeks ago Peter Kov+AMMAoQDFISI Fixed POWER CPU architecture detection
    • 9 days ago Divya Manivannan stats: average luma and chroma distortion per frame
    • 10 days ago Dnyaneshwar G asm: new algorithm for intra_ang_32 modes 3 & 33, improved over 50% than previous asm
Version 1.7.361
  • Release Date: Jul 22, 2015
    • 3 hours ago Min Chen asm: rewrite partial process code in upShift_8_avx2 to avoid Mac crash bug
    • 30 hours ago Min Chen asm: fix pixel_satd_4x16_sse2 overflow in Main12
    • 30 hours ago Min Chen asm: disable error SSE2 functions for Main12
    • 30 hours ago Min Chen asm: fix pixel_val_sse2 in Main12
    • 43 hours ago Divya Manivannan stats: fix skip cu count for 2 pass
    • 16 hours ago Deepthi Nandakumar doc: update limit-refs documentation

Version 1.7.357
  • Release Date: Jul 18, 2015
    • 3 hours ago Min Chen asm: rewrite partial process code in upShift_8_avx2 to avoid Mac crash bug
    • 30 hours ago Min Chen asm: fix pixel_satd_4x16_sse2 overflow in Main12
    • 30 hours ago Min Chen asm: disable error SSE2 functions for Main12
    • 30 hours ago Min Chen asm: fix pixel_val_sse2 in Main12
    • 43 hours ago Divya Manivannan stats: fix skip cu count for 2 pass
    • 16 hours ago Deepthi Nandakumar doc: update limit-refs documentation
Version 1.7.351
  • Release Date: Jul 17, 2015
    • 30 hours ago Min Chen asm: fix intra_pred_dc_sse2 in Main12
    • 23 hours ago Deepthi Nandakumar cli: add 12-bit to showHelp
    • 23 hours ago Deepthi Nandakumar dither: fix bitdepth check
    • 3 weeks ago Ashok Kumar Mishra entropy: removed g_puOffset table
    • 3 weeks ago Ashok Kumar Mishra analysis: used CUData helper function to get number of PUs and offset
    • 3 weeks ago Ashok Kumar Mishra analysis: removed switch-case to read the best ref index
    • 3 weeks ago Ashok Kumar Mishra Regression Test: added new command line --ref-limits for RD-5/6 in regression-tests.txt
    • 3 weeks ago Ashok Kumar Mishra analysis: model the effectiveness of --limit-ref with RD 5/6
    • 3 weeks ago Ashok Kumar Mishra analysis: respect X265_REF_LIMIT_DEPTH with RD 5/6
    • 3 weeks ago Ashok Kumar Mishra stats: RD 5/6 profile effectiveness of avoiding intra if split CUs did not select it
    • 3 weeks ago Ashok Kumar Mishra analysis: skip intra in RD 5/6 if split was analyzed and no split CUs used intra
    • 3 weeks ago Ashok Kumar Mishra analysis: at RD 5/6 avoid motion references if not used by split blocks
    • 7 weeks ago Ashok Kumar Mishra analysis: re-order RD 5/6 analysis to do splits before ME or intra

Version 1.7.338
  • Release Date: Jul 14, 2015
    • 6 hours ago Min Chen asm: enable SSE2 of sad[8x16]
    • 6 hours ago Min Chen asm: fix Main12 bugs in sad_mmx2 & sad_sse2
    • 6 hours ago Min Chen asm: Assembly SSE2/AVX2 for planecopy_sp_shl
    • 18 hours ago Dnyaneshwar G asm: fix saoCuStatsE0 segmentation fault in linux
    • 10 hours ago Steve Borho cmake: fix testbench linkage in multilib default build
    • 15 hours ago Steve Borho param: follow-up cleanups from --log-level frame removal
    • 19 hours ago Dnyaneshwar G asm: fix saoCuStatE1 testbench failure
    • 12 days ago Divya Manivannan stats: count of each CU partition per frame
    • 21 hours ago Deepthi Nandakumar log-level: remove X265_LOG_FRAME, frame level logging is supported in CSV only
    • 23 hours ago Dnyaneshwar G asm: fix linux build error- cannot override register size
    • 25 hours ago Deepthi Nandakumar cmake: remove empty file set
    • 4 days ago Rajesh Paulraj asm: frameInitLowres avx2 code for 8bpp and 10bpp
    • 6 days ago Dnyaneshwar G asm: sse4 code for saoCuStatsE1, improved 320369c->151086c
    • 7 days ago Dnyaneshwar G asm: sse4 code for saoCuStatsE0, improved 250341c->147284c
    • 7 days ago Dnyaneshwar G asm: sse4 code for saoCuStatsBO, improved 185378c->131279c
Version 1.7.323
  • Release Date: Jul 13, 2015
    • 8 hours ago Steve Borho cli: move CSV and dither features into x265-extras, export from CLI on Windows
    • 8 hours ago Steve Borho cmake: look for yasm 1.3.0 binaries as well as 1.2.0
Version 1.7.321
  • Release Date: Jul 12, 2015
    • 13 hours ago Steve Borho test: correctly report 12bit builds
    • 13 hours ago Steve Borho test: support test-bench builds with EXPORT_C_API=OFF (special namespaces)
    • 28 hours ago Min Chen fix pixelcmp testbench on 12bpp
    • 28 hours ago Min Chen Main12: fix up sample fault on 10b -> 12b input convert
    • 31 hours ago Min Chen Main12: comment on constant SAO_BIT_INC
    • 32 hours ago Min Chen asm: fix Main12 assembly up to SSSE3
Version 1.7.315
  • Release Date: Jul 12, 2015
    • 14 hours ago Steve Borho api: change --crop-rect to --display-window to avoid confusion
    • 30 hours ago Min Chen asm: use general constant INTERP_MAGIC_PS to replcae exact bit depth dependency magic constant
    • 30 hours ago Min Chen asm: replace tab_c_n32768 by pd_n32768
    • 30 hours ago Min Chen asm: fix Main12 luma_vps_sse2
    • 30 hours ago Min Chen improve duplicate MV check by reduce condition jump
    • 15 hours ago Steve Borho cli: move zones help into rate-control section (it does not affect slice types)
    • 15 hours ago Steve Borho cli: fix --log-level full --help by removing extra param_default() calls
    • 15 hours ago Steve Borho multilib: more portable string compare
    • 34 hours ago Steve Borho doc: nits

Version 1.7.302
  • Release Date: Jul 9, 2015
    • 11 hours ago Steve Borho cmake: keep all linker options in list, fixes multilib link
    • 13 hours ago Steve Borho cmake: disable shared lib on linux multlib for now
    • 31 hours ago Min Chen fix output mistake in Main12
    • 13 hours ago Steve Borho cmake: add exports def file for all Windows compilers, not just MSVC
    • 2 days ago Deepthi Nandakumar bitcost: avoid overflow when mvx and mvy costs are added together
Version 1.7.298
  • Release Date: Jul 8, 2015
    • 12 hours ago Steve Borho api: nits
    • 12 hours ago Steve Borho cmake: export C API from multilib, enables multibit shared libs (fixes #153)
    • 14 hours ago Steve Borho primitives: stubs for ARM cpu-detect functions (refs #155)
    • 11 hours ago Steve Borho cmake: export our public C API from our Windows CLI executable
    • 19 hours ago Dnyaneshwar G asm: fix 32-bit build error- undefined symbol r7d, r8d
    • 15 hours ago Steve Borho Merge
    • 3 days ago Divya Manivannan motion: fix overflow in mvcost() check failure.
    • 7 days ago Aasaipriya Chandran asm: avx2 code for weight_sp() 16bpp
    • 5 days ago Dnyaneshwar G sao: created new primitive for saoCuStatsBO
    • 5 days ago Dnyaneshwar G sao: created new primitive for saoCuStatsE0
    • 5 days ago Dnyaneshwar G sao: created new primitive for saoCuStatsE1
Version 1.7.286
  • Release Date: Jul 4, 2015
    • 11 hours ago Steve Borho build: msys multilib.sh needed a semicolon between extra libs
    • 12 hours ago Steve Borho level: signal general_one_picture_only_constraint_flag correctly
    • 13 hours ago Steve Borho docs: further improve profile determination documentation
    • 13 hours ago Steve Borho level: detect still picture profile based on just -f 1, other fixes
    • 13 hours ago Steve Borho param: expose param.totalFrames as param_parse option name 'total-frames'
    • 26 hours ago Deepthi Nandakumar sao: cleanup, remove unnecessary rdoInit, no output change

Version 1.7.281
  • Release Date: Jul 3, 2015
    • 7 hours ago Min Chen asm: fix Main12 luma_hps_sse2
    • 9 hours ago Min Chen asm: fix Main12 ssim_end_4
    • 8 hours ago Min Chen asm: main12 dynamic range fixes for DCT
    • 9 hours ago Min Chen main12: pass correct bit-depth to assembler
    • 9 hours ago Steve Borho docs: mention main12 and intra profile considerations
    • 10 hours ago Steve Borho cli: nit
    • 10 hours ago Steve Borho cli: suggest actual help command for help
    • 11 hours ago Steve Borho api: emit a frightening warning when Main12 is used
    • 11 hours ago Steve Borho level: plumb in support for Main12 and *-Intra profiles
    • 3 days ago Min Chen multilib: static functions to avoid link conflict (closes #148)
    • 36 hours ago Steve Borho cmake: plumb in build support for Main12
    • 31 hours ago Min Chen asm: fix buffer overwrite bug in upShift_8_avx2
    • 31 hours ago Min Chen testbench: fix bug in plane_copy_cp, detect outside bound now
    • 47 hours ago Rajesh Paulraj asm: pixelavg_pp[8xN] sse2 code for 10bpp
    • 2 days ago Dnyaneshwar G asm: intra_filter4x4 avx2 code, improved 8bit: 141c->118c, 10bit: 121c->88c

Version 1.7.266
  • Release Date: Jul 2, 2015
    • 20 hours ago Deepthi Nandakumar Merge with stable
    • 26 hours ago Deepthi Nandakumar param: tune grain disables rdoq-level. This provides better visual quality resultsstable
Version 1.7.265
  • Release Date: Jul 2, 2015
    • 13 hours ago Steve Borho version: cleanup macro names (remove HM oddness)
    • 13 hours ago Steve Borho cmake: remove dead LIBOCOMMON variables
    • 13 hours ago Steve Borho common: move no-asm indication into build info string
    • 14 hours ago Steve Borho cli: split x265_report_simd() from x265_setup_primitives()
    • 14 hours ago Steve Borho primitives: remove cpuid parameter from x265_setup_primitives()
    • 15 hours ago Steve Borho cli: fix multilib CPU detect, use detected cpuid from param
    • 43 hours ago Dnyaneshwar G asm: intra_filter 10bpp sse4 code
    • 4 days ago Rajesh Paulraj asm: avx2 10bit code for planecopy_cp(10660.20 -> 5685.80)
    • 36 hours ago Min Chen asm: new algorithm x265_count_nonzero_4x4_avx2, Issue #152
    • 35 hours ago Steve Borho api: nits
Version 1.7.257
  • Release Date: Jun 30, 2015
    • 15 hours ago Steve Borho common: fix for multilib checked builds, move g_checkFailures within namespace
    • 3 days ago Aarthi Thirumalai rc: fixes inconsistent output in linux because of RC Lock in CQP/CRF
    • 4 days ago Gopu Govindaswamy motion: remove mvc's sad cost calc for lowres, already measured in slicetype
    • 3 days ago Min Chen asm: cleanup unused constant and update copyright header
    • 3 days ago Dnyaneshwar G asm: intra_filter32x32 sse4 code, improved 4050c->652c over C code
    • 3 days ago Dnyaneshwar G asm: intra_filter16x16 sse4 code, improved 1952c->351c over C code
    • 3 days ago Dnyaneshwar G asm: intra_filter8x8 sse4 code, improved 990c->201c over C code
    • 3 days ago Dnyaneshwar G asm: intra_filter4x4 sse4 code and added testbench support, improved 357c->141c over C code
    • 4 days ago Rajesh Paulraj asm: sse4 10bit code for sign primitive
    • 4 days ago Rajesh Paulraj asm: avx2 10bit code for sign primitive(356.91 -> 242.00)
    • 3 days ago Dnyaneshwar G asm: fix gcc build error, invalid size for operand 1
    • 4 days ago Deepthi Nandakumar cmake: allow the CLI option to be cached
    • 5 days ago Dnyaneshwar G asm: 10bpp AVX2 code for saoCuOrgB0, improved 23127c->15595c over SSE
    • 5 days ago Dnyaneshwar G asm: 10bpp AVX2 code for saoCuOrgE3
    • 5 days ago Dnyaneshwar G asm: 10bpp AVX2 code for saoCuOrgE2
    • 5 days ago Dnyaneshwar G asm: 10bpp AVX2 code for saoCuOrgE1_2Rows, improved 900c->614c over SSE
    • 5 days ago Dnyaneshwar G asm: 10bpp AVX2 code for saoCuOrgE1, improved 492c->360c over SSE
    • 5 days ago Dnyaneshwar G asm: 10bpp AVX2 code for saoCuOrgE0, improved 974c->690c over SSE
    • 5 days ago Rajesh Paulraj asm: pixelavg_pp[12x16],[24x32] avx2 code for 10bpp
    • 5 days ago Rajesh Paulraj asm: pixelavg_pp[32xN],[64xN],48x64 avx2 code for 10bpp
    • 4 days ago Deepthi Nandakumar multilib: remove WINXP=ON from multilib scripts

Version 1.7.247
  • Release Date: Jun 27, 2015
    • 15 hours ago Steve Borho param: declare our custom strtok_r file-local to avoid multilib breakage
    • 15 hours ago Steve Borho threading: fix 32bit multilib with GCC
    • 15 hours ago Steve Borho cmake: further cleanups for high-bit-depth comment and desc string
    • 2 days ago Min Chen asm: improve AVX2 sad_x4[32xN] by new faster algorithm
    • 2 days ago Min Chen asm: AVX2 of SAD_x4[32xN]
    • 2 days ago Min Chen reduce shift operator on coeff remain code
    • 2 days ago Min Chen faster algorithm to calculate ctxSet in codeCoeffNxN()
    • 2 days ago Rajesh Paulraj asm: pixelavg_pp[16xN] avx2 code for 10bpp
    • 2 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgB0, improved 173346c->23127c over C code
    • 2 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgE3
    • 2 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgE2
    • 2 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgE1_2Rows, improved 10095c->900c, over C code
    • 3 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgE1, improved 5017c->470c, over C code
    • 5 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgE0, improved 8740c->974c, over C code
Version 1.7.243
  • Release Date: Jun 27, 2015
    • 15 hours ago Steve Borho param: declare our custom strtok_r file-local to avoid multilib breakage
    • 15 hours ago Steve Borho threading: fix 32bit multilib with GCC
    • 15 hours ago Steve Borho cmake: further cleanups for high-bit-depth comment and desc string
    • 2 days ago Min Chen asm: improve AVX2 sad_x4[32xN] by new faster algorithm
    • 2 days ago Min Chen asm: AVX2 of SAD_x4[32xN]
    • 2 days ago Min Chen reduce shift operator on coeff remain code
    • 2 days ago Min Chen faster algorithm to calculate ctxSet in codeCoeffNxN()
    • 2 days ago Rajesh Paulraj asm: pixelavg_pp[16xN] avx2 code for 10bpp
    • 2 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgB0, improved 173346c->23127c over C code
    • 2 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgE3
    • 2 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgE2
    • 2 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgE1_2Rows, improved 10095c->900c, over C code
    • 3 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgE1, improved 5017c->470c, over C code
    • 5 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgE0, improved 8740c->974c, over C code
Version 1.7.234
  • Release Date: Jun 26, 2015
    • 15 hours ago Steve Borho param: declare our custom strtok_r file-local to avoid multilib breakage
    • 15 hours ago Steve Borho threading: fix 32bit multilib with GCC
    • 15 hours ago Steve Borho cmake: further cleanups for high-bit-depth comment and desc string
    • 2 days ago Min Chen asm: improve AVX2 sad_x4[32xN] by new faster algorithm
    • 2 days ago Min Chen asm: AVX2 of SAD_x4[32xN]
    • 2 days ago Min Chen reduce shift operator on coeff remain code
    • 2 days ago Min Chen faster algorithm to calculate ctxSet in codeCoeffNxN()
    • 2 days ago Rajesh Paulraj asm: pixelavg_pp[16xN] avx2 code for 10bpp
    • 2 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgB0, improved 173346c->23127c over C code
    • 2 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgE3
    • 2 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgE2
    • 2 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgE1_2Rows, improved 10095c->900c, over C code
    • 3 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgE1, improved 5017c->470c, over C code
    • 5 days ago Dnyaneshwar G asm: 10bpp sse4 code for saoCuOrgE0, improved 8740c->974c, over C code
Version 1.7.231
  • Release Date: Jun 24, 2015
    • 14 hours ago Steve Borho doc: fix reST section quote
    • 27 hours ago Steve Borho doc: attempt to document statically linked multi-library implications
    • 27 hours ago Steve Borho doc: nit
    • 27 hours ago Steve Borho doc: fix dither docs on api page (not applicable if internal bit depth != 8)
    • 29 hours ago Steve Borho cmake: change multilib namespaces to x265_8bit/x265_10bit
    • 29 hours ago Steve Borho cmake: change msvc multilib build folders to 8bit/10bit
    • 29 hours ago Steve Borho param: clarify 8bit/10bit in logs
    • 29 hours ago Steve Borho clarify 8bit/10bit in code comments
    • 29 hours ago Steve Borho doc: replace 'bpp' in docs with 'bit' (do not imply pixels
Version 1.7.211
  • Release Date: Jun 24, 2015
    • 7 hours ago Steve Borho cpu: fix multilib compiling for some rarer build options
    • 11 hours ago Steve Borho cmake: make it more clear that the extra libs are only linked to the CLI
    • 11 hours ago Steve Borho cmake: remove hard-coded link dir hack, add multilib batch file for MSYS
    • 11 hours ago Min Chen cmake: introduce multilib support for MSVC
    • 20 hours ago Deepthi Nandakumar winxp: fix typo
    • 25 hours ago Deepthi Nandakumar winxp: partial fix for Issue #146, rename x265 to X265_NS
    • 26 hours ago David T Yuen asm: intrapred_angX_4x4 sse2 performance tweaks 10-bit
    • 29 hours ago David T Yuen asm: intrapred_angX_4x4 sse2 performance tweaks
    • 3 days ago Min Chen testbench: costCoeffNxN and enable asm code (based on Sumalatha's patch)
    • 3 days ago Praveen Tiwari param: move x265_atof into namespace "X265_NS"
    • 3 days ago Deepthi Nandakumar doc: update limit-refs behaviour for intra modes
    • 2 weeks ago Ashok Kumar Mishra analysis: allow intra mode in RD-0/4

Version 1.7.207
  • Release Date: Jun 22, 2015
    • 12 hours ago Steve Borho asm: remove useless comments
    • 11 hours ago Steve Borho cli: fix multilib link
    • 11 hours ago Steve Borho asm: fix multilib link
    • 12 hours ago Steve Borho build: fix exec bit of shell scripts
    • 22 hours ago Deepthi Nandakumar Merge with stable
    • 22 hours ago Deepthi Nandakumar doc: update strong intra smoothing explanationstable
    • 27 hours ago Dnyaneshwar G asm: avx2 code for dequant_scaling, improved 11097c->6860c, 38% over SSE4
    • 44 hours ago Dnyaneshwar G asm: dequant_scaling asm code, improved 12668c->11097c, 12% over intrinsic
    • 34 hours ago Min Chen improve fillReferenceSamples by reduce condition operators in loop
    • 47 hours ago Sumalatha Polureddy asm: avx2 code for weight_pp() for 10 bpp
    • 2 days ago Min Chen reduce VC condition branch by modify code style
    • 2 days ago Min Chen improve by convert arithmetic(signed) shift to logic(unsigned) shift
    • 2 days ago Min Chen faster algorithm to calculate signHidden cost in codeCoeffNxN()
    • 2 days ago Min Chen clean up debug code in codeCoeffNxN()
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 18, improved 1331c->884c, 31%
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 17 & 19
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 16 & 20
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 15 & 21
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 14 & 22
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 13 & 23
    • 39 hours ago Deepthi Nandakumar Merge with stable
    • 39 hours ago Deepthi Nandakumar analysis-mode: fix blocking artifacts in analysis-mode load/savestable
Version 1.7.201
  • Release Date: Jun 20, 2015
    • 12 hours ago Steve Borho asm: remove useless comments
    • 11 hours ago Steve Borho cli: fix multilib link
    • 11 hours ago Steve Borho asm: fix multilib link
    • 12 hours ago Steve Borho build: fix exec bit of shell scripts
    • 22 hours ago Deepthi Nandakumar Merge with stable
    • 22 hours ago Deepthi Nandakumar doc: update strong intra smoothing explanationstable
    • 27 hours ago Dnyaneshwar G asm: avx2 code for dequant_scaling, improved 11097c->6860c, 38% over SSE4
    • 44 hours ago Dnyaneshwar G asm: dequant_scaling asm code, improved 12668c->11097c, 12% over intrinsic
    • 34 hours ago Min Chen improve fillReferenceSamples by reduce condition operators in loop
    • 47 hours ago Sumalatha Polureddy asm: avx2 code for weight_pp() for 10 bpp
    • 2 days ago Min Chen reduce VC condition branch by modify code style
    • 2 days ago Min Chen improve by convert arithmetic(signed) shift to logic(unsigned) shift
    • 2 days ago Min Chen faster algorithm to calculate signHidden cost in codeCoeffNxN()
    • 2 days ago Min Chen clean up debug code in codeCoeffNxN()
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 18, improved 1331c->884c, 31%
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 17 & 19
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 16 & 20
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 15 & 21
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 14 & 22
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 13 & 23
    • 39 hours ago Deepthi Nandakumar Merge with stable
    • 39 hours ago Deepthi Nandakumar analysis-mode: fix blocking artifacts in analysis-mode load/savestable
Version 1.7.199
  • Release Date: Jun 19, 2015
    • 12 hours ago Steve Borho asm: remove useless comments
    • 11 hours ago Steve Borho cli: fix multilib link
    • 11 hours ago Steve Borho asm: fix multilib link
    • 12 hours ago Steve Borho build: fix exec bit of shell scripts
    • 22 hours ago Deepthi Nandakumar Merge with stable
    • 22 hours ago Deepthi Nandakumar doc: update strong intra smoothing explanationstable
    • 27 hours ago Dnyaneshwar G asm: avx2 code for dequant_scaling, improved 11097c->6860c, 38% over SSE4
    • 44 hours ago Dnyaneshwar G asm: dequant_scaling asm code, improved 12668c->11097c, 12% over intrinsic
    • 34 hours ago Min Chen improve fillReferenceSamples by reduce condition operators in loop
    • 47 hours ago Sumalatha Polureddy asm: avx2 code for weight_pp() for 10 bpp
    • 2 days ago Min Chen reduce VC condition branch by modify code style
    • 2 days ago Min Chen improve by convert arithmetic(signed) shift to logic(unsigned) shift
    • 2 days ago Min Chen faster algorithm to calculate signHidden cost in codeCoeffNxN()
    • 2 days ago Min Chen clean up debug code in codeCoeffNxN()
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 18, improved 1331c->884c, 31%
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 17 & 19
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 16 & 20
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 15 & 21
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 14 & 22
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 13 & 23
    • 39 hours ago Deepthi Nandakumar Merge with stable
    • 39 hours ago Deepthi Nandakumar analysis-mode: fix blocking artifacts in analysis-mode load/savestable

Version 1.7.195
  • Release Date: Jun 19, 2015
    • 25 hours ago Dnyaneshwar G fix issue #143 x265 is slow when it is build with GCC 5.1
    • 10 days ago Steve Borho build: introduce a new multilib build folder for gmake environments
    • 10 days ago Steve Borho cli: retrieve build strings from api pointer, rather than exported symbols
    • 21 hours ago Deepthi Nandakumar version: X265_NS prefix for build strings, max bit depth [credited to Steve Borho]
    • 21 hours ago Deepthi Nandakumar dct: rename g_entropyStateBits[credited to Steve Borho]
    • 11 days ago Steve Borho common: move remaining x265_ functions into private namespace
    • 11 days ago Steve Borho cmake: ugly hacks for adding arbitrary link library
    • 25 hours ago Deepthi Nandakumar asm: prefix primitives with X265_NS[credited to Kevin Wu]
    • 2 days ago Aasaipriya Chandran asm: avx2 interp_8tap_hv_pp for 16bpp
    • 4 days ago Min Chen move firstC2Idx, firstC2Flag and c1Next from common to local
    • 4 days ago Min Chen asm: new SSE2 primivite costC1C2Flag in codeCoeffNxN()
    • 4 days ago Aasaipriya Chandran asm: avx2 interp_8tap_hv_pp for 8bpp
    • 5 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 12 & 24
    • 6 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 11 & 25

Version 1.7.177
  • Release Date: Jun 17, 2015
    • 25 hours ago Dnyaneshwar G fix issue #143 x265 is slow when it is build with GCC 5.1
    • 10 days ago Steve Borho build: introduce a new multilib build folder for gmake environments
    • 10 days ago Steve Borho cli: retrieve build strings from api pointer, rather than exported symbols
    • 21 hours ago Deepthi Nandakumar version: X265_NS prefix for build strings, max bit depth [credited to Steve Borho]
    • 21 hours ago Deepthi Nandakumar dct: rename g_entropyStateBits[credited to Steve Borho]
    • 11 days ago Steve Borho common: move remaining x265_ functions into private namespace
    • 11 days ago Steve Borho cmake: ugly hacks for adding arbitrary link library
    • 25 hours ago Deepthi Nandakumar asm: prefix primitives with X265_NS[credited to Kevin Wu]
    • 2 days ago Aasaipriya Chandran asm: avx2 interp_8tap_hv_pp for 16bpp
    • 4 days ago Min Chen move firstC2Idx, firstC2Flag and c1Next from common to local
    • 4 days ago Min Chen asm: new SSE2 primivite costC1C2Flag in codeCoeffNxN()
    • 4 days ago Aasaipriya Chandran asm: avx2 interp_8tap_hv_pp for 8bpp
    • 5 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 12 & 24
    • 6 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 11 & 25

Version 1.7.169
  • Release Date: Jun 16, 2015
    • 22 hours ago Divya Manivannan stats: write encoder summary in the console to libx265
    • 22 hours ago Divya Manivannan stats: fix build warning in mingw
    • 44 hours ago Divya Manivannan stats: remove frame level console logging
    • 2 days ago Divya Manivannan stats: CLI now dumps frame level statistics into csv file
    • 2 days ago Divya Manivannan stats: add x265_frame_stats structure to x265_picture
    • 2 days ago Divya Manivannan cli: CLI-only option --csv-log-level now indicates level of csv logging
    • 2 days ago Divya Manivannan cli: --csv is now a CLI only option
    • 2 days ago Divya Manivannan stats: move FrameStats from ratecontrol.h to framedata.h
    • 2 days ago David T Yuen asm: dst4 sse2 8bpp and 10bpp
    • 2 days ago Rajesh Paulraj asm: interp_4tap_vert_X[6xN] avx2 10bit code for i420,i422
    • 46 hours ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vps/vsp/vss[4x4, 4x8, 4x16][i444] 16bpp
    • 46 hours ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vps/vsp/vss[8x4, 8x8, 8x16, 8x32][i444] 16bpp
    • 46 hours ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vps/vsp/vss[8x4, 8x8, 8x16, 8x32][i422] 16bpp
    • 2 days ago Aasaipriya Chandran asm: chroma_vpp/vps/vsp/vss[4x4, 4x8, 4x16][i422] avx2 code for 16bpp
    • 46 hours ago Aasaipriya Chandran asm: chroma_vpp/vps/vsp/vss[8x12][i422] avx2 code for 16bpp
    • 47 hours ago Aasaipriya Chandran asm: chroma_vpp/vss/vsp/vpp[8x4][i420] 16bpp
    • 47 hours ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vps/vss/vsp[8x6][i420] 16bpp
    • 47 hours ago Aasaipriya Chandran asm:avx2 code for chroma_vpp/vps/vsp/vss[8x8][i420] 16bpp
    • 47 hours ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vsp/vps/vss[4x16][i420] and chroma_vpp/vsp/vps/vss[4x32][i422] for 16bpp
    • 2 days ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vsp/vps/vss[4x8][i420] 16bpp
    • 2 days ago Aasaipriya Chandran asm: chroma_vpp/vps/vsp/vss[4x2, 4x4, 8x2][i420] 16bpp - modification of doClip variable
    • 2 days ago David T Yuen asm: count_nonzero ssse3 to sse2
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 9,10,26 & 27
    • 2 days ago Rajesh Paulraj asm: interp_4tap_vert_X[64xN] avx2 10bit code for i444
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 8 & 28
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 7 & 29
    • 3 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 6 & 30
    • 3 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 5 & 31
    • 3 days ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vsp/vps/vss[4x4][i420] 16bpp
    • 3 days ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vsp/vps/vss[4x2][i420] 16bpp
    • 3 days ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vps/vss/vsp[8x2][i420] 16bpp
    • 3 days ago Min Chen asm: improve costCoeffRemain by bypass uncoded coeff
    • 3 days ago Min Chen split fast RD path in codeCoeffNxN()
    • 3 days ago Min Chen reduce conditional jump in codeCoeffNxN()
    • 3 days ago Min Chen move variant from common to local in codeCoeffNxN()
    • 3 days ago Deepthi Nandakumar primitives: disable costCoeffRemain until foolproof testcases have been added
    • 3 days ago Min Chen asm: SSE4 of costCoeffRemain in codeCoeffNxN()
    • 3 days ago Divya Manivannan stats: rename fields of FrameStats structure
    • 3 days ago Dnyaneshwar G asm: avx2 code for intra_pred_ang32x32 mode 4 & 32, improved 8709c->4618c, 47%
    • 3 days ago Rajesh Paulraj asm: interp_4tap_vert_X[i444][16xN, 32xN, 12x16, 24x32] avx2 10bit code
    • 3 days ago Rajesh Paulraj asm: interp_4tap_vert_X[i422][16xN, 32xN, 12x32, 24x64] avx2 10bit code
    • 3 days ago Rajesh Paulraj asm: interp_4tap_vert_X[48x64] avx2 10bit code for i444
    • 3 days ago Rajesh Paulraj asm: interp_4tap_vert_X[12x16],[24x32] avx2 10bit code for i420
    • 3 days ago Rajesh Paulraj asm: interp_4tap_vert_X[32xN] avx2 10bit code for i420
    • 4 days ago Dnyaneshwar G asm: avx2 code for intra_pred_ang32x32 mode 2,3,33 & 34
    • 7 days ago Steve Borho multilib: add shim implementations of API introspection functions
    • 7 days ago Steve Borho param: make encoder lib use private param functions when not exporting C API
    • 7 days ago Steve Borho cmake: introduce EXPORT_C_API build variable
    • 7 days ago Steve Borho cli: hoist param_default() call from showHelp
    • 4 days ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vps/vsp/vss[8x16, 8x32] i420 and chroma_vpp/vps/vsp/vss[8x64]for i422 for 16bpp
    • 7 days ago Rajesh Paulraj asm: interp_4tap_vert_X[16xN] avx2 10bit code for i420
    • 9 days ago Rajesh Paulraj asm: align tab_ChromaCoeffV constant to 32 bytes bound
    • 7 days ago Divya Manivannan stats: remove --cu-stats option
    • 9 days ago Divya Manivannan smoke-test: remove --cu-stats option from the command line
Version 1.7.166
  • Release Date: Jun 14, 2015
    • 22 hours ago Divya Manivannan stats: write encoder summary in the console to libx265
    • 22 hours ago Divya Manivannan stats: fix build warning in mingw
    • 44 hours ago Divya Manivannan stats: remove frame level console logging
    • 2 days ago Divya Manivannan stats: CLI now dumps frame level statistics into csv file
    • 2 days ago Divya Manivannan stats: add x265_frame_stats structure to x265_picture
    • 2 days ago Divya Manivannan cli: CLI-only option --csv-log-level now indicates level of csv logging
    • 2 days ago Divya Manivannan cli: --csv is now a CLI only option
    • 2 days ago Divya Manivannan stats: move FrameStats from ratecontrol.h to framedata.h
    • 2 days ago David T Yuen asm: dst4 sse2 8bpp and 10bpp
    • 2 days ago Rajesh Paulraj asm: interp_4tap_vert_X[6xN] avx2 10bit code for i420,i422
    • 46 hours ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vps/vsp/vss[4x4, 4x8, 4x16][i444] 16bpp
    • 46 hours ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vps/vsp/vss[8x4, 8x8, 8x16, 8x32][i444] 16bpp
    • 46 hours ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vps/vsp/vss[8x4, 8x8, 8x16, 8x32][i422] 16bpp
    • 2 days ago Aasaipriya Chandran asm: chroma_vpp/vps/vsp/vss[4x4, 4x8, 4x16][i422] avx2 code for 16bpp
    • 46 hours ago Aasaipriya Chandran asm: chroma_vpp/vps/vsp/vss[8x12][i422] avx2 code for 16bpp
    • 47 hours ago Aasaipriya Chandran asm: chroma_vpp/vss/vsp/vpp[8x4][i420] 16bpp
    • 47 hours ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vps/vss/vsp[8x6][i420] 16bpp
    • 47 hours ago Aasaipriya Chandran asm:avx2 code for chroma_vpp/vps/vsp/vss[8x8][i420] 16bpp
    • 47 hours ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vsp/vps/vss[4x16][i420] and chroma_vpp/vsp/vps/vss[4x32][i422] for 16bpp
    • 2 days ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vsp/vps/vss[4x8][i420] 16bpp
    • 2 days ago Aasaipriya Chandran asm: chroma_vpp/vps/vsp/vss[4x2, 4x4, 8x2][i420] 16bpp - modification of doClip variable
    • 2 days ago David T Yuen asm: count_nonzero ssse3 to sse2
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 9,10,26 & 27
    • 2 days ago Rajesh Paulraj asm: interp_4tap_vert_X[64xN] avx2 10bit code for i444
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 8 & 28
    • 2 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 7 & 29
    • 3 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 6 & 30
    • 3 days ago Dnyaneshwar G asm: 10bpp avx2 code for intra_pred_ang32x32 mode 5 & 31
    • 3 days ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vsp/vps/vss[4x4][i420] 16bpp
    • 3 days ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vsp/vps/vss[4x2][i420] 16bpp
    • 3 days ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vps/vss/vsp[8x2][i420] 16bpp
    • 3 days ago Min Chen asm: improve costCoeffRemain by bypass uncoded coeff
    • 3 days ago Min Chen split fast RD path in codeCoeffNxN()
    • 3 days ago Min Chen reduce conditional jump in codeCoeffNxN()
    • 3 days ago Min Chen move variant from common to local in codeCoeffNxN()
    • 3 days ago Deepthi Nandakumar primitives: disable costCoeffRemain until foolproof testcases have been added
    • 3 days ago Min Chen asm: SSE4 of costCoeffRemain in codeCoeffNxN()
    • 3 days ago Divya Manivannan stats: rename fields of FrameStats structure
    • 3 days ago Dnyaneshwar G asm: avx2 code for intra_pred_ang32x32 mode 4 & 32, improved 8709c->4618c, 47%
    • 3 days ago Rajesh Paulraj asm: interp_4tap_vert_X[i444][16xN, 32xN, 12x16, 24x32] avx2 10bit code
    • 3 days ago Rajesh Paulraj asm: interp_4tap_vert_X[i422][16xN, 32xN, 12x32, 24x64] avx2 10bit code
    • 3 days ago Rajesh Paulraj asm: interp_4tap_vert_X[48x64] avx2 10bit code for i444
    • 3 days ago Rajesh Paulraj asm: interp_4tap_vert_X[12x16],[24x32] avx2 10bit code for i420
    • 3 days ago Rajesh Paulraj asm: interp_4tap_vert_X[32xN] avx2 10bit code for i420
    • 4 days ago Dnyaneshwar G asm: avx2 code for intra_pred_ang32x32 mode 2,3,33 & 34
    • 7 days ago Steve Borho multilib: add shim implementations of API introspection functions
    • 7 days ago Steve Borho param: make encoder lib use private param functions when not exporting C API
    • 7 days ago Steve Borho cmake: introduce EXPORT_C_API build variable
    • 7 days ago Steve Borho cli: hoist param_default() call from showHelp
    • 4 days ago Aasaipriya Chandran asm: avx2 code for chroma_vpp/vps/vsp/vss[8x16, 8x32] i420 and chroma_vpp/vps/vsp/vss[8x64]for i422 for 16bpp
    • 7 days ago Rajesh Paulraj asm: interp_4tap_vert_X[16xN] avx2 10bit code for i420
    • 9 days ago Rajesh Paulraj asm: align tab_ChromaCoeffV constant to 32 bytes bound
    • 7 days ago Divya Manivannan stats: remove --cu-stats option
    • 9 days ago Divya Manivannan smoke-test: remove --cu-stats option from the command line
Version 1.7.135
  • Release Date: Jun 11, 2015
    • David T Yuen testbench: add missing index for chromaPartStr
Version 1.7.109
  • Release Date: Jun 7, 2015
    • David T Yuen testbench: add missing index for chromaPartStr
Version 1.7.108
  • Release Date: Jun 6, 2015
    • 12 hours ago Sumalatha Polureddy asm: avx2 code for chroma sse_pp[16x32, 32x64] for i422default tip
    • 20 hours ago Min Chen fix Issue #141: TestBench failed on AMD FX8350
    • 17 hours ago David T Yuen asm: filterPixelToShort 8-bit and 10-bit sse2
    • 23 hours ago Min Chen asm: AVX2 of sa8d[32x32], 7.6K -> 6.7K cycles
    • 23 hours ago Min Chen move tmpCoeff outside fast RD path
    • 23 hours ago Min Chen merge NxN and 4x4 fast RD path
    • 23 hours ago Min Chen merge ctxIndMap4x4[] into tab_cnt[][]
    • 23 hours ago Min Chen asm: reuse costCoeffGroupNxN in 4x4 path
    • 23 hours ago Min Chen asm: sse4 version of costCoeffGroupNxN in codeCoeffNxN
    • 27 hours ago Aasaipriya Chandran asm: chroma_hps[6x8, 6x16] for high bit depth
    • 32 hours ago Sumalatha Polureddy asm: avx2 code for chroma sse_pp[16x16, 32x32] for i420
    • 35 hours ago Sumalatha Polureddy asm: avx2 code for denoiseDct for 10 bpp

Version 1.7.86
  • Release Date: May 29, 2015
    • 19 hours ago Gopu Govindaswamy encoder: fix line feed issue
    • 41 hours ago Ashok Kumar MishraRegression Test: added new command line --ref-limits in regression-tests.txt
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for sse_ss[64x64] for 10 bpp (7039 -> 4051)
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for sse_ss[32x32] for 10 bpp (1935 -> 1030)

Version 1.7.58
  • Release Date: May 27, 2015
    • 19 hours ago Gopu Govindaswamy encoder: fix line feed issue
    • 41 hours ago Ashok Kumar MishraRegression Test: added new command line --ref-limits in regression-tests.txt
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for sse_ss[64x64] for 10 bpp (7039 -> 4051)
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for sse_ss[32x32] for 10 bpp (1935 -> 1030)
Version 1.7.51
  • Release Date: May 23, 2015
    • 19 hours ago Gopu Govindaswamy encoder: fix line feed issue
    • 41 hours ago Ashok Kumar MishraRegression Test: added new command line --ref-limits in regression-tests.txt
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for sse_ss[64x64] for 10 bpp (7039 -> 4051)
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for sse_ss[32x32] for 10 bpp (1935 -> 1030)

Version 1.7.41
  • Release Date: May 22, 2015
    • 19 hours ago Gopu Govindaswamy encoder: fix line feed issue
    • 41 hours ago Ashok Kumar MishraRegression Test: added new command line --ref-limits in regression-tests.txt
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for sse_ss[64x64] for 10 bpp (7039 -> 4051)
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for sse_ss[32x32] for 10 bpp (1935 -> 1030)

Version 1.7.37
  • Release Date: May 22, 2015
    • 13 hours ago Steve Borho doc: update --tune grain docs
    • 14 hours ago Steve Borho Merge with stable
    • 15 hours ago Steve Borho cli: delay calling showHelp until a param is allocated and defaultedstable
    • 29 hours ago David T Yuen asm: interp_4tap_vert_pX_4xN sse2
    • 15 hours ago Steve Borho cli: delay calling showHelp until a param is allocated and defaulted
    • 2 months ago Deepthi Nandakumar docs: document --limit-refs
    • 20 hours ago Deepthi Nandakumar analysis: re-order cost calculation for early-outs
    • 2 days ago Ashok Kumar Mishra analysis: model the effectiveness of --limit-ref with RD 0/4
    • 2 days ago Ashok Kumar Mishra stats: with the CU reference limit, even 8x8 can have skipped motion searches
    • 2 days ago Ashok Kumar Mishra cli: connect --limit-refs to param.limitReferences
    • 2 days ago Ashok Kumar Mishra analysis: respect X265_REF_LIMIT_DEPTH with RD 0/4
    • 2 days ago Ashok Kumar Mishra stats: RD 0/4 profile effectiveness of avoiding intra if split CUs did not select it
    • 2 days ago Ashok Kumar Mishra analysis: skip intra in RD 0/4 if split was analyzed and no split CUs used intra
    • 2 days ago Ashok Kumar Mishra stats: profile effectiveness of reference limit masks
    • 2 days ago Ashok Kumar Mishra analysis: at RD 0/4 avoid motion references if not used by split blocks
    • 2 days ago Ashok Kumar Mishra analysis: re-order RD 0/4 analysis to do splits before ME or intra
    • 7 days ago Divya Manivannan asm: filter_vpp, filter_vps for 64xN in avx2
    • 22 hours ago Rajesh Paulraj asm: avx2 10bit code for luma_hpp[4xN]
    • 23 hours ago Dnyaneshwar G asm: removed duplicate constants in intrapred8.asm 8bpp, these constants are already defined into const-a.asm
    • 23 hours ago Dnyaneshwar G asm: removed some duplicate constants in intrapred16.asm 16bpp
    • 24 hours ago Deepthi Nandakumar param: tune grain disables rdoq-level. This provides better visual quality results
    • 25 hours ago Sumalatha Polureddy asm: avx2 code for sad_x4[48x64] (33937 -> 15279) for 10 bpp
    • 30 hours ago Steve Borho doc: fix formatting of code sample
Version 1.7.16
  • Release Date: May 20, 2015
    • 19 hours ago Gopu Govindaswamy search: add lowres MV into search MV candidate list for search ME
    • 18 hours ago Dnyaneshwar G asm: removed duplicate and redundant constants
    • 22 hours ago Sumalatha Polureddy asm: avx2 code for sad_x4[64xN] for 10 bpp
    • 22 hours ago Rajesh Paulraj asm: avx2 10bit code for luma_hpp[48x64] (82440.47 -> 44731.61)
    • 22 hours ago Rajesh Paulraj asm: avx2 10bit code for luma_hpp[24x32] (18855.08 -> 10742.66)
    • 23 hours ago Rajesh Paulraj asm: avx2 10bit code for luma_hpp[12x16] (5154.47 -> 3632.88)
    • 23 hours ago Rajesh Paulraj asm: avx2 10bit code for luma_hpp[32xN],[64xN]
    • 24 hours ago Rajesh Paulraj asm: avx2 10bit code for luma_hpp[16xN]
    • 24 hours ago Rajesh Paulraj asm: avx2 10bit code for luma_hpp[8xN]
    • 23 hours ago Sumalatha Polureddy asm: avx2 code for sad_x4[32xN] for 10 bpp
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for sad_x4[16xN] for 10 bpp
    • 42 hours ago Dnyaneshwar G asm: avx2 code for psyCost_pp 8x8, 16x16, 32x32 & 64x64, improved over 40% than previous asm
    • 26 hours ago Sumalatha Polureddy asm: avx2 code for sad_x3[48x64] for 10 bpp (25339 -> 11730)
Version 1.7.2
  • Release Date: May 20, 2015
    • 7 hours ago Steve Borho Merge with stable
    • 4 days ago Steve Borho api: introduce a less version strict API query
    • 7 hours ago Steve Borho Added tag 1.7 for changeset 8425278def1estable
    • 12 hours ago Steve Borho api: fix x265.h documentation for x265_max_bit_depthstable 1.7
    • 3 days ago Min Chen improve by replace condition operator to mask based
    • 3 days ago Min Chen reuse coeffFlag to reduce memory operator on coeff[] memory
    • 3 days ago Min Chen modify logic to remove lastCG in Quant::signBitHidingHDQ()
    • 3 days ago Min Chen faster algorithm to find firstNZPosInCG & lastNZPosInCG in Quant::signBitHidingHDQ()
    • 16 hours ago Min Chen improve Quant::signBitHidingHDQ by scanPosLast and findPosFirstLast
    • 16 hours ago Min Chen asm: modify API on findPosFirstLast to support all zeros block
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for sad_x3[64xN] for 10 bpp
    • 24 hours ago Sumalatha Polureddy asm: avx2code fore sad_x3[32xN] for 10bpp
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for sad_x3[16xN] for 10 bpp
    • 28 hours ago David T Yuen Call macros to reduce code size of primitive setup
    • 28 hours ago David T Yuen asm: interp_4tap_vert_ps_64xN and interp_4tap_vert_ps_48x64 sse2
    • 28 hours ago David T Yuen asm: interp_4tap_vert_ps_32xN sse2
    • 29 hours ago David T Yuen asm: interp_4tap_vert_ps_24xN sse2
    • 29 hours ago David T Yuen asm: interp_4tap_vert_ps_16xN sse2
    • 29 hours ago David T Yuen asm: interp_4tap_vert_ps_12xN sse2
    • 29 hours ago David T Yuen asm: interp_4tap_vert_ps_8xN sse2
    • 29 hours ago David T Yuen asm: interp_4tap_vert_ps_8xN sse2
    • 29 hours ago David T Yuen asm: interp_4tap_vert_ps_6xN sse2
    • 29 hours ago David T Yuen asm: interp_4tap_vert_ps_4xN sse2
    • 30 hours ago David T Yuen asm: interp_4tap_vert_ps_4x2 sse2
    • 30 hours ago David T Yuen asm: interp_4tap_vert_ps_2xN sse2
    • 4 days ago Gopu Govindaswamy lowres: cache the lowres maxNumBlocks and reuse this in other places
    • 3 days ago Steve Borho api: do not log warnings from x265_api_get() on typical failuresstable
    • 4 days ago Dnyaneshwar G asm: addAvg high_bit_depth avx2 asm for chroma sizes width >= 8, reused code from luma
    • 5 days ago Sumalatha Polureddy asm: avx2 code for sad[64x64] (13997 -> 5214) for 10 bpp
    • 5 days ago Divya Manivannan asm: filter_vpp, filter_vps for 48x64 in avx2
    • 5 days ago Dnyaneshwar G asm: addAvg avx2 code for high_bit_depth sizes >= 8, improved over +AH4-45% than previous code
    • 5 days ago Sumalatha Polureddy asm: avx2 code for sad[64x48] (10791 -> 4053) for 10 bpp
    • 5 days ago Divya Manivannan asm: filter_vpp, filter_vps for 24x64 in avx2
    • 5 days ago Sumalatha Polureddy asm: avx2 code for sad[64x32] (7156 -> 2625) for 10 bpp
    • 5 days ago Divya Manivannan asm: filter_vpp, filter_vps for 4x32 in avx2
    • 5 days ago Sumalatha Polureddy asm: avx2 code for sad[64x16] (3620 -> 1279) for 10 bpp
Version 1.6.451
  • Release Date: May 15, 2015
    • 7 hours ago Min Chen asm: AVX2 version sa8d[16x16], 1913c(AVX) -> 1620c(AVX2)
    • 7 hours ago Min Chen asm: enable sa8d_8x8_avx2, 375c -> 322c
    • 7 hours ago Min Chen two inputs version of signOf
    • 7 hours ago Min Chen sao: improve calcSaoStatsCu, use local buffer to reduce array index operators
    • 18 hours ago Deepthi Nandakumar Merge with stable
    • 19 hours ago Deepthi Nandakumar pixel-util: correct typo, which prevented Windows header files from being includedstable
    • 23 hours ago Deepthi Nandakumar cmake: spacing nitsstable

Version 1.6.442
  • Release Date: May 15, 2015
    • 7 hours ago Min Chen asm: AVX2 version sa8d[16x16], 1913c(AVX) -> 1620c(AVX2)
    • 7 hours ago Min Chen asm: enable sa8d_8x8_avx2, 375c -> 322c
    • 7 hours ago Min Chen two inputs version of signOf
    • 7 hours ago Min Chen sao: improve calcSaoStatsCu, use local buffer to reduce array index operators
    • 18 hours ago Deepthi Nandakumar Merge with stable
    • 19 hours ago Deepthi Nandakumar pixel-util: correct typo, which prevented Windows header files from being includedstable
    • 23 hours ago Deepthi Nandakumar cmake: spacing nitsstable

Version 1.6.435
  • Release Date: May 13, 2015
    • 8 hours ago David T Yuen asm: interp_4tap_vert_pp sse2
    • 8 hours ago David T Yuen asm: interp_4tap_vert_pp_32xN sse2
    • 15 hours ago Steve Borho entropy: fix build warning in checked GCC builds
    • 16 hours ago Steve Borho Merge with stable
    • 16 hours ago Steve Borho cli: we can no longer use compiled bit-depth for ditherstable
    • 29 hours ago David T Yuen asm: interp_4tap_vert_pp sse2
    • 29 hours ago David T Yuen asm: interp_4tap_vert_pp sse2
    • 30 hours ago David T Yuen asm: interp_4tap_vert_pp sse2
    • 31 hours ago Min Chen reduce shift operators in cost on coeffRemain
    • 31 hours ago Min Chen reduce conditional operators on c1Flag loop in codeCoeffNxN()
    • 31 hours ago Min Chen convert for_loop to do_while in codeCoeffNxN(), because all of count are positive
    • 31 hours ago Min Chen convert for_loop to do_while in codeCoeffNxN(), because all of scanPosSigOff positive now
    • 31 hours ago Min Chen improve codeCoeffNxN by bypass check on one coeff block
    • 24 hours ago Dnyaneshwar G asm: enable psyCost_ss avx2 for HIGH_BIT_DEPTH
    • 43 hours ago Dnyaneshwar G asm: avx2 code for high_bit_depth psyCost_pp_4x4, reduced 400c->250c
    • 25 hours ago Dnyaneshwar G asm: avx2 code for high_bit_depth intra_pred_planar32x32
    • 2 days ago Dnyaneshwar G asm: avx2 code for high_bit_depth intra_planar_16x16
    • 25 hours ago Dnyaneshwar G asm: avx2 code for high_bit_depth intra_dc_16x16
    • 25 hours ago Dnyaneshwar G asm: intra_pred_dc32 high_bit_depth code
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for sad[32x64] for 10 bpp(6222 -> 2427)
    • 26 hours ago Divya Manivannan asm: filter_vpp, filter_vps for 2x16 in avx2
    • 31 hours ago Steve Borho Merge with stable
    • 31 hours ago Steve Borho api: improve error handling in x265_api_get() bit depth checksstable
Version 1.6.417
  • Release Date: May 12, 2015
    • 12 hours ago David T Yuen asm: interp_4tap_horiz_pp sse3
    • 15 hours ago Steve Borho Merge with stable
    • 17 hours ago Steve Borho cli: allow --output-depth to select output|internal bit depthstable
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for sad[32x16],[32x24],[32x32] for 10 bpp
    • 25 hours ago Sumalatha Polureddy asm: avx2 code for sad[32x8] for 10 bpp(833 -> 304)
    • 25 hours ago Sumalatha Polureddy asm: avx2 code for sad[16x64] for 10 bpp (3060->1353)
    • 17 hours ago Steve Borho slicetype: fix build with detailed stats enabledstable
Version 1.6.411
  • Release Date: May 10, 2015
    • 14 hours ago Steve Borho Merge with default (prep for 1.7)
    • 3 days ago Jie Zhang param: space nitsdefault
    • 3 days ago Jie Zhang param: added qcomp into x265 info on ABR or CRF ratecontrol mode
    • 31 hours ago Min Chen use new combo nextState and bitsCost table to reduce memory and address operators in codeCoeffNxN()
    • 29 hours ago David T Yuen asm: interp_4tap_vert_pp sse2
    • 34 hours ago David T Yuen asm: interp_4tap_vert_pp sse2
    • 35 hours ago David T Yuen asm: interp_4tap_vert_pp sse2
    • 35 hours ago David T Yuen asm: interp_4tap_vert_pp sse2
    • 35 hours ago David T Yuen asm: interp_4tap_vert_pp sse2
    • 35 hours ago David T Yuen asm: interp_4tap_vert_pp sse2
    • 34 hours ago Tom Vaughan cli.rst - qgsize formatting fixed, fixed explanation of --strict-cbr
    • 2 days ago Dnyaneshwar G asm: avx2 code for high_bit_depth satd_48x64, improved over +AH4-50% than previous asm
    • 2 days ago Dnyaneshwar G asm: avx2 code for high_bit_depth satd_64xN, improved over +AH4-50% than previous asm
    • 2 days ago Dnyaneshwar G asm: avx2 code for high_bit_depth satd_32xN, improved over +AH4-50% than previous asm
    • 2 days ago Dnyaneshwar G asm: avx2 code for high_bit_depth satd_16xN, improved over +AH4-50% than previous asm
    • 2 days ago Dnyaneshwar G asm: avx2 code for high_bit_depth satd_16x8
Version 1.6.402
  • Release Date: May 9, 2015
    • 33 hours ago Min Chen inline mvcost() to reduce address operators
    • 13 hours ago Steve Borho api: new API functions need to be exported from Win DLL and added to x265_api
    • 5 weeks ago Aarthi Thirumalai param: add function to print reconfigured param options
    • 5 weeks ago Aarthi Thirumalai api: introduce api to reconfigure encoder param
    • 6 weeks ago Aarthi Thirumalai encoder: introduce an instance to hold the latest param setting in the encoder
    • 6 weeks ago Aarthi Thirumalai frameEncoder: copy param state for each FE and worker thread from input frame
    • 6 weeks ago Aarthi Thirumalai frame: add x265_param* to each frame
    • 36 hours ago Steve Borho api: make all input string params use const char* type
    • 43 hours ago Aasaipriya Chandran chroma_hpp[6x16] for i422 - improved 1035c->868c
    • 43 hours ago Aasaipriya Chandran asm: luma_hps[4x4, 4x8, 4x16] for 16bpp - improved 1172c->993c, 1573c->1232c, 2412c->1912c
    • 26 hours ago Divya Manivannan asm: filter_vpp, filter_vps for 16x24 in avx2
Version 1.6.396
  • Release Date: May 7, 2015
    • 33 hours ago Min Chen inline mvcost() to reduce address operators
    • 13 hours ago Steve Borho api: new API functions need to be exported from Win DLL and added to x265_api
    • 5 weeks ago Aarthi Thirumalai param: add function to print reconfigured param options
    • 5 weeks ago Aarthi Thirumalai api: introduce api to reconfigure encoder param
    • 6 weeks ago Aarthi Thirumalai encoder: introduce an instance to hold the latest param setting in the encoder
    • 6 weeks ago Aarthi Thirumalai frameEncoder: copy param state for each FE and worker thread from input frame
    • 6 weeks ago Aarthi Thirumalai frame: add x265_param* to each frame
    • 36 hours ago Steve Borho api: make all input string params use const char* type
    • 43 hours ago Aasaipriya Chandran chroma_hpp[6x16] for i422 - improved 1035c->868c
    • 43 hours ago Aasaipriya Chandran asm: luma_hps[4x4, 4x8, 4x16] for 16bpp - improved 1172c->993c, 1573c->1232c, 2412c->1912c
    • 26 hours ago Divya Manivannan asm: filter_vpp, filter_vps for 16x24 in avx2
Version 1.6.385
  • Release Date: May 6, 2015
    • 13 hours ago Steve Borho param: x265_free has no return code
    • 15 hours ago Steve Borho rdcost: consistent clipping for chroma QP, combine conditionals
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for sad 16x4,16x8,16x12,16x16,16x32 for 10bpp
    • 30 hours ago Min Chen fix warning C4701: potentially uninitialized local variable 'scanPos' used
Version 1.6.381
  • Release Date: May 5, 2015
    • 10 hours ago Steve Borho cli: improve help for --crqpoffs and --cbqpoffs
    • 11 hours ago Steve Borho rdcost: fix cut-paste bugs in chroma QP offset R-D calculations
    • 13 hours ago Steve Borho cli: nit
    • 13 hours ago Steve Borho cli: header nit
    • 2 days ago Steve Borho cudata: add comment for clipMv
    • 2 days ago Steve Borho search: move up selectMVP() in file, no behavior change
    • 2 days ago Steve Borho api: fixup master-display docs
    • 2 days ago Steve Borho sei: add content light level info SEI
    • 2 days ago Steve Borho doc: give a more useful master-display example
    • 21 hours ago Sumalatha Polureddy asm: use x264 code for sse2 sad[16x64]
    • 22 hours ago Sumalatha Polureddy asm: use x264 code for sse2 sad[16xN] except for 16x64
    • 26 hours ago Aasaipriya Chandran asm: chroma_hps[2x16] for i422 - improved 796c->559c
    • 26 hours ago Aasaipriya Chandran asm: chroma_hps[24x64] for i422 - improved 8428c->6562c
    • 26 hours ago Aasaipriya Chandran asm: chroma_hps[48x64] for i444 - improved 16378c->11853c
    • 4 days ago Deepthi Nandakumar api: add error message and checks in api_get()
Version 1.6.366
  • Release Date: May 4, 2015
    • 15 hours ago Steve Borho sei: remaining bits of master-display SEI changes
Version 1.6.365
  • Release Date: May 3, 2015
    • 2 days ago David T Yuen asm: interp_8tap_hv_pp_8x8 sse3
    • 42 hours ago David T Yuen asm: interp_8tap_horiz pp and ps sse2
    • 11 hours ago Steve Borho sei: the mastering display color primaries will typically be RGB
    • 14 hours ago Steve Borho search: use selectMVP() helper function for --pme
    • 11 hours ago Steve Borho search: fix non-det when an AMVP candidate is discarded by frame parallelism

Version 1.6.360
  • Release Date: May 1, 2015
    • 6 days ago Steve Borho search: allow AMP to use motion estimation for 64x64 CUs
    • 35 hours ago Steve Borho search: do not clip MVP in setSearchRange()
    • 35 hours ago Steve Borho search: introduce selectMVP helper method
    • 37 hours ago Steve Borho search: cleanup checkBestMVP(), no behavior change
    • 18 hours ago Divya Manivannan asm: filter_vpp, filter_vps for 2x4 in avx2
    • 20 hours ago Divya Manivannan asm: filter_vpp, filter_vps for 8x12 in avx2
    • 21 hours ago Divya Manivannan asm: filter_vpp, filter_vps for 12x32 in avx2
    • 18 hours ago Min Chen asm: downgrade x265_interp_8tap_hv_pp_8x8 from SSE4 to SSSE3
    • 18 hours ago Min Chen fix check failure in Entropy::writeCoefRemainExGolomb()
    • 20 hours ago Min Chen simplify logic on get coeff remain cost in codeCoeffNxN()
    • 20 hours ago Min Chen faster clip operator on goRiceParam
    • 20 hours ago Min Chen simplify compute on get codeNumber length
    • 20 hours ago Min Chen improve compute on baseLevel by 2-bits encode code
    • 20 hours ago Min Chen fast RD path on encode coeff remain code in codeCoeffNxN()
    • 20 hours ago Min Chen remove reduce check on firstC2FlagIdx
    • 20 hours ago Min Chen pre-compute abs coeff and simplify scan table
    • 20 hours ago Min Chen convert sigCtx table from [4][4] to [16]
    • 43 hours ago Aasaipriya Chandran asm: chroma_hpp[48x64] for i444 - improved 17498c->13381c
    • 22 hours ago Deepthi Nandakumar doc: replace sublayer with enhancement layer
    • 26 hours ago Deepthi Nandakumar api: clarify docs and use of x265_api_get()
Version 1.6.340
  • Release Date: Apr 30, 2015
    • 14 hours ago Steve Borho api: document x265_api_get()'s ability to forward API requests
    • 20 hours ago Divya Manivannan asm: filter_vpp, filter_vps for 32x64, 32x48 in avx2
    • 21 hours ago Divya Manivannan asm: filter_vpp, filter_vps for 8x64 in avx2
    • 22 hours ago Divya Manivannan asm: filter_vpp, filter_vps for 16x64 in avx2
    • 24 hours ago Divya Manivannan asm: filter_vsp, filter_vss for 64xN, 48x64 in avx2
    • 26 hours ago Divya Manivannan asm: filter_vsp, filter_vss for 2x4 in avx2
    • 35 hours ago Steve Borho cli: add -P short option for --profile
    • 34 hours ago Steve Borho api: allow libx265 to forward x265_api_get() callsa
Version 1.6.332
  • Release Date: Apr 29, 2015
    • 10 hours ago Steve Borho frameencoder: cleanups
    • 17 hours ago Min Chen simplify logic on posOffset in codeCoeffNxN()
    • 17 hours ago Min Chen asm: remove interp4_hps_shuf, it is duplicate to interp4_hpp_shuf
    • 17 hours ago Min Chen asm: use prefix const to avoid unaligned crash
    • 19 hours ago Min Chen asm: remove tab_c_526336, it is duplicate to pd_526336
    • 18 hours ago Sumalatha Polureddy asm: avx2 code for sad[16x8] for 10 bpp (398 -> 254)
    • 19 hours ago Divya Manivannan asm: filter_vsp, filter_vss for 4x32 in avx2
    • 19 hours ago Divya Manivannan asm: filter_vsp, filter_vss for 12x32 in avx2
    • 20 hours ago Divya Manivannan asm: filter_vsp, filter_vss for 16x24 in avx2
    • 21 hours ago Divya Manivannan asm: filter_vsp, filter_vss for 2x16 in avx2
    • 16 hours ago Aarthi Thirumalai log: make qTreeCnt as stack arrays to avoid non determinism in 2 pass
Version 1.6.317
  • Release Date: Apr 28, 2015
    • 12 hours ago Steve Borho cudata: cu index is no longer necessary again
    • 12 hours ago Steve Borho analysis: remove m_aqQP[], determine AQ QPs on demand
    • 53 minutes ago Steve Borho frameencoder: cleanups, no behavior change
    • 58 minutes ago Steve Borho frameencoder: recover vbvCost and intraVbvCost logic
    • 12 hours ago Steve Borho rc: extract final average QP from the coded CTU structure
    • 13 hours ago Steve Borho level: do not try to configure color space in x265_param_apply_profile()
    • 15 hours ago Steve Borho Merge with stable
    • 19 hours ago Min Chen rdoQuant: reduce address operators by swap order on array significantBits[][]
    • 19 hours ago Min Chen force type convert since multiplication result up to 34-bits
    • 19 hours ago Min Chen modify lambda from int64 to int32 because dynamic range less than 21 bits
    • 19 hours ago Min Chen modify m_psyRdoqScale from int64 to int32 because dynamic range is [0,50]*256
    • 18 hours ago Divya Manivannan sao: add comment for the reason of two versions of sao primitives
    • 19 hours ago Divya Manivannan asm: filter_vsp[6x16], filter_vss[6x16] in avx2
    • 21 hours ago Divya Manivannan asm: filter_vsp[8x12], filter_vss[8x12] for I422 in avx2
    • 23 hours ago Divya Manivannan asm: filter_vsp and filter_vss for Nx64, 32x48 in I422
    • 23 hours ago Sumalatha Polureddy asm: avx2 code for chroma vpp filter for i444
    • 25 hours ago Sumalatha Polureddy asm: avx2 code for chroma vpp filter for i422
    • 6 days ago Aarthi Thirumalai rc: fix cost issues in predicting row size during mid frame vbv encodes.stable
    • 3 days ago Steve Borho analysis: always configure quant QP directly after setting RD lambda
    • 3 days ago Steve Borho analysis: configure slave quant QP prior to pmode intra RDO for RD 0..4
    • 3 days ago Steve Borho analysis: hoist all adaptive-quant work to recursive callers
    • 3 days ago Steve Borho analysis: keep per-CU AQ QPs in cuGeom index order, simplify arguments
    • 3 days ago Steve Borho analysis: simplify CTU QP init loops
    • 3 days ago Steve Borho analysis: rename m_qp to m_aqQP for clarity
Version 1.6.298
  • Release Date: Apr 25, 2015
    • 9 hours ago Steve Borho search: rename setQP() method to setLambdaFromQP
    • 9 hours ago Steve Borho analysis: nits
    • 17 hours ago Rajesh Paulraj asm: avx2 10bit code for add_ps for chroma sizes 16xN, 32xN, reuse luma code
    • 17 hours ago Rajesh Paulraj asm: avx2 10bit code for add_ps[16x16],[32x32],[64x64]
    • 22 hours ago Divya Manivannan asm: add pixel restoration part in saoCuOrgE2 primitive
    • 22 hours ago Divya Manivannan asm: avx2 code for sign primitive: improve 204c->114c
    • 19 hours ago Sumalatha Polureddy asm: avx2 code for chroma vps filter for i444
    • 20 hours ago Sumalatha Polureddy asm: avx2 code for chroma vps filter for i422
    • 22 hours ago Sumalatha Polureddy asm: avx2 code for chroma vsp filter for i444
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for chroma vsp filter for i422
    • 17 hours ago Min Chen rdoQuant: optimize getSigCtxInc()
    • 17 hours ago Min Chen rdoQuant: split coeff cost into psy and non-psy path
    • 17 hours ago Min Chen fix build fault on WinXP
Version 1.6.285
  • Release Date: Apr 25, 2015
    • 12 hours ago Steve Borho smoke-test: set a higher VBV tolerance for the smoke test
    • 16 hours ago Rajesh Paulraj asm: avx2 10bit code for sub_ps for chroma sizes 16xN, 32xN, reuse luma code
    • 16 hours ago Rajesh Paulraj asm: avx2 10bit code for sub_ps[16x16],[32x32],[64x64]
    • 16 hours ago Divya Manivannan asm: saoCuOrgE2[1] avx2 code: improve 449c->292c
    • 17 hours ago Divya Manivannan asm: saoCuOrgE2[0] avx2 code: improve 154c->128c
    • 17 hours ago Divya Manivannan sao: modify saoCuOrgE2 primitive to handle width=16 separately
    • 20 hours ago Sumalatha Polureddy asm: avx2 code for chroma vss filter for i444
    • 23 hours ago Sumalatha Polureddy asm: avx2 code chroma vss filter for i422

Version 1.6.277
  • Release Date: Apr 24, 2015
    • 3 hours ago Steve Borho cli: document --qpstep
    • 9 hours ago David T Yuen Invoke macro to setup sse2 intrapred dc and planar primitives
    • 10 hours ago Steve Borho Merge with stable
    • 10 hours ago Steve Borho doc: correct --qpstepstable
    • 14 hours ago Steve Borho threadpool: use Win7 version macro directly, for more clarity
    • 17 hours ago Min Chen testbench: fix table fault when trSize more than 8
    • 17 hours ago Min Chen asm: avx2+-bmi2 version of scanPosLast, 27.6k -> 6.8k cycles
    • 17 hours ago Min Chen testbench: fix testbench crash when no coeff in block
    • 17 hours ago Min Chen testbench: support BMI2
    • 17 hours ago Min Chen asm: rename findPosLast to scanPosLast and modify its API
    • 17 hours ago Min Chen asm: fix bug in generic version findPosLast_x64 and improve testbench on it
    • 16 hours ago Rajesh Paulraj asm: avx2 10bit code for scale2D_64to32
    • 24 hours ago Dnyaneshwar G asm: avx2 code for satd_48x64 and 64xN, improved over +AH4-50% than SSE
    • 24 hours ago Dnyaneshwar G asm: avx2 code for satd_16xN, improved over +AH4-50% than SSE code
    • 23 hours ago Divya Manivannan asm: saoCuOrgE3 avx2 code for width>16: improve 508c->427c
    • 24 hours ago Divya Manivannan sao: remove saoCuOrgE3_2Rows function and modify saoCuOrgE3 primitive to handle width=16 seperately
    • 26 hours ago David T Yuen asm: interp_4tap_horiz_pp sse3
Version 1.6.260
  • Release Date: Apr 22, 2015
    • 86 minutes ago Steve Borho Merge with stable
    • 96 minutes ago Steve Borho doc: correct frame thread docsstable
    • 11 hours ago Steve Borho threadpool: fix 'enumeral and non-enumeral type in conditional ...' warnings
    • 11 hours ago Steve Borho asm: disable new findPosLast - it is causing check failures / crashes
    • 20 hours ago Gopu Govindaswamy slicetype: select best mvp using neighbor mvs satd cost for Lowres ME
    • 21 hours ago Dnyaneshwar G asm: new optimized algorithm for satd, improved +AH4-25% over previous algorithm
    • 21 hours ago Sumalatha Polureddy asm: avx code for chroma satd functions for all partitions of 422
    • 23 hours ago Min Chen asm: generic x64 version of findPosLast
    • 34 hours ago Steve Borho doc: if you break cpu auto-detection, you get to keep both halves
    • 42 hours ago Min Chen asm: ssse3 version of findPosFirstLast, 365c -> 75c
    • 2 days ago Min Chen rdoQuant: move cgRdStats.sigCost0 outside from loop
    • 4 days ago Min Chen rdoQuant: fast zero-coeff path
    • 4 days ago Min Chen rdoQuant: improve coeff group block clean code
    • 4 days ago Min Chen rdoQuant: fast zero cost compute path
    • 2 days ago Min Chen rdoQuant: improve encoder +AH4-3.5% by modify all zero-coeff group scan and cost compute logic
    • 41 hours ago Divya Manivannan sao: modify saoCuOrgE3_2Rows C code and add sse4 code
    • 13 days ago Steve Borho tests: add coverage for --qg-size
    • 12 days ago Steve Borho param: show quant-group size in logs, move AQ config into its own line
    • 12 days ago Steve Borho encoder: give param->rc.qgSize a sane default when dqp is not used
    • 12 days ago Steve Borho encoder: ignore param->rc.qgSize when delta-qp coding is disabled
    • 2 weeks ago Deepthi Nandakumar search: add RDcost measurement of DeltaQP to lower rdLevels
    • 4 days ago Deepthi Nandakumar entropy: after encodeCU, the CU structures need to be reset with the right QP
    • 4 days ago Deepthi Nandakumar AQ: Re-enable fine grained adaptive quantization
Version 1.6.239
  • Release Date: Apr 19, 2015
    • 14 hours ago David T Yuen asm: leading space nit
    • 11 hours ago Steve Borho rc: test case nit
Version 1.6.237
  • Release Date: Apr 19, 2015
    • 11 hours ago Xinyue Lu cli: fix incorrect timebase source
    • 15 hours ago David T Yuen asm: interp_4tap_horiz_pp_4x32_sse3
    • 15 hours ago David T Yuen asm: interp_4tap_horiz_pp_4x16_sse3
    • 15 hours ago David T Yuen asm: interp_4tap_horiz_pp_4x8_sse3
    • 15 hours ago David T Yuen asm: interp_4tap_horiz_pp_4x4_sse3
    • 15 hours ago David T Yuen asm: interp_4tap_horiz_pp_4x2_sse3
    • 15 hours ago David T Yuen asm: interp_4tap_horiz_pp_2x16_sse3
    • 15 hours ago David T Yuen asm: interp_4tap_horiz_pp_2x8_sse3
    • 16 hours ago David T Yuen asm: interp_4tap_horiz_pp_2x4_sse3
    • 21 hours ago Praveen Tiwari asm: intra_allangs4x4 improved by +AH4-61% over SSE4
    • 20 hours ago Aasaipriya Chandran asm: chroma_hps[4xN, 8xN, 16xN, 32xN, 24x32] for i444
    • 20 hours ago Aasaipriya Chandran asm: chroma_hps[4xN, 8xN, 16xN, 32xN, 2x8]
    • 21 hours ago Aasaipriya Chandran asm: chroma_hpp[4xN, 8xN, 16xN, 32xN, 12x16, 24x32] for i444
    • 21 hours ago Rajesh Paulraj asm: avx2 10bit code for chroma_p2s[16xN],[24xN],[32xN], reuse luma code
    • 23 hours ago Sumalatha Polureddy asm: avx2 code for chroma addAvg for all partitions
    • 26 hours ago Rajesh Paulraj asm: avx2 10bit code for convert_p2s[24xN],[48x64]
    • 27 hours ago Aasaipriya Chandran asm: chroma_hpp[64x64, 64x48, 64x32, 64x16] for i444 - improved 22990c->14176c, 17897c->10791c, 12050c->7186c, 5655c->3266c
    • 31 hours ago Steve Borho rc: unix eoln for rate-control-tests.txt
    • 45 hours ago Mahesh Pittala regression: typo in rc tests
    • 47 hours ago Rajesh Paulraj asm: avx2 10bit code for convert_p2s[32xN],[64xN]
    • 43 hours ago Rajesh Paulraj asm: avx2 10bit code for convert_p2s[16xN]
    • 46 hours ago Aasaipriya Chandran asm: chroma_hpp i422[4xN, 8xN, 16xN, 32xN]
    • 47 hours ago Aasaipriya Chandran asm: chroma_hps[64x64, 64x48, 64x32, 64x16] for i444 - improved 21540c->14767c, 18551c->14129c,17096c->12742c, 6216c->3923c
    • 13 hours ago Steve Borho Backed out changeset: 7be1172ec816
    • 2 days ago Dnyaneshwar G asm: new optimized algorithm for satd, improved +AH4-30% over previous algorithm
    • 2 days ago Rajesh Paulraj asm: sse version 10bit code for chroma_p2s, reuse luma code
    • 2 days ago Rajesh Paulraj asm: sse4 10bit code for chroma_p2s[2xN] for i420, i422
    • 2 days ago Rajesh Paulraj asm: sse4 10bit code for chroma_p2s[6xN] for i420, i422
    • 2 days ago Rajesh Paulraj asm: ssse3 10bit code for chroma_p2s[4x2],[8x2],[8x6]
    • 2 days ago Rajesh Paulraj asm: ssse3 10bit code for convert_p2s[12xN],[48x64]
    • 2 days ago Sumalatha Polureddy asm: avx2 code for chroma sub_ps module, reused luma code
    • 2 days ago Sumalatha Polureddy asm: add macro to sub_ps module to reduce code size

Version 1.6.213
  • Release Date: Apr 17, 2015
    • 20 hours ago Sumalatha Polureddy asm: avx2 code for chroma add_ps, reused luma code
    • 22 hours ago Divya Manivannan asm: saoCuOrgE3 avx2 code: 502c->473c
    • 22 hours ago Aasaipriya Chandran asm: chroma_hpp[2x16] for i422 - improved 595c->500c
    • 22 hours ago Aasaipriya Chandran asm: chroma_hpp[24x64] for i422 - improved 9272c->8212c
    • 22 hours ago Aasaipriya Chandran asm: chroma_hpp[12x32] for i422 - improved 2997c->2295c
    • 81 minutes ago mahesh pittala regression: typo in rc tests
    • 42 minutes ago Min Chen asm: improve algorithm logic on saoCuOrgE3
    • 17 hours ago Min Chen improve rdoQuant() by block fill on non-zero coeff group
    • 17 hours ago Min Chen improve rdoQuant() by use non-zero coeff group mask to reduce count of coeff scan
    • 17 hours ago Min Chen improve rdoQuant() by reduce count of code group scan
    • 17 hours ago Rajesh Paulraj asm: ssse3 10bit code for convert_p2s[24xN]
    • 17 hours ago Rajesh Paulraj asm: ssse3 10bit code for convert_p2s[32xN],[64xN]
    • 17 hours ago Rajesh Paulraj asm: ssse3 10bit code for convert_p2s[16xN]
    • 17 hours ago Rajesh Paulraj asm: ssse3 10bit code for convert_p2s[8xN]
    • 18 hours ago Rajesh Paulraj asm: ssse3 10bit code for convert_p2s[4xN]
    • 20 hours ago Sumalatha Polureddy asm: avx code for chroma copy_ss 32x64, reused luma code (2616 -> 1313)
    • 22 hours ago Dnyaneshwar G asm: avx2 code for satd_32xN
    • 26 hours ago Divya Manivannan sao: add saoCuOrgE3_2Rows function to process 2 rows
    • 24 hours ago Praveen Tiwari asm: intra_pred_ang32_18 improved by +AH4-45% over SSE4
    • 43 hours ago Sumalatha Polureddy asm: improve sub_ps[16x16] (477 -> 461) and reduce code size
    • 25 hours ago Min Chen disable SIGPIPE on Windows platform
    • 33 hours ago David T Yuen asm: intra pred all_angs_pred_4x4 sse2
    • 44 hours ago Min Chen simplify rdoQuant() logic on ctxSet
    • 6 days ago Sumalatha Polureddy asm: improve avx2 code sub_ps[32x32] 1402 -> 1360
    • 36 hours ago Steve Borho doc: clarify that --pools strings might need shell escaping (closes #121)
    • 36 hours ago Steve Borho api: add SMPTE ST 2086 mastering display color metadata
    • 5 days ago Peixuan Zhang cli: add an output preview feature, activated by --recon-y4m-exec
    • 2 days ago Steve Borho input: use poke() method of ThreadSafeInteger appropriately
    • 2 days ago Steve Borho cmake: use explicit cpp file lists for input/ and output/
    • 2 days ago Steve Borho cmake: specify default Windows target O/S as Win7, to enable NUMA APIs
    • 5 days ago Aarthi Thirumalai rc: add helper function to decide the VBV predictor type for each frame

Version 1.6.205
  • Release Date: Apr 16, 2015
    • 20 hours ago Sumalatha Polureddy asm: avx2 code for chroma add_ps, reused luma code
    • 22 hours ago Divya Manivannan asm: saoCuOrgE3 avx2 code: 502c->473c
    • 22 hours ago Aasaipriya Chandran asm: chroma_hpp[2x16] for i422 - improved 595c->500c
    • 22 hours ago Aasaipriya Chandran asm: chroma_hpp[24x64] for i422 - improved 9272c->8212c
    • 22 hours ago Aasaipriya Chandran asm: chroma_hpp[12x32] for i422 - improved 2997c->2295c
    • 81 minutes ago mahesh pittala regression: typo in rc tests
    • 42 minutes ago Min Chen asm: improve algorithm logic on saoCuOrgE3
    • 17 hours ago Min Chen improve rdoQuant() by block fill on non-zero coeff group
    • 17 hours ago Min Chen improve rdoQuant() by use non-zero coeff group mask to reduce count of coeff scan
    • 17 hours ago Min Chen improve rdoQuant() by reduce count of code group scan
    • 17 hours ago Rajesh Paulraj asm: ssse3 10bit code for convert_p2s[24xN]
    • 17 hours ago Rajesh Paulraj asm: ssse3 10bit code for convert_p2s[32xN],[64xN]
    • 17 hours ago Rajesh Paulraj asm: ssse3 10bit code for convert_p2s[16xN]
    • 17 hours ago Rajesh Paulraj asm: ssse3 10bit code for convert_p2s[8xN]
    • 18 hours ago Rajesh Paulraj asm: ssse3 10bit code for convert_p2s[4xN]
    • 20 hours ago Sumalatha Polureddy asm: avx code for chroma copy_ss 32x64, reused luma code (2616 -> 1313)
    • 22 hours ago Dnyaneshwar G asm: avx2 code for satd_32xN
    • 26 hours ago Divya Manivannan sao: add saoCuOrgE3_2Rows function to process 2 rows
    • 24 hours ago Praveen Tiwari asm: intra_pred_ang32_18 improved by +AH4-45% over SSE4
    • 43 hours ago Sumalatha Polureddy asm: improve sub_ps[16x16] (477 -> 461) and reduce code size
    • 25 hours ago Min Chen disable SIGPIPE on Windows platform
    • 33 hours ago David T Yuen asm: intra pred all_angs_pred_4x4 sse2
    • 44 hours ago Min Chen simplify rdoQuant() logic on ctxSet
    • 6 days ago Sumalatha Polureddy asm: improve avx2 code sub_ps[32x32] 1402 -> 1360
    • 36 hours ago Steve Borho doc: clarify that --pools strings might need shell escaping (closes #121)
    • 36 hours ago Steve Borho api: add SMPTE ST 2086 mastering display color metadata
    • 5 days ago Peixuan Zhang cli: add an output preview feature, activated by --recon-y4m-exec
    • 2 days ago Steve Borho input: use poke() method of ThreadSafeInteger appropriately
    • 2 days ago Steve Borho cmake: use explicit cpp file lists for input/ and output/
    • 2 days ago Steve Borho cmake: specify default Windows target O/S as Win7, to enable NUMA APIs
    • 5 days ago Aarthi Thirumalai rc: add helper function to decide the VBV predictor type for each frame

Version 1.6.174
  • Release Date: Apr 13, 2015
    • 32 hours ago Steve Borho cli: nits, remove dead code
    • 33 hours ago Steve Borho encoder: rename functions to avoid name conflicts with Thread::stop()
    • 36 hours ago Steve Borho nal: annexb warning fixes for MSVC
Version 1.6.172
  • Release Date: Apr 12, 2015
    • 17 hours ago Divya Manivannan asm: saoCuOrgB0 avx2 code: 23780c->18441cdefault tip
    • 22 hours ago Sumalatha Polureddy asm: avx code for chroma sa8d, reused luma code
    • 23 hours ago Praveen Tiwari asm: intra_pred_ang8_16 improved by +AH4-3% over SSE4
    • 24 hours ago Praveen Tiwari asm: intra_pred_ang8_20 improved by +AH4-4% over SSE4
    • 18 hours ago Min Chen avoid calculate rateIncUp and rateIncDown when sigHide disabled
    • 18 hours ago Min Chen change costUncoded[] coordinate system from Raster to Zigzag
    • 25 hours ago Rajesh Paulraj asm: avx2 8bpp code for chroma_p2s[32xN],[24xN], reuse the luma code
    • 25 hours ago Rajesh Paulraj asm: avx2 8bpp code for convert_p2s[24xN]
    • 25 hours ago Dnyaneshwar G asm: avx2 code for planecopy_sp
    • 29 hours ago Xinyue Lu cli: annex_b format switch
    • 15 hours ago Steve Borho asm: fix eoln in comment
Version 1.6.160
  • Release Date: Apr 10, 2015
    • 14 hours ago Steve Borho Merge with stable
    • 18 hours ago Praveen Tiwari asm: intra_pred_ang8_21 improved by +AH4-5% over SSE4
    • 18 hours ago Praveen Tiwari asm: intra_pred_ang8_22 improved by +AH4-14% over SSE4
    • 18 hours ago Praveen Tiwari asm: intra_pred_ang8_23 improved by +AH4-18% over SSE4
    • 19 hours ago Praveen Tiwari asm: intra_pred_ang8_15 improved by +AH4-5% over SSE4
    • 20 hours ago Praveen Tiwari asm: intra_pred_ang8_14 improved by +AH4-15% over SSE4
    • 20 hours ago Praveen Tiwari asm: intra_pred_ang8_13 improved by +AH4-16% over SSE4
    • 23 hours ago Praveen Tiwari asm: correct register count
    • 24 hours ago Praveen Tiwari asm: intra_pred_ang16_13 improved by +AH4-9% over SSE4
    • 25 hours ago Praveen Tiwari asm: intra_pred_ang16_12 improved by +AH4-20% over SSE4
    • 17 hours ago Rajesh Paulraj asm: sse4 8bpp code for chroma_p2s[2xN]
    • 18 hours ago Min Chen optimize c1c2 context set update logic in rdoQuant
    • 15 hours ago Steve Borho cmake: do not allow full path of libnuma to be used in x265.pc
    • 3 weeks ago Aarthi Thirumalai rc: tune initial predictor values for better frame size predictions in vbv lookaheadstable
    • 9 days ago Aarthi Thirumalai rc: separate frame bits predictor objects for BRef and B framesstable
Version 1.6.147
  • Release Date: Apr 9, 2015
    • 11 hours ago Steve Borho Merge with stable
    • 26 hours ago Steve Borho slicetype: run pre-lookahead within slicetypeDecide using a bonded task groupstable
    • 26 hours ago Steve Borho slicetype: simplify locking scheme to a single lockstable
    • 17 hours ago Rajesh Paulraj asm: improve avx2 8bpp code for convert_p2s[64xN]
    • 17 hours ago Rajesh Paulraj asm: improve avx2 8bpp code for convert_p2s[32xN]
    • 20 hours ago Praveen Tiwari asm: intra_pred_ang16_3 improved by +AH4-25% over SSE4
    • 21 hours ago Praveen Tiwari asm: intra_pred_ang16_4 improved by +AH4-23% over SSE4
    • 24 hours ago Praveen Tiwari asm: intra_pred_ang16_5 improved by +AH4-16% over SSE4
    • 25 hours ago Praveen Tiwari asm: intra_pred_ang16_6 improved by +AH4-19% over SSE4
    • 17 hours ago Min Chen optimize maxAbsLevel cost compute logic in rdoQuant()
    • 21 hours ago Divya Manivannan asm: saoCuOrgE1_2Rows avx2 code: 657c->525c
    • 21 hours ago Min Chen asm: avx2 version convert_p2s[48x64], 4069c -> 3043c
    • 21 hours ago Divya Manivannan asm: saoCuOrgE1 avx2 code: 403c->331c
    • 43 hours ago Divya Manivannan sao: add C and sse4 code of saoCuOrgE1 to process 2 rows

Version 1.6.114
  • Release Date: Apr 8, 2015
    • 4 hours ago Steve Borho level: allow unbounded level 8.5 to be used for lossless encodes
    • 8 hours ago Xinyue Lu cli: rewrite pts_queue to use new/delete, not to confuse the leak tool
    • 18 hours ago Min Chen improve rdoQuant by more parameters on getSigCoeffGroupCtxInc and calcPatternSigCtx
    • 18 hours ago Min Chen fix count of shift overflow bug in Quant::getSigCoeffGroupCtxInc
    • 18 hours ago Min Chen improve rdoQuant by reduce type convert and condition check
    • 19 hours ago Rajesh Paulraj asm: ssse3 8bpp code for chroma_p2s i422, reuse luma code
    • 21 hours ago Rajesh Paulraj asm: ssse3 8bpp code for chroma_p2s[8x6](4.74x) for i420
    • 21 hours ago Rajesh Paulraj asm: sse4 8bpp code for chroma_p2s[6xN] for i420, i422
    • 21 hours ago Rajesh Paulraj asm: ssse3 8bpp code for convert_p2s[12xN],[24xN],[48x64]
    • 20 hours ago Sumalatha Polureddy asm: improve old avx2 code for sad[64x48]
    • 21 hours ago Sumalatha Polureddy asm: improve the old avx2 code for sad[64x64]
    • 21 hours ago Divya Manivannan asm: saoCuOrgE0 avx2 code: 756c->629c
    • 22 hours ago Divya Manivannan sao: modify C and SSE4 code for saoCuOrgE0 to process 2 rows
    • 23 hours ago Praveen Tiwari asm: intra_pred_ang4_3 improved by +AH4-41% over SSE4
    • 23 hours ago Praveen Tiwari asm: intra_pred_ang4_4 improved by +AH4-44% over SSE4
    • 24 hours ago Praveen Tiwari asm: intra_pred_ang4_5 improved by +AH4-41% over SSE4
    • 24 hours ago Praveen Tiwari asm: intra_pred_ang4_6 improved by +AH4-36% over SSE4
    • 24 hours ago Praveen Tiwari asm: intra_pred_ang4_7 improved by +AH4-42% over SSE4
    • 25 hours ago Praveen Tiwari asm: intra_pred_ang4_8 improved by +AH4-24% over SSE4
    • 24 hours ago Sumalatha Polureddy asm: improve the old avx2 code for sad[32x24]
    • 26 hours ago Aasaipriya Chandran asm: luma_hvpp[16x16] - 11.39x 5226c
    • 27 hours ago Aasaipriya Chandran asm: chroma_hps[24x32] avx2 - improved 4458c->3583c
    • 27 hours ago Aasaipriya Chandran asm: luma_hps[24x32] avx2 - improved 11545c->6843c
    • 27 hours ago Aasaipriya Chandran asm: luma_hps[12x16] avx2 - improved 3779c->2482c
    • 36 hours ago Steve Borho api: add --allow-non-conformance param, default to False

Version 1.6.89
  • Release Date: Apr 6, 2015
    • 33 hours ago Steve Borho cli: tweak output file logging
    • 34 hours ago David T Yuen asm: intra_pred_ang4_18
    • 34 hours ago David T Yuen asm: intra_pred_ang4_26_sse2
    • 34 hours ago David T Yuen asm: intra_pred4_x filtering
    • 34 hours ago Steve Borho Merge with stable
    • 34 hours ago Steve Borho threading: intrdoduce poke() method for ThreadSafeIntegerstable
    • 2 days ago Steve Borho encoder: do not disable the thread pool if lookahead-slices is enabled
    • 2 days ago Xinyue Lu cli: move raw bitstream output to separate file
    • 2 days ago David T Yuen asm: intra_pred_ang4_18_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_17_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_16_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_15_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_14_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_13_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_12_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_11_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_26_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_10_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_9_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_8_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_7_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_6_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_5_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_4_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_3_sse2 16-bit
    • 2 days ago David T Yuen asm: intra_pred_ang4_2_sse2 16-bit
Version 1.6.64
  • Release Date: Apr 5, 2015
    • 11 hours ago Steve Borho cmake: bump X265_BUILD for x265_cleanup() behavior change
    • 12 hours ago Steve Borho api: make x265_cleanup() a NOP if an encoder is still open
    • 13 hours ago Steve Borho cmake: avoid strict-overflow warnings in slicetype.cpp from GCC 4.9
    • 14 hours ago Steve Borho asm: only 4:4:4 chroma 4-tap filters are configured in asm-primitives.cpp
    • 16 hours ago Rajesh Paulraj asm: sse4 chroma_p2s[4x2](2.29x), ssse3 chroma_p2s[8x2](3.60x) for i420
    • 17 hours ago Rajesh Paulraj asm: ssse3 code for chroma_p2s for i420, i422, i444, reuse the luma code
    • 17 hours ago Rajesh Paulraj asm: ssse3 8bpp code for convert_p2s[32xN],[64xN]
    • 17 hours ago Rajesh Paulraj asm: ssse3 8bpp code for convert_p2s[8xN],convert_p2s[16xN]
    • 17 hours ago Rajesh Paulraj asm: sse4 8bpp code for convert_p2s[4xN]
    • 17 hours ago Rajesh Paulraj primivites: rename luma_p2s to convert_p2s and move into PU
    • 19 hours ago Min Chen improve fillReferenceSamples by merge pixel fill
    • 19 hours ago Min Chen asm: reduce 1 register in quant_avx2
    • 19 hours ago Min Chen asm: general calSign to accelerate sao
    • 20 hours ago Praveen Tiwari asm: reduce code size with macro 'INTRA_PRED_TRANS_STORE_4x4'
    • 21 hours ago Praveen Tiwari asm: intra_pred_ang4_9 improved by +AH4-35% over SSE4
    • 21 hours ago Praveen Tiwari asm: intra_pred_ang4_11 improved by +AH4-31% over SSE4
    • 21 hours ago Praveen Tiwari asm: intra_pred_ang4_12 improved by +AH4-35% over SSE4
    • 22 hours ago Praveen Tiwari asm: intra_pred_ang4_13 improved by +AH4-43% over SSE4
    • 23 hours ago Praveen Tiwari asm: intra_pred_ang4_14 improved by +AH4-43% over SSE4
    • 23 hours ago Praveen Tiwari asm: intra_pred_ang4_15 improved by +AH4-53% over SSE4
    • 23 hours ago Praveen Tiwari asm: intra_pred_ang4_16 improved by +AH4-49% over SSE4
    • 23 hours ago Praveen Tiwari asm: intra_pred_ang4_17 improved by +AH4-57% over SSE4
    • 20 hours ago Dnyaneshwar G asm: avx2 code for intra_dc_32x32
    • 24 hours ago Dnyaneshwar G asm: avx2 code for intra_planar_32x32
    • 24 hours ago Dnyaneshwar G asm: avx2 code for intrapred_planar16x16
    • 25 hours ago Dnyaneshwar G asm: remove duplicate constant pw_256 and alignment nits

Version 1.6.38
  • Release Date: Apr 4, 2015
    • 12 hours ago Steve Borho cli: improve log message
    • 12 hours ago Steve Borho cli: use general_log for recon file data
    • 12 hours ago Steve Borho common: remove redundant check for log-level, already at the top of the function
    • 33 hours ago David T Yuen asm: intra pred 4x4 modes 19-25 and 27-33
    • 34 hours ago David T Yuen asm: intra_pred_ang4_18_sse2
    • 34 hours ago David T Yuen asm: intra_pred_ang4_17_sse2
    • 34 hours ago David T Yuen asm: intra_pred_ang4_16_sse2
    • 35 hours ago David T Yuen asm: intra_pred_ang4_15_sse2
    • 35 hours ago David T Yuen asm: intra_pred_ang4_14_sse2
    • 35 hours ago David T Yuen asm: intra_pred_ang4_13_sse2
    • 35 hours ago David T Yuen asm: intra_pred_ang4_12_sse2
    • 35 hours ago David T Yuen asm: intra_pred_ang4_11_sse2
    • 34 hours ago David T Yuen asm: intra_pred_ang4_26_sse2
    • 35 hours ago David T Yuen asm: intra_pred_ang4_10_sse2
    • 19 hours ago Min Chen asm: reduce count of register in calSign, 213c -> 202c
    • 21 hours ago Praveen Tiwari asm: reduce code size with macro 'INTRA_PRED_STORE_4x4'
    • 22 hours ago Praveen Tiwari asm: intra_pred_ang4_19 improved by +AH4-60% over SSE4
    • 22 hours ago Praveen Tiwari asm: intra_pred_ang4_20 improved by +AH4-52% over SSE4
    • 22 hours ago Praveen Tiwari asm: intra_pred_ang4_21 improved by +AH4-50% over SSE4
    • 23 hours ago Praveen Tiwari asm: intra_pred_ang4_22 improved by +AH4-46% over SSE4
    • 23 hours ago Praveen Tiwari asm: intra_pred_ang4_23 improved by +AH4-48 over SSE4
    • 24 hours ago Praveen Tiwari asm: intra_pred_ang4_25 improved +AH4-37% over SSE4
    • 24 hours ago Praveen Tiwari asm: intra_pred_ang4_24 improved by +AH4-38% over SSE4
    • 25 hours ago Praveen Tiwari asm: intra_pred_ang4_33 improved by +AH4-44% over SSE4
    • 25 hours ago Praveen Tiwari asm: intra_pred_ang4_32 improved by +AH4-47% over SSE4
    • 25 hours ago Praveen Tiwari asm: intra_pred_ang4_31, improved by +AH4-43% over SSE4
    • 20 hours ago Xinyue Lu cli: unify x265 log function
    • 20 hours ago Sumalatha Polureddy asm: avx2 code for weight_sp() for 8bpp
    • 42 hours ago Praveen Tiwari asm: intra_pred_ang4_30 improve by +AH4-38% over SSE4, 160.00c -> 99.99c
    • 42 hours ago Praveen Tiwari asm: intra_pred_ang4_29 improved by +AH4-45% over SSE4, 157.78c -> 88.14c
    • 42 hours ago Praveen Tiwari asm: intra_pred_ang4_28 improved by +AH4-38% over SSE4, 152.87c -> 94.98c
    • 43 hours ago Praveen Tiwari asm: intra_pred_ang4_27 improved by +AH4-35% over SSE4, 146.71c -> 95.03c
    • 44 hours ago Praveen Tiwari asm: cleanup unused constant table
    • 47 hours ago Praveen Tiwari asm: intra_pred_ang32_24 improved by +AH4-5% over AVX2
    • 47 hours ago Praveen Tiwari asm: intra_pred_ang32_23 improved by +AH4-10% over AVX2, 1925.55c -> 1738.47c
    • 13 hours ago Steve Borho Merge with stable
    • 13 hours ago Steve Borho Added tag 1.6 for changeset cbeb7d8a4880stable
    • 13 hours ago Steve Borho cli: use multi-lib APIs as a (weak) demonstration
    • 13 hours ago Steve Borho Merge with stable
    • 13 hours ago Steve Borho api: introduce x265_api_get()stable 1.6
    • 13 hours ago Steve Borho doc: api doc nitsstable
    • 35 hours ago David T Yuen asm: intra_pred_ang4_9_sse2 improved +AH4-5% 605.00 -> 572.56 with nits and tweaks
    • 35 hours ago David T Yuen asm: intra_pred_ang4_8_sse2 improved +AH4-5% 609.99 -> 577.50 with nits and tweaks
    • 36 hours ago David T Yuen asm: intra_pred_ang4_7_sse2 improved +AH4-6.5% 634.99 -> 592.50 with nits and tweaks
    • 16 hours ago David T Yuen asm: intra_pred_ang4_6_sse2 improved +AH4-2% 622.50 -> 609.99 with nits and tweaks
    • 34 hours ago David T Yuen asm: intra_pred_ang4_5_sse2 improved +AH4-2.5% 642.50 -> 627.50 with nits and tweaks
    • 34 hours ago Steve Borho api: remove public funcdef and export for deprecated x265_setup_primitives()stable
    • 36 hours ago Steve Borho api: comment nitstable
    • 2 days ago Santhoshini Sekar frameencoder: account for hpel iteration steps in motion search rangestable
    • 41 hours ago David T Yuen asm: intra_pred_ang4_4_sse2 improved +AH4-3% 642.49 -> 624.99 with nits and tweaks
    • 42 hours ago David T Yuen asm: intra_pred_ang4_3_sse2 improved +AH4-4.5% 684.95 -> 654.99 with nits and tweaks
    • 45 hours ago Min Chen x86inc: alignment all of const to 32-bytes
    • 2 days ago Dnyaneshwar G picyuv: replace for loops with memcpy
    • 2 days ago Dnyaneshwar G asm: avx2 code for idst4x4
    • 2 days ago Dnyaneshwar G asm: avx2 code for dst4x4
    • 2 days ago David T Yuen asm: intra_pred_ang4_2 improved by +AH4-4% 134.99 -> 129.95
    • 2 days ago Praveen Tiwari asm: intra_pred_ang32_22 improved by +AH4-5% over AVX2, 2308.11c -> 2207.80c
    • 2 days ago Praveen Tiwari asm: intra_pred_ang32_21 improved by +AH4-27% over SSE4, 3439.25c -> 2504.30c
    • 2 days ago Praveen Tiwari asm: reduce binary size
Version 1.5.460
  • Release Date: Apr 1, 2015
    • 5 hours ago Steve Borho Merge with stable
    • 8 hours ago Steve Borho cmake: support PGO for Intel C+-+- (icpc)stable
    • 14 hours ago Steve Borho Merge with stable
    • 24 hours ago Steve Borho cmake: use flags which ignore -fprofile-use warningsstable
    • 25 hours ago Steve Borho cmake: add support for statically linking with C runtime with MinGWstable
    • 24 hours ago Praveen Tiwari asm: intra_pred_ang32_23 improved +AH4-5% over AVX2 code, 1925.55c -> 1833.59c
    • 39 hours ago Praveen Tiwari asm: avx2 intra_pred_ang32_22, improved by +AH4-32%, 3470.37c -> 2308.11c
    • 2 days ago Praveen Tiwari asm: avx2 intra_pred_ang32_23, improved by +AH4-45% over SSE4
    • 18 hours ago Deepthi Nandakumar Merge with stable
    • 42 hours ago Sreelakshmy V G rc: depreciate amortized frame and fraction by 1.1 at each I framestable
    • 23 hours ago Deepthi Nandakumar doc: update fast presetstable

Version 1.5.454
  • Release Date: Apr 1, 2015
    • 4 hours ago Steve Borho cli: rename input/output classes to be more specific
    • 4 hours ago Steve Borho Merge with stable
    • 18 hours ago Santhoshini Sekar rc: fix bug in ABR 2 pass, do not overwrite cutree offsets read from stats filestable
    • 6 hours ago Steve Borho frameencoder: use atomics to determine full completion of wpp and filter workstable
    • 12 hours ago Steve Borho Merge with stable
    • 18 hours ago Xinyue Lu cli: prevent Windows from falling into standby
    • 24 hours ago Dnyaneshwar G asm: replace movd+-pshufd+-vinserti128 with vpbroadcastd instruction
    • 12 hours ago Steve Borho cmake: avoid CMP0054 warnings in main scriptstable
    • 19 hours ago Xinyue Lu cli: save and restore console title before and after encoding
    • 15 hours ago Steve Borho doc: fix typo in --lookahead-slicesstable
    • 36 hours ago Steve Borho cmake: add option for -march=nativestable
    • 37 hours ago Steve Borho cmake: introduce fprofile optionsstable
    • 3 days ago Steve Borho Merge with stable
    • 3 days ago Steve Borho regression: refine max-refs tests, since we validate with the HM decoderstable
    • 3 days ago Min Chen split scale_t to scale1D_t and scale2D_t
    • 3 days ago Min Chen intra: replace 64x64 scaled reference pixel copy loop by memcpy

Version 1.5.445
  • Release Date: Mar 29, 2015
    • 13 hours ago Steve Borho Merge with stable
    • 14 hours ago Steve Borho param: add missing rdoq-level and deblock offsets to param2string()stable
    • 26 hours ago Dnyaneshwar G asm: scale2D_64to32 avx2 code
    • 44 hours ago Praveen Tiwari asm: intra_pred_ang32_24 improved by +AH4-53% over SSE4
    • 47 hours ago Praveen Tiwari asm: intra_pred_ang32_25 improved by +AH4-53% over SSE4
    • 2 days ago Praveen Tiwari asm: intra_pred_ang32_33 improved by +AH4-35% over SSE4
    • 16 hours ago Steve Borho search: comment nitsstable
    • 16 hours ago Steve Borho regression: test-file format can't handle commas in argumentsstable
    • 26 hours ago Steve Borho regression: overhaul test coverage, many more features covered in less test timestable
Version 1.5.439
  • Release Date: Mar 28, 2015
    • 11 hours ago Steve Borho Merge default into stable, prep for 1.6
    • 15 hours ago Steve Borho docs: update threading details of lookaheaddefault
    • 15 hours ago Steve Borho docs: make the --b-adapt docs less embarrasingly bad
    • 15 hours ago Steve Borho docs: add documentation for --lookahead-slices
    • 15 hours ago Steve Borho param: expose 'lookahead-slices' for param.lookaheadSlices (help hidden at H1)
    • 15 hours ago Steve Borho slicetype: respect new m_param->lookaheadSlices param
    • 16 hours ago Steve Borho api: add param.lookaheadSlices
    • 16 hours ago Steve Borho api: comment nits
    • 20 hours ago Sumalatha Polureddy asm: fix alignment issue in ssd
    • 27 hours ago Aasaipriya Chandran asm: chroma_hps[6x8] avx2 - improved 670c->602c
    • 27 hours ago Steve Borho slicetype: re-enable batching, keep coop slices disabled
    • 28 hours ago Steve Borho check sched failures
    • 35 hours ago Steve Borho slicetype: disable b-adapt 2 work batching, temporarily
    • 35 hours ago Steve Borho slicetype: nit
    • 35 hours ago Steve Borho slicetype: use auto-vars to catch race hazards
    • 37 hours ago Steve Borho slicetype: cleanup runtime stat-collection, fewer ifdefs within the functions
    • 37 hours ago Steve Borho slicetype: add a check for cost estimate failure
    • 2 days ago Min Chen improve codeCoeffNxN by new fast RD path
    • 2 days ago Min Chen improve codeCoeffNxN by merge loop of encodeBin
    • 2 days ago Min Chen improve codeCoeffNxN by inline getSigCtxInc()

Version 1.5.420
  • Release Date: Mar 26, 2015
    • 14 hours ago Steve Borho Backed out changeset: 30e713269c6f
    • 15 hours ago Steve Borho search: use pre-calculated size index for psy-energy (nit)
    • 15 hours ago Steve Borho analysis: initialize merge costs at RD levels 0..4
    • 20 hours ago Praveen Tiwari asm: intra_pred_ang32_32 improved by +AH4-46% over SSE4
    • 21 hours ago Praveen Tiwari asm: intra_pred_ang32_31 improved by +AH4-54% over SSE4
    • 21 hours ago Praveen Tiwari asm: intra_pred_ang32_30 improved by +AH4-57% over SSE4 asm
    • 21 hours ago Praveen Tiwari asm: intra_pred_ang32_29 improved by +AH4-58%, 3763.98c -> 1562.61c over SSE4
    • 21 hours ago Praveen Tiwari asm: intra_pred_ang32_28 improved by +AH4-56%, 3234.68c -> 1421.76c over SSE4
    • 21 hours ago Praveen Tiwari asm: intra_pred_ang32_27 improved by +AH4-61%, 3402.39c -> 1322.11c over SSE4
    • 24 hours ago Dnyaneshwar G asm: psyCost_ss avx2 code for all sizes(4x4,8x8,16x16,32x32,64x64)
    • 24 hours ago Sumalatha Polureddy asm: avx2 code for see_pp for chroma 16x16, 32x32, reused luma code
    • 26 hours ago Sumalatha Polureddy asm: avx code for satd for chroma sizes 420, reused the luma code
    • 26 hours ago Sumalatha Polureddy asm: call avx code for copy_ss[32x32] and chroma copy_ss[32x32]
    • 26 hours ago Sumalatha Polureddy asm: avx2 code for sub_ps for chroma sizes 16x16, 32x32, reused the code from luma
    • 26 hours ago Sumalatha Polureddy asm: avx2 code for add_ps for chroma sizes 16x16, 32x32, reused the code from luma
    • 26 hours ago Sumalatha Polureddy asm: avx2 code sse_pp[32x32] and sse_pp[64x64] for 8 bpp
    • 27 hours ago Aasaipriya Chandran asm: chroma_hpp[24x32] avx2 - improved 4525c->4289c
    • 27 hours ago Aasaipriya Chandran asm: chroma_hpp[12x16] avx2 - improved 1483c->1176c
    • 27 hours ago Aasaipriya Chandran asm: chroma_hps[2x8] avx2 - improved 502c->378c
    • 27 hours ago Aasaipriya Chandran asm: chroma_hps[2x4] avx2 - improved 348c->274c
    • 27 hours ago Aasaipriya Chandran asm: chroma_hps[8x2, 8x4, 8x6, 8x16, 8x32] avx2 - improved 298c->228c, 397c->312c, 455c->348c, 936c->736c, 1696c->1319c
    • 23 hours ago Gopu Govindaswamy cmake: use CMAKE_CURRENT_SOURCE_DIR as location to run hg commands find a version
    • 33 hours ago Steve Borho slicetype: do not re-calculate AQ cost of B-frames
    • 33 hours ago Steve Borho rc: nit
    • 21 hours ago Deepthi Nandakumar search: disable psyEnergy checks
Version 1.5.395
  • Release Date: Mar 25, 2015
    • 11 hours ago Steve Borho slicetype: spleling
    • 11 hours ago Steve Borho slicetype: fix crash when lookaheadDepth is 0
    • 11 hours ago Steve Borho param: bframes can match lookaheadDepth if both are zero (fixes #118)
    • 12 hours ago Steve Borho analysis: only perform checks if merge mode was selected
    • 41 hours ago Dnyaneshwar G asm: psyCost_pp avx2 code for BLOCK_4x4
    • 47 hours ago Dnyaneshwar G asm: psyCost_pp avx2 code for BLOCK(8x8,16x16,32x32,64x64)
    • 27 hours ago Sumalatha Polureddy asm: avx2 code for ssd_s[16x16] for 8bpp
    • 35 hours ago David T Yuen asm: intra_pred_ang4_9_sse2
    • 35 hours ago David T Yuen asm: intra_pred_ang4_8_sse2
    • 35 hours ago David T Yuen asm: intra_pred_ang4_7_sse2
    • 36 hours ago David T Yuen asm: intra_pred_ang4_6_sse2
    • 36 hours ago David T Yuen asm: intra_pred_ang4_5_sse2
    • 36 hours ago David T Yuen asm: intra_pred_ang4_4_sse2
    • 36 hours ago David T Yuen asm: intra_pred_ang4_3_sse2
    • 36 hours ago David T Yuen asm: intra_pred_ang4_2_sse2
Version 1.5.380
  • Release Date: Mar 24, 2015
    • 7 hours ago Steve Borho analysis: do not check bidir cost (pmode RD5/6) if not B SLICE
    • 7 hours ago Steve Borho analysis: add missing initCosts() calls for RD 5/6 merge
    • 7 hours ago Steve Borho analysis: add missing initCosts() calls for --cu-lossless
    • 9 hours ago Steve Borho analysis: add invalidate(), ok() methods to Mode for debugging purposes
    • 7 hours ago Steve Borho Merge with stable
    • 7 hours ago Steve Borho analysis: fix merge logic for RD levels 0..4stable
    • 15 hours ago Steve Borho regression: remove '--scaling-list default' from i444 test
    • 15 hours ago Steve Borho cmake: use CMAKE_SOURCE_DIR as location to run hg/git commands to find version
    • 5 days ago Dnyaneshwar G asm: addAvg avx2 code for chroma sizes width >= 8, reused code from luma
    • 26 hours ago Aasaipriya Chandran asm: luma_hps[64x64 , 64x48 , 64x32 , 64x16] avx2 - improved 58533c->28919c, 43517c->22248c, 31254c->15788c, 17540c->8854c
    • 27 hours ago Divya Manivannan asm: filter_hpp[2x8] in avx2: 301c->249c

Version 1.5.370
  • Release Date: Mar 24, 2015
    • 5 hours ago Steve Borho encoder: add explicit synchronization in frame thread startup
    • 14 hours ago Steve Borho smoke-test: nits
    • 15 hours ago Steve Borho regression: do not use --pools 0 to disable pool features
    • 15 hours ago Steve Borho frameencoder: inline compressCTU()
    • 15 hours ago Steve Borho wavefront: paranoid bitmap clears

Version 1.5.364
  • Release Date: Mar 22, 2015
    • 104 minutes ago Steve Borho threadpool: more bitmasks must be sleepbitmap_t
    • 10 hours ago Steve Borho threadpool: explicit upcast of sleep bit to bitmap type
    • 16 hours ago Min Chen asm: x64 version findPosLast, 73815c -> 25890c (2.85x)
    • 2 days ago Rajesh Paulraj asm: avx2 sub_ps 8bpp - use loop for 32x32 and 64x64
    • 27 hours ago Aasaipriya Chandran asm: luma_hps[48x64] avx2 - improved 42120c->21918c
    • 27 hours ago Aasaipriya Chandran asm: luma_hps[32x8, 32x16, 32x24, 32x32, 32x64] avx2 - improved 6143c->3007c, 9343c->4767c, 12806c->6238c, 16065c->8125c, 29073c->14615c
    • 28 hours ago Divya Manivannan asm: filter_hpp[6x8] in avx2: 559c->415c
    • 28 hours ago Divya Manivannan asm: filter_hpp[8x6] in avx2: 360c->234c
    • 28 hours ago Divya Manivannan asm: filter_hpp[8x2] in avx2: 150c->107c
    • 28 hours ago Divya Manivannan asm: luma_hpp[4x8, 4x16] in avx2: 624c->387c, 1141c->735c
    • 24 hours ago Ashok Kumar Mishra removed extra condition check in topSkipMinDepth()
    • 28 hours ago Steve Borho threadpool: wait until each worker is idle before stopping them
    • 19 hours ago Dnyaneshwar G cmake: fix policy CMP0054 warning for cmake version > 3.0
    • 18 hours ago Mahesh Pittala regression: fix inputsequence filename format
    • 2 days ago Sumalatha Polureddy asm: avx2 code for sad[48x64] for 8bpp
    • 2 days ago Sumalatha Polureddy asm: avx2 code for sad[32x64] for 8bpp
    • 2 days ago Dnyaneshwar G asm: addAvg avx2 code for luma width >= 8
    • 2 days ago Divya Manivannan asm: filter_hpp[2x4] in avx2: 185c->161c
    • 2 days ago Divya Manivannan asm: filter_vsp[6x8], filter_vss[6x8] in avx2: 863c->653c, 823c->524c
    • 2 days ago Divya Manivannan asm: filter_vsp[2x8], filter_vss[2x8] in avx2: 482c->346c, 431c->306c
    • 2 days ago Divya Manivannan asm: filter_vsp[32x24], filter_vss[32x24] in avx2: 14478c->5704c, 11464c->6074c
    • 2 days ago Divya Manivannan asm: filter_vsp[8x16, 8x32], filter_vss[8x16, 8x32] in avx2
    • 2 days ago Praveen Tiwari asm: intra_pred_ang32: reduce register count
    • 2 days ago Praveen Tiwari asm: intrapred8.asm: reduce asm code size
    • 2 days ago Praveen Tiwari asm: intra_pred_ang32_26: improve performance by +AH4-65% over SSE4
    • 2 days ago Praveen Tiwari asm: intrapred8.asm: asm code size reduction
    • 2 days ago Praveen Tiwari asm: intra_pred_ang32_2: improve performance by +AH4-48% over SSSE3
    • 2 days ago Praveen Tiwari asm: intra_pred_ang32_34: improve performance by +AH4-53% over SSSE3
    • 2 days ago Aasaipriya Chandran asm: chroma_hpp[16x4, 16x8, 16x12, 16x32] for i420 avx2 - improved 444c-> 276c, 824c->504c, 1144c->730c, 2913c->1871c
    • 2 days ago Aasaipriya Chandran asm: luma_hpp[12x16] avx2 - improved 2899c->1808c
    • 2 days ago Aasaipriya Chandran asm: luma_hpp[24x32] avx2 - improved 6186c->5068c
Version 1.5.334
  • Release Date: Mar 20, 2015
    • 6 hours ago Steve Borho encoder: msvc does not like const char* passed to free()
    • 6 hours ago Steve Borho encoder: add missing free() statements for strdup() in x265_param_parse
    • 9 hours ago Steve Borho regression: --cu-lossless needs --rd 3 or higher, combine tests
    • 17 hours ago Steve Borho rc-test: missing hyphen

Version 1.5.330
  • Release Date: Mar 19, 2015
    • 8 hours ago Steve Borho rc-test: reduce veryslow encodes, fix a couple broken tests
    • 8 hours ago Steve Borho smoke-test: ABR tests need to force -FN, remove 1080p slow test case
    • 2 days ago Aarthi Thirumalai rc: release RC lock after RateControlStart for CQP and CRF(without VBV)
    • 26 hours ago Sumalatha Polureddy asm: comment satd_4x4 module for XOP (fixes #114)
    • 24 hours ago Deepthi Nandakumar quant[BUG]: correct scaling list type
    • 16 hours ago David T Yuen bug: fixed gcc compiler error: expected primary-expression before ???)??? token
    • 26 hours ago Divya Manivannan asm: filter_vsp[8x6], filter_vss[8x6] in avx2: 716c->434c, 705c->405c
    • 26 hours ago Divya Manivannan asm: filter_vsp[8x2], filter_vss[8x2] in avx2: 370c->183c, 331c->175c
    • 26 hours ago Divya Manivannan asm: filter_vsp[32x8, 16x8], filter_vss[32x8, 16x8] in avx2
    • 26 hours ago Divya Manivannan asm: filter_vsp[16x4], filter_vss[16x4] in avx2: 987c->575c, 862c->600c
    • 26 hours ago Divya Manivannan asm: filter_vsp[16x12], filter_vss[16x12] in avx2: 2394c->1473c, 1912c->1550c
    • 26 hours ago Divya Manivannan asm: filter_vpp[12x16], filter_vps[12x16] in avx2: 1116c->968c, 971c->804c
    • 26 hours ago Sumalatha Polureddy asm: Comment satd_4x4 module for XOP
    • 43 hours ago Praveen Tiwari asm-intrapred8.asm: introduce macro 'INTRA_PRED_ANG16_MC5' to reduce code size
    • 43 hours ago Praveen Tiwari intra_pred_ang16_22: improve speedup by +AH4-19 over SSE4
    • 45 hours ago Praveen Tiwari asm-intra_pred_ang16_23: improve speedup by +AH4-22% over SSE4
    • 2 days ago Praveen Tiwari asm-intra_pred_ang16_24: improved speedup by +AH4-40 over SSE4
    • 20 hours ago Deepthi Nandakumar param: explicitly cast to int, prevent warning
    • 30 hours ago Steve Borho api: allow x265_cleanup() to reset the configured CTU size (closes #110)
    • 31 hours ago Steve Borho param: allow --cip shorthand for constrained intra
    • 31 hours ago Steve Borho encoder: disable some params implicitly if they are for disabled features
    • 31 hours ago Steve Borho param: nits
    • 31 hours ago Steve Borho param: auto-wrap tools list, add more tool options and group by category
    • 32 hours ago Min Chen asm: wrapper for findPosLast loop
    • 24 hours ago Deepthi Nandakumar presets: both rdoq and psy-rdoq are disabled at medium/faster presets.
    • 29 hours ago Aarthi Thirumalai ratecontrol: enable code erroneously disabled in 638a3af86fa1
Version 1.5.304
  • Release Date: Mar 18, 2015
    • 9 hours ago Steve Borho cleanups
    • 12 hours ago Aarthi Thirumalai dpb: fix decoder crash with --temporal-layers option
    • 14 hours ago Aarthi Thirumalai rc: fix buffer underflow issue in vbv
    • 26 hours ago Sumalatha Polureddy asm: avx2 code for sad[32x24] for 8bpp
    • 26 hours ago Sumalatha Polureddy asm: avx2 code for sad[32x16] for 8bpp
    • 26 hours ago Sumalatha Polureddy asm: avx2 code for sad[32x8] for 8bpp
    • 26 hours ago Sumalatha Polureddy asm: avx2 code for sad[64x64] for 8bpp
    • 26 hours ago Sumalatha Polureddy asm: avx2 code for sad[64x48] for 8bpp
    • 26 hours ago Sumalatha Polureddy asm: avx2 code for sad[64x32] for 8bpp
    • 26 hours ago Sumalatha Polureddy asm: avx2 code for sad[64x16] for 8bpp
    • 41 hours ago Praveen Tiwari asm: intra_pred_ang16_28: use macro 'INTRA_PRED_ANG16_MC2' to reduce code size
    • 41 hours ago Praveen Tiwari asm: intra_pred_ang16_33: use macro 'INTRA_PRED_ANG16_MC3' to reduce code size
    • 41 hours ago Praveen Tiwari asm: intrapred8.asm: added macro 'INTRA_PRED_ANG16_MC4' to reduce code size
    • 41 hours ago Praveen Tiwari asm: intrapred8.asm: added macro 'INTRA_PRED_ANG16_MC3' to reduce code size
    • 41 hours ago Praveen Tiwari asm: intrapred8.asm: added macro 'INTRA_PRED_ANG16_MC2' to reduce code size
    • 41 hours ago Praveen Tiwari asm: intra_pred_ang16_33: improve +AH4-16% on SSE4
    • 42 hours ago Praveen Tiwari asm: intra_pred_ang16_32: improve +AH4-45% on SSE4
    • 42 hours ago Praveen Tiwari asm: intra_pred_ang16_31: improve +AH4-28% on SSE4
    • 42 hours ago Praveen Tiwari asm: intra_pred_ang16_30: improve +AH4-39% on SSE4
    • 26 hours ago Rajesh Paulraj asm: avx2 8bpp code for sub_ps[16x16](8.06x), sub_ps[32x32](10.20x),
    • 26 hours ago Divya Manivannan asm: filter_vsp[12x16], filter_vss[12x16] in avx2: 3987c->1530c, 3305c->1482c
    • 27 hours ago Divya Manivannan asm: filter_vsp[8x4], filter_vss[8x4] in avx2: 598c->293c, 586c->290c
Version 1.5.282
  • Release Date: Mar 18, 2015
    • 6 hours ago Steve Borho typo
    • 7 hours ago Steve Borho tweak smoke test coverage
    • 11 hours ago Min Chen asm: improve sad[32x32] 10% by unroll loop
    • 10 hours ago Steve Borho Merge with stable
    • 17 hours ago Mahesh Pittala Add few more commandlines for smoke test, fix/add rate control testsstable
    • 14 hours ago Steve Borho pixel: remove unnecessary check statementstable
    • 13 hours ago Min Chen asm: remove unused IACA test codestable
    • 21 hours ago Aarthi Thirumalai rc: bug fix for enforcing qpFile in 2 pass(fixes issue #111)
    • 26 hours ago Aasaipriya Chandran asm: chroma_hpp[4x8, 4x16] for i420 avx2 - improved 346c->322c, 610c->586c
    • 27 hours ago Aasaipriya Chandran asm: chroma_hpp[8x4, 8x16, 8x32] for i420 avx2 - improved 289c->220c, 928c->618c, 1802c->1128c
    • 27 hours ago Sumalatha Polureddy asm: avx2 code for sad[32x32] for 8bpp
    • 27 hours ago Aasaipriya Chandran asm: chroma_hpp[32x16, 32x24, 32x8] for i420 - improved 2966c->1647c, 4514c->2627c, 1494c->870c
    • 27 hours ago Aasaipriya Chandran asm : chroma_hpp[4x2] for i420 avx2 - improved 138c->134c
    • 3 days ago Praveen Tiwari asm-intra-pred8.asm: replace 'lea' instruction with faster 'add' instruction
    • 3 days ago Praveen Tiwari asm-intra_pred_ang16_25: reduce const table address size
    • 3 days ago Praveen Tiwari asm-intra_pred_ang16_29: improved, 866.95c -> 493.20c over SSE4 asm code
    • 3 days ago Praveen Tiwari asm-intra_pred_ang16_27: improved, 645.84c -> 415.14c over SSE4 asm code
    • 3 days ago Praveen Tiwari asm-intrapred8.asm: use macro 'INTRA_PRED_ANG16_MC1' to shorten asm code length
    • 3 days ago Praveen Tiwari asm-intrapred8.asm: rename macro 'INTRA_PRED_ANG16_25' to 'INTRA_PRED_ANG16_MC1'
    • 3 days ago Praveen Tiwari asm-intra_pred_ang16_28: improved, 865.16c -> 456.44c
    • 27 hours ago Divya Manivannan asm: filter_vsp[2x4], filter_vss[2x4] in avx2: 292c->189c, 248c->184c
    • 27 hours ago Divya Manivannan asm: filter_vsp[4x2], filter_vss[4x2] in avx2: 237c->137c, 206c->118c
    • 27 hours ago Divya Manivannan asm: filter_vsp[4x16], filter_vss[4x16] in avx2: 985c->599c, 877c->491c
    • 27 hours ago Divya Manivannan asm: filter_vsp[4x8], filter_vss[4x8] in avx2: 673c->339c, 608c->263c
    • 21 hours ago Deepthi Nandakumar predict: check for out of bounds ref indices before weight tables are used
    • 23 hours ago Deepthi Nandakumar merge: check merge reference indices
    • 23 hours ago Deepthi Nandakumar merge: set zero MV correctly
Version 1.5.258
  • Release Date: Mar 16, 2015
    • 15 hours ago Steve Borho Merge with stable
    • 15 hours ago Steve Borho common: move debug break at the end of check failure logicstable
    • 16 hours ago Steve Borho encoder: remove unnecessary funcdef
    • 16 hours ago Steve Borho encoder: split out VPS initialization, just to be consistent
    • 28 hours ago Steve Borho Merge with stable
    • 32 hours ago Steve Borho pixel: fix stride alignment checks to be byte basedstable
    • 37 hours ago Steve Borho common: in checked builds, use hard abort when errors are foundstable
    • 37 hours ago Steve Borho threading: in posix timedWait(), clear timeout flag if event was triggeredstable
    • 40 hours ago Steve Borho Merge with stable
    • 40 hours ago Steve Borho tests: rdoq-level 2 not supported on the stable branch at this timestable

Version 1.5.251
  • Release Date: Mar 15, 2015
    • 11 hours ago Steve Borho tests: add default branch features
    • 5 hours ago Steve Borho Merge with stable
    • 12 hours ago Steve Borho test: add lists of regression tests to be used by external test harnessstable
    • 14 hours ago Steve Borho Merge with stable
    • 14 hours ago Steve Borho encoder: X265_MAX_FRAME_THREADS did not exist on stablestable
    • 15 hours ago Steve Borho Merge with stable
    • 30 hours ago David T Yuen asm: intra pred planar32 sse2 high bit
    • 30 hours ago David T Yuen asm: intra pred planar32 sse2
    • 16 hours ago Aarthi Thirumalai rc: fix bug in CRF caused by e0f834c69cb2
    • 28 hours ago Sreelakshmy V G analysis: add logic for calculate qp for a given cu size
    • 26 hours ago Deepthi Nandakumar encoder: set frame thread count correctly when no-wpp is enabledstable
    • 21 hours ago Deepthi Nandakumar presets[OUTPUT CHANGE]: change superfast and ultrafast presets
    • 2 days ago Aarthi Thirumalai rc: clip qp after initial vbv-lookahead estimation.
    • 2 days ago Aarthi Thirumalai rc: recompute planned frame size when using vbv with --qpfile
    • 2 days ago Aasaipriya Chandran asm: chroma_hps[32x8, 32x16, 32x24] for i420 avx2 - improved 1843c->1210c, 3149c->2001c, 4440c->2906c
    • 2 days ago Aasaipriya Chandran asm: chroma_hps[16x4, 16x8, 16x12, 16x32] for i420 avx2 - improved 743c->468c, 1065c->681c, 1399c->894c, 2961c->1844c
    • 2 days ago Aasaipriya Chandran asm : chroma_hps[4x8 , 4x16] for i420 avx2 - improved 557c->377c, 873c->588c
    • 2 days ago Aasaipriya Chandran asm : chroma_hps[4x2] for i420 avx2 - improved 329c->233c
    • 26 hours ago Deepthi Nandakumar encoder: set frame thread count correctly when no-wpp is enabled
    • 44 hours ago Divya Manivannan asm: filter_vsp[16x32, 24x32, 32x32], filter_vss[16x32, 24x32, 32x32] in avx2
    • 45 hours ago Divya Manivannan asm: filter_vsp[16x16, 32x16], filter_vss[16x16, 32x16]: 3042c->1875c, 5844c->3724c, 2646c->1988c, 4655c->4040c
    • 2 days ago Divya Manivannan asm: filter_vsp[8x8], filter_vss[8x8] in avx2: 887c->525c, 828c->524c
    • 2 days ago Divya Manivannan asm: filter_vsp[4x4], filter_vss[4x4] in avx2: 407c->198c, 361c->180c
    • 42 hours ago Praveen Tiwari asm-intra_pred_ang16_25: improved, 781.13c -> 466.16c
    • 43 hours ago Praveen Tiwari asm-intra_pred_ang8_11: improved, 317.84c -> 230.29c over SSE4 asm code
    • 28 hours ago Min Chen asm: improve +AH4-5% on AVX2 interp_8tap_horiz_ps_4xN
    • 28 hours ago Divya Manivannan asm: avx2 code for filter_vpp[16x4], filter_vps[16x4]: 303c->293c, 311c->253c
    • 28 hours ago Deepthi Nandakumar entropy: add check failure
Version 1.5.228
  • Release Date: Mar 14, 2015
    • 3 days ago Ashok Kumar Mishra cache spatial and temporal PMVs to find the best MV for each reference index
    • 13 hours ago Steve Borho cmake: separate aggressive check flags from CHECKED_BUILD option
    • 43 hours ago Divya Manivannan asm: filter_vpp[24x32], filter_vps[24x32] in avx2: 2901c->2784c, 2876c->2511c
    • 47 hours ago Divya Manivannan asm: filter_vpp[16x32], filter_vps[16x32] in avx2: 1944c->1870c, 1907c->1628c
    • 47 hours ago Divya Manivannan asm: filter_vpp[16x12], filter_vps[16x12] in avx2: 753c->728c, 740c->628c
    • 46 hours ago Praveen Tiwari asm-intra_pred_ang8_12: improved, 359.01c -> 259.03c over SSE4 asm code
    • 47 hours ago Praveen Tiwari asm-intra_pred_ang8_24: improved, 215.89c -> 177.69c over SSE4 asm code
Version 1.5.221
  • Release Date: Mar 12, 2015
    • 4 hours ago Steve Borho frameencoder: initialize m_localTldIdx
    • 4 hours ago Steve Borho encoder: fix --no-wpp with --nr and/or DETAILED_CU_STATS
    • 5 hours ago Steve Borho spleling
    • 6 hours ago Steve Borho motion: initialize more data members
    • 22 hours ago Deepthi Nandakumar param: initialize rdoqLevel
    • 23 hours ago Deepthi Nandakumar doc: correct description for rd 4.
    • 7 hours ago Steve Borho search: no longer needs a pool pointer
    • 7 hours ago Steve Borho analysis: initialize member pointers to NULL
    • 12 hours ago Steve Borho encoder: fix crash with --pools none
    • 24 hours ago Praveen Tiwari asm-intrapred8.asm: prefix 'db' to constant tables
    • 2 days ago Praveen Tiwari asm-avx2: intra_pred, align const
    • 26 hours ago Aasaipriya Chandran asm : chroma_hps[8x8] for i420 avx2 - improved 629c->503c
    • 26 hours ago Aasaipriya Chandran asm : chroma_hps[4x4] for i420 avx2 - improved 336c->214c
    • 26 hours ago Aasaipriya Chandran asm : chroma_hps[16x16]for i420 avx2 - improved 1624c->1093c
    • 26 hours ago Aasaipriya Chandran asm : chroma_hps[32x32] for i420 avx2 - improved 5927c->3933c
    • 45 hours ago Sumalatha Polureddy asm: avx2 code for sad_x4[8x8] for 8bpp
    • 2 days ago Sumalatha Polureddy asm: avx2 code for sad_x3[8x16] for 8bpp
    • 2 days ago Sumalatha Polureddy asm: avx2 code for sad_x3[8x8] for 8bpp
    • 2 days ago Sumalatha Polureddy asm: avx2 code for sad_x3[8x4] for 8bpp
Version 1.5.206
  • Release Date: Mar 11, 2015
    • 23 hours ago Deepthi Nandakumar pool: Threadpool::stop will only stop all worker threads, not destruct them.
    • 25 hours ago Steve Borho pool: separate worker thread stoppage from destructor
Version 1.5.200
  • Release Date: Mar 11, 2015
    • 7 hours ago Steve Borho search: fix GCC warnings and nits
    • 7 hours ago Min Chen asm: split 8bpp version of all_angs from intrapred8.asm
    • 10 hours ago Min Chen asm: alignment constant to 32-bytes bound in intrapred8.asm
    • 14 hours ago Steve Borho Merge with stable
    • 15 hours ago Steve Borho param: disallow encodes without room for P frame in lookaheadstable
    • 3 days ago Steve Borho asm: properly disable x265_stack_align() when ENABLE_ASSEMBLY is OFFstable
    • 15 hours ago Steve Borho param: disallow encodes without room for P frame in lookahead
    • 38 hours ago Steve Borho testbench: add flushes for stdout() before printing errors to stderr
    • 2 days ago Steve Borho cli: nits
    • 3 days ago Steve Borho asm: properly disable x265_stack_align() when ENABLE_ASSEMBLY is OFF
    • 5 days ago Gopu Govindaswamy dqp: add logic to encode DQP at depth zero and above as well
    • 22 hours ago Sreelakshmy V G slicetype[OUTPUT CHANGE]: fix the count of total 16x16 blocks for averaging of qp_adj
    • 22 hours ago Sreelakshmy V G slicetype: Rename variables for better clarity and understanding
    • 4 days ago Praveen Tiwari asm-avx2: intra_pred_ang8_25, (42.92x)
    • 4 days ago Divya Manivannan asm-avx2: filter_vpp[2x8], filter_vps[2x8]: 490c->368c, 438c->298c
    • 4 days ago Divya Manivannan asm-avx2: filter_vpp[6x8], filter_vps[6x8]: 415c->396c, 364c->351c
Version 1.5.186
  • Release Date: Mar 8, 2015
    • 13 hours ago Steve Borho cmake: fix -fsanitize=address use with ENABLE_TESTS=ON
    • 13 hours ago Steve Borho cmake: fix -fsanitize=address use with ENABLE_SHARED=ON
    • 28 hours ago Praveen Tiwari asm-avx2: intra_pred_ang8_9
    • 28 hours ago Praveen Tiwari asm-avx2: intra_pred_ang8_27
    • 15 hours ago Steve Borho asm: upgrade runtime warning to explicit compile error
    • 15 hours ago Steve Borho cmake: use the more aggressive form -fstack-protector-all
    • 15 hours ago Steve Borho cmake: -fstack-protector requires libssp on MinGW
    • 15 hours ago Steve Borho cmake: remove obsolete CC_HAS_FSANITIZE section
    • 15 hours ago Steve Borho cmake: generalize adding additional linker options
    • 15 hours ago Steve Borho cmake: use list append for numa lib (nit)
    • 25 hours ago Aarthi Thirumalai rc: fix crash in strict cbr when totalFrames = 0
Version 1.5.175
  • Release Date: Mar 7, 2015
    • 5 hours ago Steve Borho asm: prevent assembly use in 32bit HBD builds
    • 6 hours ago Steve Borho api: add support for transfer characteristics added in HEVC V2
    • 5 hours ago Steve Borho asm: simply AVX2 code by ignoring 32bit builds and moving functions together
    • 8 hours ago David T Yuen asm: intra pred planar16 sse2 high bit
    • 8 hours ago David T Yuen asm: intra pred planar16 sse2
    • 8 hours ago David T Yuen asm: intra pred planar8 sse2 high bit
    • 8 hours ago David T Yuen asm: intra pred planar8 sse2
    • 9 hours ago David T Yuen asm: intra pred dc32 sse2 high bit
    • 9 hours ago David T Yuen asm: intra pred dc32 sse2
    • 10 hours ago David T Yuen asm: intra pred dc16 sse2 high bit
    • 11 hours ago David T Yuen asm: intra pred dc8 sse2 high bit
    • 11 hours ago David T Yuen asm: intra pred dc4 sse2 high bit
    • 20 hours ago Divya Manivannan asm: filter_vpp[8x2], filter_vps[8x2]: 141c->118c, 131c->113c
    • 21 hours ago Divya Manivannan asm: filter_vpp[8x32], filter_vps[8x32]: 1028c->937c, 902c->860c
    • 22 hours ago Divya Manivannan asm: filter_vpp[4x16], filter_vps[4x16]: 786c->590c, 651c->489c
    • 25 hours ago Divya Manivannan asm: filter_vpp[32x24, 32x16, 32x8], filter_vps[32x24, 32x16, 32x8]
    • 12 hours ago Steve Borho frameencoder: use a bonded worker thread to perform weight analysis, add stat
    • 21 hours ago Praveen Tiwari sao.cpp: init additional pixels for SAO
    • 7 days ago Steve Borho api: make RDOQ level externally configurable, make two levels visible
    • 6 days ago Steve Borho predict: don't pass a Predict member variable to a Predict method
    • 6 days ago Steve Borho predict: move 4:4:4 chroma sample filtering into initAdiPatternChroma()
    • 6 days ago Steve Borho predict: move intra functions together in predict.cpp (no change)
    • 40 hours ago Steve Borho intra: pull the simple 1:2:1 pixel filtering into a performance primitive
    • 6 days ago Steve Borho param: remove --aq-mode 1 from --tune grain
Version 1.5.151
  • Release Date: Mar 5, 2015
    • 24 hours ago Divya Manivannan asm: filter_vpp[8x6], filter_vps[8x6]: improve 277c->226c, 264c->217c
    • 25 hours ago Divya Manivannan asm: filter_vpp[4x2], filter_vps[4x2]: improve 142c->130c, 126c->121c
    • 29 hours ago David T Yuen asm: intra pred planar4 sse2 high bit
    • 29 hours ago David T Yuen asm: intra pred planar4 sse2
    • 2 days ago David T Yuen asm: intra pred dc16 sse2
    • 5 days ago David T Yuen asm: intra pred dc8 sse2
    • 28 hours ago Min Chen asm: improve algorithm on luma_hps[8xN]
    • 2 days ago Min Chen asm: improve intra_pred_dc4_sse4 by merge reduce code
    • 8 hours ago Steve Borho frameencoder: avoid race hazard in end of frame logic
    • 11 hours ago Steve Borho threading: make Windows Event::timedWait() match behavior of POSIX version
    • 17 hours ago Steve Borho frameencoder: use a timed wait to avoid a help-wanted flag race hazard
    • 2 days ago Steve Borho encoder: divide init/start and stop/delete to prevent race hazards
    • 2 days ago Steve Borho slicetype: clear m_helpWanted with input lock held to prevent race
    • 2 days ago Steve Borho me: split initialization to avoid harmless race hazard
    • 15 hours ago Steve Borho weightb: avoid signed left-shifts (which cause -ftrapv exceptions)
    • 18 hours ago Steve Borho primitives: ensure x265_cpu_cpuid() clears eax even when built without asm
    • 18 hours ago Steve Borho frameencoder: nit
    • 17 hours ago Steve Borho threading: nit
    • 5 days ago Steve Borho encoder: protect against logging inf stats if no RDO was performed
    • 17 hours ago Steve Borho cmake: more general support for clang/gcc -fsanitize options
    • 2 days ago Steve Borho cmake: add -fstack-protector to checked builds, if available
    • 2 days ago Steve Borho cmake: ignore older versions of libnuma
Version 1.5.88
  • Release Date: Feb 26, 2015
    • 12 hours ago Steve Borho search: misc variable renames and comment improvements
    • 12 hours ago Steve Borho search: prune temp members from MergeData, there was no caching
    • 12 hours ago Steve Borho search: improve comments and check strings in mergeEstimation()
    • 13 hours ago Steve Borho search: pass PredictionUnit to mergeEstimation - removes redundant data
    • 14 hours ago Steve Borho predict: introduce PredictionUnit structure to maintain current PU geometry
    • 5 hours ago Steve Borho slicetype: remove unused intraPenalty flag to batched estimates
    • 4 hours ago Steve Borho slicetype: workaround for non-determinism
    • 4 hours ago Steve Borho lowres: use CHECKED_MALLOC_ZERO()
    • 5 hours ago Steve Borho slicetype: varname nit
    • 11 hours ago Steve Borho slicetype: zero row-satds prior to cooperative (slice based) cost estimate
    • 12 hours ago Steve Borho slicetype: fix I slice cost estimates
    • 12 hours ago Steve Borho slicetype: nit
    • 12 hours ago Steve Borho slicetype: fencIntra[] alignment is implied by 8x8 size of prediction[]
    • 12 hours ago Steve Borho threadpool: only mention NUMA node in pool message if multiple are present
    • 13 hours ago Steve Borho entropy: nit - initialize m_pad
    • 26 hours ago Steve Borho docs: cleanup restructured text for --pools
Version 1.5.77
  • Release Date: Feb 26, 2015
    • 12 hours ago Steve Borho search: misc variable renames and comment improvements
    • 12 hours ago Steve Borho search: prune temp members from MergeData, there was no caching
    • 12 hours ago Steve Borho search: improve comments and check strings in mergeEstimation()
    • 13 hours ago Steve Borho search: pass PredictionUnit to mergeEstimation - removes redundant data
    • 14 hours ago Steve Borho predict: introduce PredictionUnit structure to maintain current PU geometry
    • 5 hours ago Steve Borho slicetype: remove unused intraPenalty flag to batched estimates
    • 4 hours ago Steve Borho slicetype: workaround for non-determinism
    • 4 hours ago Steve Borho lowres: use CHECKED_MALLOC_ZERO()
    • 5 hours ago Steve Borho slicetype: varname nit
    • 11 hours ago Steve Borho slicetype: zero row-satds prior to cooperative (slice based) cost estimate
    • 12 hours ago Steve Borho slicetype: fix I slice cost estimates
    • 12 hours ago Steve Borho slicetype: nit
    • 12 hours ago Steve Borho slicetype: fencIntra[] alignment is implied by 8x8 size of prediction[]
    • 12 hours ago Steve Borho threadpool: only mention NUMA node in pool message if multiple are present
    • 13 hours ago Steve Borho entropy: nit - initialize m_pad
    • 26 hours ago Steve Borho docs: cleanup restructured text for --pools
Version 1.5.61
  • Release Date: Feb 24, 2015
    • 3 hours ago Steve Borho stats: fix accounting of pmode and pme time, do not double-count master thread
    • 3 hours ago Steve Borho analysis: fixup --pmode and --min-cu-size > 8
    • 4 hours ago Steve Borho slicetype: only count intraMbs when estimating P slices
    • 4 hours ago Steve Borho slicetype: avoid bonded task group if lookahead slices not in use
    • 6 hours ago Steve Borho slicetype: fine tuning for slicetypeDecide batch work
    • 7 hours ago Steve Borho cli: improve CLI help for --pools
    • 8 hours ago Steve Borho cmake: bump build number for --pools vs --threads API change
    • 4 days ago Steve Borho slicetype: refactor lookahead to use bonded task groups
    • 4 days ago Steve Borho pool: introduce bonded task groups
    • 4 days ago Steve Borho NUMA based thread pools
    • 11 hours ago Steve Borho search: use aligned mallocs for transform skip temp buffers
    • 12 hours ago Steve Borho search: use a single aligned malloc for intra analysis buffers
    • 12 hours ago Steve Borho search: typo
    • 12 hours ago Steve Borho Merge with stable
    • 12 hours ago Steve Borho search: hoist bufScale allocation out of its limited scopestable
    • 14 hours ago Steve Borho encoder: w/s nit
    • 14 hours ago Steve Borho encoder: declare defaultAnalysisFileName as static (close #105)
    • 15 hours ago Steve Borho cudata: rename CUGeom encodeIdx to absPartIdx, no behavior change
    • 16 hours ago Steve Borho search: make bMergeOnly argument to predInterSearch more of a suggestion
    • 20 hours ago Deepthi Nandakumar ratecontrol: add check failure
    • 20 hours ago Divya Manivannan testbench: change in stride value

Version 1.5.41
  • Release Date: Feb 22, 2015
    • 33 hours ago David T Yuen asm: dct8 sse2 1.88x improvement over c code
    • 2 days ago Steve Borho param: nits

Version 1.5.34
  • Release Date: Feb 20, 2015
  • 7 hours ago Steve Borho analysis: remove redundant skip flag check
    8 hours ago Steve Borho stats: define a new macro which only increments a counter (no elapsed time)
    9 hours ago Steve Borho analysis: cleanup load/save member variables
    3 days ago Santhoshini Sekar cli: add new option --min-cu-size
    13 hours ago Steve Borho api: introduce --no-mixed-refs type feature interface
    13 hours ago Steve Borho api: nit reorder of defines for better clarity
    14 hours ago Santhoshini Sekar api: allow minimum CU size to be configured
    22 hours ago Praveen Tiwari asm-see: intra_pred_ang4_2, fix xmm register count
    2 days ago Santhoshini Sekar rename variable g_maxFullDepth to g_unitSizeDepth, NUM_CU_PARTITIONS to NUM_4x4_PARTITIONS
    23 hours ago Deepthi Nandakumar Merge with stable
    3 days ago Aarthi Thirumalai slicetype: fix the BRef cost estimates in vbv lookahead.stable
Version 1.5.20
  • Release Date: Feb 17, 2015
    • 2 hours ago Steve Borho Merge with stable
    • 19 hours ago Praveen Tiwari SSE-ASM: fix quant primitive output mismatchstable
    • 21 hours ago Deepthi Nandakumar testbench: include 4x4 sizes alsostable
    • 22 hours ago Deepthi Nandakumar testbench: update quant tests, to capture the right transform sizesstable
    • 19 hours ago Praveen Tiwari SSE-ASM: fix quant primitive output mismatch
    • 21 hours ago Deepthi Nandakumar testbench: include 4x4 sizes also
    • 3 days ago Deepthi Nandakumar entropy: refine check failures
    • 22 hours ago Deepthi Nandakumar testbench: update quant tests, to capture the right transform sizes
    • 25 hours ago Deepthi Nandakumar encoder: add return for clean console outputs
    • 25 hours ago Deepthi Nandakumar param: if maxTUSize > maxCUSize, do not abort. Display a warning and reduce maxTUSize.
    • 3 days ago Deepthi Nandakumar entropy: rename encodeTransform variables for more clarity. No logic change.
    • 29 hours ago Steve Borho Merge with stable
    • 3 days ago Jie Zhang doc: fix aq-mode value on presetstable
Version 1.5.13
  • Release Date: Feb 16, 2015
    • 13 hours ago Steve Borho encoder: do not allow encode to continue if check failure has occurred
    • 2 days ago Steve Borho api: drop fast-cbf option
    • 47 hours ago Santhoshini Sekar CLI: remove wrong short option given to max-tu-size
Version 1.5.12
  • Release Date: Feb 15, 2015
    • 13 hours ago Steve Borho encoder: do not allow encode to continue if check failure has occurred
    • 2 days ago Steve Borho api: drop fast-cbf option
    • 47 hours ago Santhoshini Sekar CLI: remove wrong short option given to max-tu-size
Version 1.5.9
  • Release Date: Feb 12, 2015
    • 27 hours ago Santhoshini Sekar cli: add cli option max-tu-size and support for it
    • 12 hours ago Steve Borho api: reorder pic and params for greater clarity, improve and correct comments
    • 13 hours ago Steve Borho y4m: use ios::good() rather than !ios::eof() (refs #80)
    • 13 hours ago Steve Borho y4m: nit cleanups in parseHeader()
    • 16 hours ago Steve Borho Merge with stable
    • 30 hours ago Steve Borho add a short readme in the root folder for code hosting sites
    • 2 days ago Gopu Govindaswamy analysis: dump and reuse bestCand for skip mode
    • 2 days ago Gopu Govindaswamy analysis: improve Intra Information sharing using analysis-mode
    • 28 hours ago Steve Borho Added tag 1.5 for changeset 9f0324125f53stable
Version 1.4.542
  • Release Date: Feb 12, 2015
    • 10 hours ago Steve Borho Merge with stable
    • 10 hours ago Steve Borho search: use chroma reconQt buffer instead of recon picture (not yet updated)stable
    • 32 hours ago Steve Borho nit: remove remaining :: scope prefixes from memset and memcpy
    • 32 hours ago Steve Borho recon: nits
    • 14 hours ago Steve Borho entropy: we prefer to keep variables without -1/+-1 signal offsets
    • 16 hours ago Steve Borho entropy: w/s nits
    • 23 hours ago Aarthi Thirumalai entropy: fix bug in coding HRD parameters based on the temporal layers.

Version 1.4.536
  • Release Date: Feb 10, 2015
    • 9 hours ago Steve Borho Merge with stable
    • 9 hours ago Steve Borho frame: account for pixel size when initializing recon buffers for SAOstable
    • 9 hours ago Steve Borho picyuv: nits
    • 14 hours ago Steve Borho Merge with stable
    • 20 hours ago Praveen Tiwari sao: avoid access beyond boundsstable
    • 2 days ago Steve Borho encoder: drop warnings about range extensionsstable
    • 3 days ago Steve Borho stats: do not crash if no thread pool was used
Version 1.4.533
  • Release Date: Feb 7, 2015
    • 13 hours ago Steve Borho stats: capture recon generation time at RD levels 0 and 1 as "RDO" time
    • 13 hours ago Steve Borho stats: add more comments
    • 14 hours ago Steve Borho stats: include time spent in slicetype decisions
    • 14 hours ago Steve Borho slicetype: use access macros for alloc and free
    • 15 hours ago Steve Borho stats: note that PSNR/SSIM are measured directly after loop filters per row
    • 26 hours ago Praveen Tiwari blockcopy_pp_8x12: fixed crash on 32-bit

Version 1.4.515
  • Release Date: Feb 6, 2015
    • 11 hours ago Steve Borho dpb: fix msvc int to bool conversion warning
    • 22 hours ago Min Chen improve rdoQuant by split path on different probability
    • 3 months ago Nicolas Morey-Chaisemartin Make FrameEncoder partially virtual so it can be overloaded
    • 45 hours ago Aarthi Thirumalai cli: add --[no-]temporal-layers
    • 45 hours ago Aarthi Thirumalai encoder: Add support for temporal layering of the encoded bitstream.
    • 43 hours ago Praveen Tiwari blockcopy_pp_8x64: sse2 asm code optimization
    • 43 hours ago Praveen Tiwari blockcopy_pp_8x32: sse2 asm code optimization
    • 43 hours ago Praveen Tiwari blockcopy_pp_8x16: sse2 asm code optimization
    • 43 hours ago Praveen Tiwari blockcopy_pp_8x8: sse2 asm code optimization
    • 43 hours ago Praveen Tiwari blockcopy_pp_8x12: sse2 asm code optimization
    • 47 hours ago Praveen Tiwari blockcopy_pp_6x8 sse2 asm code optimization
    • 44 hours ago Praveen Tiwari blockcopy_pp_8x6: optimize register uses
    • 22 hours ago Min Chen c_model: correct weight_sp round parameters check
    • 22 hours ago Deepthi Nandakumar level: fix VPS uninitialized issue
    • 23 hours ago Deepthi Nandakumar doc: update doc about level/tier

Version 1.4.496
  • Release Date: Feb 4, 2015
    • 4 hours ago Steve Borho primitives: drop redundant enums and pointer aliasing
    • 5 hours ago Steve Borho stats: keep intra and inter RDO performance stats per-depth
    • 8 hours ago Steve Borho stats: fix double-counting of chroma intra time
    • 25 hours ago Deepthi Nandakumar level: choose high tier only if the determined level is greater than or equal to
    • 26 hours ago Rajesh Paulraj primitive: rename functions and add testharness code
    • 25 hours ago Praveen Tiwari blockcopy_pp[4x16:4x32], sse2 asm code optimization
    • 26 hours ago Praveen Tiwari blockcopy_pp_4x8 sse2 asm code optimization
    • 44 hours ago Praveen Tiwari blockfill_s_8x8 sse2 asm code optimization
    • 27 hours ago Deepthi Nandakumar level: VBV buffer size needs to be checked before determining tier
Version 1.4.487
  • Release Date: Feb 4, 2015
    • 9 hours ago Steve Borho encoder: remove redundant 'worker', fit within 80 chars
    • 9 hours ago Steve Borho encoder: account for worker time spent in loop filters
    • 10 hours ago Steve Borho encoder: show unnaccounted time and add CPU utilization numbers
    • 10 hours ago Steve Borho encoder: initialize m_threadPool pointer to NULL
    • 10 hours ago Steve Borho encoder: do not report stats if no CU time was recorded
    • 19 hours ago Min Chen remove unneeded intermedia variant in coeff scan loop
    • 13 hours ago Steve Borho api: fix spelling empirically
    • 14 hours ago Steve Borho Merge with stable
    • 34 hours ago Steve Borho pmode: do not call findJob() from task master, avoid double counting of time
    • 36 hours ago Steve Borho search: fix indentation of new brace scope, no logic change
    • 36 hours ago Steve Borho search: seperate intra analysis from RDO in estIntraPredQT(), improve var names
    • 2 days ago Steve Borho search: remove unused zeroPixel
    • 2 days ago Steve Borho search: add compile-time optional detailed CU stats
    • 2 days ago Steve Borho frameencoder: simplify noise reduction update logic flow
    • 2 days ago Steve Borho api: give type name to x265_cli_cspstable
    • 2 days ago Steve Borho rdcost: use a more gradual fall-off function for psy-rd at high QPstable
    • 2 days ago Steve Borho doc: update frame threading docs
    • 2 days ago Steve Borho api: set a limit on the number of frame encoders
    • 2 days ago Steve Borho threading: create a utility class for measuring elapsed time in functions
    • 21 hours ago Praveen Tiwari blockfill_s_32x32 sse2 asm code optimization
    • 22 hours ago Praveen Tiwari blockfill_s_16x16 sse2 asm code optimization
    • 20 hours ago Deepthi Nandakumar Merge with stable
    • 23 hours ago Gopu Govindaswamy threading: use InterlockedExchangeAdd for ATOMIC_ADDstable
    • 4 days ago Deepthi Nandakumar encoder: whitespace nits and document fixesstable
    • 2 days ago Steve Borho rc: fix comment text that was pasted from the HEVC specstable
    • 3 days ago Steve Borho merge default into stable, prep for 1.5 tagstable
    • 3 days ago Steve Borho nit: replace hard-coded 51 with QP_MAX_SPEC
    • 5 days ago Santhoshini Sekar pixelHarness: add testharness code for estimateCUPropagateCost
    • 3 days ago Min Chen improve codeCoeffNxN by calculate context in scanLast loop
    • 3 days ago Satoshi Nakagawa quant: add m_tqBypass
    • 4 days ago Steve Borho encoder: allow 8 frame threads with 4k and many core servers
    • 4 days ago Steve Borho encoder: no longer warn when disabling psy-rdo[q] for rdlevel reasons
    • 4 days ago Steve Borho param: enable psy-rd and psy-rdoq by default
    • 4 days ago Steve Borho rdcost: auto down-scale psy-rd at higher QPs
    • 4 days ago Steve Borho api: nits

Version 1.4.465
  • Release Date: Feb 3, 2015
    • 29 hours ago Steve Borho cmake: bump build number for X265_LOG_FRAME
    • 30 hours ago Steve Borho cli: improve and document return codes
    • 30 hours ago Steve Borho cli: move a param validation into the encoder with other param validations
    • 30 hours ago Steve Borho cli: remove a comment that was out of date 18 months ago
    • 2 days ago Steve Borho encoder: abort on failure to open CSV log file for write
    • 2 days ago Steve Borho stats: introduce X265_LOG_FRAME for file level CSV logging without console logs
    • 34 hours ago Deepthi Nandakumar vps: frameOnlyConstraintFlag is true if fieldSeqFlag is false.
    • 34 hours ago Deepthi Nandakumar stats: nits
    • 2 days ago Steve Borho stats: document the new columms in per-frame CSV files
    • 2 days ago Steve Borho stats: report frame wall time spent waiting for decided frames
    • 2 days ago Steve Borho stats: report wall time of frame encoder with no active worker threads
    • 2 days ago Steve Borho stats: report wall time of wait for reference rows
    • 2 days ago Steve Borho stats: include loop filter processing and all overhead in worker wall time
    • 2 days ago Steve Borho stats: report times in milliseconds
    • 2 days ago Steve Borho stats: report row0wait and frame end overhead seperate from wall time
    • 2 days ago Steve Borho stats: count the number of times top dependencies block worker threads
    • 2 days ago Steve Borho stats: keep timestamps instead of elapsed times, to allow more flexibility
    • 2 days ago Steve Borho frameencoder: use uint32_t more consistently for rows and columns
    • 2 days ago Steve Borho stats: add frame statistic for average WPP benefit
    • 2 days ago Steve Borho stats: keep running count of number of active worker threads per frame encoder
    • 2 days ago Steve Borho threading: add ATOMIC_ADD
    • 4 days ago Steve Borho cli: add a (mostly harmless) missing initializer
Version 1.4.463
  • Release Date: Jan 31, 2015
    • 29 hours ago Steve Borho cmake: bump build number for X265_LOG_FRAME
    • 30 hours ago Steve Borho cli: improve and document return codes
    • 30 hours ago Steve Borho cli: move a param validation into the encoder with other param validations
    • 30 hours ago Steve Borho cli: remove a comment that was out of date 18 months ago
    • 2 days ago Steve Borho encoder: abort on failure to open CSV log file for write
    • 2 days ago Steve Borho stats: introduce X265_LOG_FRAME for file level CSV logging without console logs
    • 34 hours ago Deepthi Nandakumar vps: frameOnlyConstraintFlag is true if fieldSeqFlag is false.
    • 34 hours ago Deepthi Nandakumar stats: nits
    • 2 days ago Steve Borho stats: document the new columms in per-frame CSV files
    • 2 days ago Steve Borho stats: report frame wall time spent waiting for decided frames
    • 2 days ago Steve Borho stats: report wall time of frame encoder with no active worker threads
    • 2 days ago Steve Borho stats: report wall time of wait for reference rows
    • 2 days ago Steve Borho stats: include loop filter processing and all overhead in worker wall time
    • 2 days ago Steve Borho stats: report times in milliseconds
    • 2 days ago Steve Borho stats: report row0wait and frame end overhead seperate from wall time
    • 2 days ago Steve Borho stats: count the number of times top dependencies block worker threads
    • 2 days ago Steve Borho stats: keep timestamps instead of elapsed times, to allow more flexibility
    • 2 days ago Steve Borho frameencoder: use uint32_t more consistently for rows and columns
    • 2 days ago Steve Borho stats: add frame statistic for average WPP benefit
    • 2 days ago Steve Borho stats: keep running count of number of active worker threads per frame encoder
    • 2 days ago Steve Borho threading: add ATOMIC_ADD
    • 4 days ago Steve Borho cli: add a (mostly harmless) missing initializer
Version 1.4.433
  • Release Date: Jan 28, 2015
    • 10 hours ago Steve Borho Merge
    • 12 hours ago amarcu5 Adds documentation for zones
    • 2 days ago Steve Borho profile: illuminate pre-lookahead tasks of downscale and AQ init
    • 2 days ago Steve Borho profile: re-enable frame encode tasks
    • 2 days ago Steve Borho profile: name the file read thread
    • 2 days ago Steve Borho profile: give an compile error if PPA and VTUNE are enabled
    • 5 days ago Steve Borho encoder: proper indentation for the zero-latency loop, no logic changes
    • 5 days ago Steve Borho encoder: if zero-latency, encode each picture in single call
    • 5 days ago Steve Borho encoder: white-space, comment nits
    • 5 days ago Gopu Govindaswamy analysis: allocate and initialize interData ref index
    • 5 days ago Steve Borho cli: allow the CLI to be bit-depth independent on non-Windows platforms
    • 5 days ago Steve Borho slice: signal sps_max_latency_increase_plus1 more accurately (refs #99)
    • 5 days ago Steve Borho level: make --tune zero-latency have zero-latency at the decoder (closes #99)
    • 5 days ago Steve Borho param: make --tune zero-latency actually have zero-latency at the encoder
Version 1.4.432
  • Release Date: Jan 28, 2015
    • 10 hours ago Steve Borho Merge
    • 12 hours ago amarcu5 Adds documentation for zones
    • 2 days ago Steve Borho profile: illuminate pre-lookahead tasks of downscale and AQ init
    • 2 days ago Steve Borho profile: re-enable frame encode tasks
    • 2 days ago Steve Borho profile: name the file read thread
    • 2 days ago Steve Borho profile: give an compile error if PPA and VTUNE are enabled
    • 5 days ago Steve Borho encoder: proper indentation for the zero-latency loop, no logic changes
    • 5 days ago Steve Borho encoder: if zero-latency, encode each picture in single call
    • 5 days ago Steve Borho encoder: white-space, comment nits
    • 5 days ago Gopu Govindaswamy analysis: allocate and initialize interData ref index
    • 5 days ago Steve Borho cli: allow the CLI to be bit-depth independent on non-Windows platforms
    • 5 days ago Steve Borho slice: signal sps_max_latency_increase_plus1 more accurately (refs #99)
    • 5 days ago Steve Borho level: make --tune zero-latency have zero-latency at the decoder (closes #99)
    • 5 days ago Steve Borho param: make --tune zero-latency actually have zero-latency at the encoder
Version 1.4.424
  • Release Date: Jan 23, 2015
    • 10 hours ago David T Yuen Added 10bit support to ssse3 dct16 and dct32 intrinsics
    • 4 weeks ago Yann.OA cmake: fix pkgconfig install destination
    • 32 hours ago Steve Borho asm: slight improvements to macro names
    • 32 hours ago Steve Borho asm: remove LUMA_SP_FILTERS, it was redundant with LUMA_FILTERS
    • 32 hours ago Steve Borho asm: split setupAssemblyPrimitives() between 8bpp and 16bpp versions
    • 18 hours ago Steve Borho rc: explicit type conversion to avoid compiler warning

Version 1.4.413
  • Release Date: Jan 21, 2015
    • 9 hours ago Steve Borho cmake: bump X265_BUILD for zones and strict-cbr changes
    • 9 hours ago Steve Borho asm: nits
    • 11 hours ago David T Yuen Added high bit support to sse3 intrinsics
    • 4 weeks ago Adam MarcusSupport for tweaking rate control using zones

Version 1.4.409
  • Release Date: Jan 21, 2015
    • 4 hours ago Steve Borho asm: cleanups
    • 4 hours ago Steve Borho pixelharness: cleanup
    • 4 hours ago Steve Borho asm: remove obsolete comment
    • 26 hours ago David T Yuen asm: idct16 intrinsic 28900->25000 improvement over previous intrinsic
    • 26 hours ago Min Chen asm: fix broken on weight_sp and weight_pp on 8bpp mode
    • 26 hours ago Min Chen avoid warning on variant correction in weight_sp_c()
    • 33 hours ago Min Chen asm: rewrite and fix bug in weight_sp_sse4 on HIGH_BIT_DEPTH mode
    • 33 hours ago Min Chen asm: rewrite and fix bug in weight_pp_sse4 on HIGH_BIT_DEPTH mode
    • 38 hours ago Divya Manivannan asm: psyCost_ss_64x64 in sse4: improve 501123c->159906c
    • 38 hours ago Divya Manivannan asm: psyCost_ss_32x32 in sse4: improve 136848c->39754c
    • 38 hours ago Divya Manivannan asm: psyCost_ss_16x16 in sse4: improve 31052c->9946c
    • 34 hours ago Deepthi Nandakumar x265: update copyright header
    • 39 hours ago Deepthi Nandakumar predict: disable conditional-expression-constant warnings
    • 2 days ago Steve Borho primitives: better document the data structures and their use
Version 1.4.397
  • Release Date: Jan 19, 2015
    • 22 hours ago Satoshi Nakagawa more use CUGeom
    • 20 hours ago Steve Borho asm: pickup missing primitives
    • 20 hours ago Steve Borho pixel: consistent naming of blockfill functions
    • 21 hours ago Steve Borho asm: remove _ from ALL_LUMA_TU_TYPED_S to allow more users
    • 21 hours ago Steve Borho primitives: move intra pred functions into CU for improved cache coherency
    • 22 hours ago Steve Borho asm: move template into x265 namespace
    • 22 hours ago Steve Borho asm: new ALL_LUMA_BLOCKS for all square blocks, TU or CU
    • 22 hours ago Steve Borho asm: cleanup AVC oriented macros
    • 23 hours ago Steve Borho asm: chroma all-block macros, remove large amounts of redundancy
    • 25 hours ago Steve Borho asm: introduce helper macros to unify function sets
    • 26 hours ago Steve Borho primitives: consistent naming of chroma macros, cleanup asm-primitives.cpp
    • 27 hours ago Steve Borho mbdstharness: use NUM_TR_SIZE, reintroduce speed tests for dst/idst4x4
    • 29 hours ago Steve Borho primitives: reorder header for better clarity
    • 3 days ago Steve Borho cli: move option tables and help functions into x265cli.h
    • 3 days ago Satoshi Nakagawa fix lossless
    • 3 days ago Praveen Tiwari all_angs_pred_8x8: optimize mov instruction
    • 3 days ago Steve Borho framefilter: use unsigned 32bit ints consistently to avoid gcc warnings
    • 4 days ago Min Chen intrapred: constant input for intra_pred
    • 4 days ago Dnyaneshwar G intrapred: nits
    • 4 days ago Steve Borho framefilter: fix PSNR calculations
    • 4 days ago Deepthi Nandakumar doc: add more details to strict-cbr
    • 4 days ago Steve Borho predict: minor cleanups
    • 4 days ago Steve Borho intrapred: nit
    • 4 days ago Steve Borho slicetype: simplify lowres intra logic
    • 4 days ago Steve Borho search: cleanup comments and streamline logic
    • 4 days ago Steve Borho search: fix GCC warning about set but unused variable
Version 1.4.395
  • Release Date: Jan 19, 2015
    • 22 hours ago Satoshi Nakagawa more use CUGeom
    • 20 hours ago Steve Borho asm: pickup missing primitives
    • 20 hours ago Steve Borho pixel: consistent naming of blockfill functions
    • 21 hours ago Steve Borho asm: remove _ from ALL_LUMA_TU_TYPED_S to allow more users
    • 21 hours ago Steve Borho primitives: move intra pred functions into CU for improved cache coherency
    • 22 hours ago Steve Borho asm: move template into x265 namespace
    • 22 hours ago Steve Borho asm: new ALL_LUMA_BLOCKS for all square blocks, TU or CU
    • 22 hours ago Steve Borho asm: cleanup AVC oriented macros
    • 23 hours ago Steve Borho asm: chroma all-block macros, remove large amounts of redundancy
    • 25 hours ago Steve Borho asm: introduce helper macros to unify function sets
    • 26 hours ago Steve Borho primitives: consistent naming of chroma macros, cleanup asm-primitives.cpp
    • 27 hours ago Steve Borho mbdstharness: use NUM_TR_SIZE, reintroduce speed tests for dst/idst4x4
    • 29 hours ago Steve Borho primitives: reorder header for better clarity
    • 3 days ago Steve Borho cli: move option tables and help functions into x265cli.h
    • 3 days ago Satoshi Nakagawa fix lossless
    • 3 days ago Praveen Tiwari all_angs_pred_8x8: optimize mov instruction
    • 3 days ago Steve Borho framefilter: use unsigned 32bit ints consistently to avoid gcc warnings
    • 4 days ago Min Chen intrapred: constant input for intra_pred
    • 4 days ago Dnyaneshwar G intrapred: nits
    • 4 days ago Steve Borho framefilter: fix PSNR calculations
    • 4 days ago Deepthi Nandakumar doc: add more details to strict-cbr
    • 4 days ago Steve Borho predict: minor cleanups
    • 4 days ago Steve Borho intrapred: nit
    • 4 days ago Steve Borho slicetype: simplify lowres intra logic
    • 4 days ago Steve Borho search: cleanup comments and streamline logic
    • 4 days ago Steve Borho search: fix GCC warning about set but unused variable
Version 1.4.382
  • Release Date: Jan 15, 2015
    • 94 minutes ago Aarthi Thirumalai rc: fix bug in using strict-cbr option
    • 91 minutes ago Steve Borho primitives: fix for 4:4:4 chroma copy_pp
    • 4 hours ago Steve Borho primitives: improve and update comments, reorder for clarity
    • 5 hours ago Steve Borho primitives: cleanup sse_pp aliasing
    • 5 hours ago Steve Borho asm: move HIGH_BIT_DEPTH aliases into setupAliasPrimitives()
    • 5 hours ago Steve Borho testbench: validate correctness of aliases, but not performance
    • 6 hours ago Steve Borho asm: use available copy and pixelcmp methods and aliases
    • 7 hours ago Steve Borho primitives: rename C primitive functions to match coding style
    • 8 hours ago Steve Borho loopfilter: place C primitives in anonymous namespace
    • 8 hours ago Steve Borho search: ignore MSVC warnings about constant conditional expressions
    • 8 hours ago Steve Borho primitives: new CHROMA CU enums for clarity and safety
    • 8 hours ago Steve Borho asm: use BLOCK_ enums when setting CU primitives for safety
    • 8 hours ago Steve Borho primitives: zero 4x4 sa8d pointer for chroma 4:4:4
    • 8 hours ago Steve Borho asm: remove invalid primitive assignment
    • 9 hours ago Steve Borho primitives: cleanup aliasing of p2s
    • 9 hours ago Steve Borho primitives: rename primitive functions to follow our naming style
    • 21 hours ago Steve Borho primitives: add chroma CU copy_pp, use new CU based methods where applicable
    • 21 hours ago Steve Borho primitives: create a CU copy_pp alias for square blocks, cleanup aliasing
    • 21 hours ago Steve Borho primitives: remove luma_ prefix from many primitives
    • 21 hours ago Steve Borho pixel: cleanup definitions for C primitives, fill missing chroma pointers
    • 21 hours ago Steve Borho asm: fix 16bpp build
    • 25 hours ago Steve Borho primitives: move sse_pp from PU to CU
    • 26 hours ago Steve Borho primitives: rename luma_addAvg to just addAvg
    • 26 hours ago Steve Borho primitives: remove unused funcdef
    • 26 hours ago Steve Borho primitives: move luma_copy_ss from PU to CU
    • 26 hours ago Steve Borho primitives: remove unused sse_sp primitives (no longer used by intra)
    • 27 hours ago Steve Borho primitives: remove obsolete comment
    • 27 hours ago Steve Borho primitives: fixups for 16bpp
    • 27 hours ago Steve Borho primitives: move sse_ss from luma PU to CU struct
    • 28 hours ago Steve Borho primitives: add comments for bidir functions to differentiate them
    • 28 hours ago Steve Borho search: use chroma sse_pp to avoid partition size lookup
    • 28 hours ago Steve Borho primitives: luma_copy_ps and luma_copy_sp are CU only, same for chroma
    • 29 hours ago Steve Borho primitives: remove sa8d_inter, unused by the encoder
    • 8 days ago Steve Borho primitives: add sa8d and sse_pp aliases for chroma square blocks
    • 4 hours ago Divya Manivannan asm: psyCost_ss_8x8 in sse4: improve 7652c->2515c
    • 2 hours ago Dnyaneshwar G intrapred: remove _new suffix
    • 30 hours ago Dnyaneshwar G intrapred: remove deprecated intrapred angular functions
    • 4 days ago Dnyaneshwar G 16bpp: updated asm code for intrapred_ang_4x4 mode 2 to 34
    • 4 days ago Aasaipriya intrapred: updated asm code for intra_ang_4x4 mode 2 to 34
    • 8 hours ago Dnyaneshwar G intrapred: remove deprecated all_angs_pred and planar_pred functions and unit test code
    • 6 days ago Dnyaneshwar G intrapred: updated asm code for all_ang_pred_32x32
    • 6 days ago Dnyaneshwar G intrapred: cleanup all_angs_pred_4x4, 8x8 and 16x16 older asm code
    • 6 days ago Dnyaneshwar G intrapred: updated asm code for all_ang_pred_16x16
    • 6 days ago Dnyaneshwar G intrapred: updated asm code for all_ang_pred_8x8
    • 7 days ago Praveen Tiwari all_angs_pred_new_4x4: asm code
    • 7 days ago Dnyaneshwar G 16bpp: cleanup intra_ang8x8, intra_ang16x16 and intra_ang32x32 older asm code
    • 7 hours ago Dnyaneshwar G 16bpp: updated asm and unit test code for intrapred_ang_8x8, 16x16 and 32x32 mode 2 to 34
    • 11 days ago Dnyaneshwar G intrapred: cleanup intra_ang8x8, intra_ang16x16 and intra_ang32x32 older asm code
    • 11 days ago Dnyaneshwar G intrapred: updated asm and unit test code for intra_ang_32x32 mode 2 to 34
    • 13 days ago Dnyaneshwar G intrapred: updated asm and unit test code for intra_ang_16x16 mode 2 to 34
    • 13 days ago Dnyaneshwar G intrapred: fix 32-bit build crash due to stack broken for intra_ang8x8
    • 30 hours ago Dnyaneshwar G intrapred: updated asm and unit test code for intra_ang_8x8 mode 2 to 34
    • 2 weeks ago Dnyaneshwar G intrapred: cleanup intra_pred_dc and intra_pred_planar older code
    • 3 weeks ago Dnyaneshwar G intrapred: updated HIGH_BIT_DEPTH asm and unit test code for intra_planar 4x4, 8x8, 16x16 and 32x32
    • 3 weeks ago Dnyaneshwar G intrapred: updated asm and unit test code for intra_planar 4x4, 8x8, 16x16 and 32x32
    • 3 weeks ago Dnyaneshwar G intra_pred_dc: fix 32-build error, invalid register used
    • 4 weeks ago Dnyaneshwar G intra_pred_dc: cleanup, remove unused asm code
    • 30 hours ago Dnyaneshwar G intra_pred_dc: cleanup, remove unused functions
    • 4 weeks ago Dnyaneshwar G high_bit_depth: intra_pred_dc_new updated asm and unit test code
    • 4 weeks ago Dnyaneshwar G intra_pred_dc_new: updated asm and unit test code
    • 30 hours ago Praveen Tiwari scale1D_128to64_new: cleanup
    • 30 hours ago Praveen Tiwari scale1D_128to64: cleanup depricated primitive
    • 31 hours ago Praveen Tiwari scale1D_128to64_new primitive: ASM code and unit test code
    • 22 hours ago Ashok Kumar Mishra intra refactoring: intra prediction with modified planar, angular prediction functions
    • 28 hours ago Ashok Kumar Mishra intra refactoring: removed storing left neighbouring pixels in 144 bytes stride - new
    • 23 hours ago Deepthi Nandakumar rc: remove unused variable
    • 27 hours ago Deepthi Nandakumar rc: For strict cbr short segment encodes, set lookahead depth equal to frame count.
    • 28 hours ago Deepthi Nandakumar x265: remove rateTolerance from cli/param interface
    • 27 hours ago Deepthi Nandakumar ratecontrol: reorder strictCbr condition checks
    • 6 days ago Aarthi Thirumalai rc: control bitrate overshoots during midframe encodes in strict-cbr
    • 3 weeks ago Aarthi Thirumalai rc: restrict low tolerance for bitrate fluctuations to strict-cbr option
    • 3 weeks ago Aarthi Thirumalai cli: remove tune cbr and introduce strict-cbr as a param option instead
    • 29 hours ago Steve Borho primitives: move luma_sub_ps, luma_add_ps from PU to CU
Version 1.4.376
  • Release Date: Jan 14, 2015
    • 94 minutes ago Aarthi Thirumalai rc: fix bug in using strict-cbr option
    • 91 minutes ago Steve Borho primitives: fix for 4:4:4 chroma copy_pp
    • 4 hours ago Steve Borho primitives: improve and update comments, reorder for clarity
    • 5 hours ago Steve Borho primitives: cleanup sse_pp aliasing
    • 5 hours ago Steve Borho asm: move HIGH_BIT_DEPTH aliases into setupAliasPrimitives()
    • 5 hours ago Steve Borho testbench: validate correctness of aliases, but not performance
    • 6 hours ago Steve Borho asm: use available copy and pixelcmp methods and aliases
    • 7 hours ago Steve Borho primitives: rename C primitive functions to match coding style
    • 8 hours ago Steve Borho loopfilter: place C primitives in anonymous namespace
    • 8 hours ago Steve Borho search: ignore MSVC warnings about constant conditional expressions
    • 8 hours ago Steve Borho primitives: new CHROMA CU enums for clarity and safety
    • 8 hours ago Steve Borho asm: use BLOCK_ enums when setting CU primitives for safety
    • 8 hours ago Steve Borho primitives: zero 4x4 sa8d pointer for chroma 4:4:4
    • 8 hours ago Steve Borho asm: remove invalid primitive assignment
    • 9 hours ago Steve Borho primitives: cleanup aliasing of p2s
    • 9 hours ago Steve Borho primitives: rename primitive functions to follow our naming style
    • 21 hours ago Steve Borho primitives: add chroma CU copy_pp, use new CU based methods where applicable
    • 21 hours ago Steve Borho primitives: create a CU copy_pp alias for square blocks, cleanup aliasing
    • 21 hours ago Steve Borho primitives: remove luma_ prefix from many primitives
    • 21 hours ago Steve Borho pixel: cleanup definitions for C primitives, fill missing chroma pointers
    • 21 hours ago Steve Borho asm: fix 16bpp build
    • 25 hours ago Steve Borho primitives: move sse_pp from PU to CU
    • 26 hours ago Steve Borho primitives: rename luma_addAvg to just addAvg
    • 26 hours ago Steve Borho primitives: remove unused funcdef
    • 26 hours ago Steve Borho primitives: move luma_copy_ss from PU to CU
    • 26 hours ago Steve Borho primitives: remove unused sse_sp primitives (no longer used by intra)
    • 27 hours ago Steve Borho primitives: remove obsolete comment
    • 27 hours ago Steve Borho primitives: fixups for 16bpp
    • 27 hours ago Steve Borho primitives: move sse_ss from luma PU to CU struct
    • 28 hours ago Steve Borho primitives: add comments for bidir functions to differentiate them
    • 28 hours ago Steve Borho search: use chroma sse_pp to avoid partition size lookup
    • 28 hours ago Steve Borho primitives: luma_copy_ps and luma_copy_sp are CU only, same for chroma
    • 29 hours ago Steve Borho primitives: remove sa8d_inter, unused by the encoder
    • 8 days ago Steve Borho primitives: add sa8d and sse_pp aliases for chroma square blocks
    • 4 hours ago Divya Manivannan asm: psyCost_ss_8x8 in sse4: improve 7652c->2515c
    • 2 hours ago Dnyaneshwar G intrapred: remove _new suffix
    • 30 hours ago Dnyaneshwar G intrapred: remove deprecated intrapred angular functions
    • 4 days ago Dnyaneshwar G 16bpp: updated asm code for intrapred_ang_4x4 mode 2 to 34
    • 4 days ago Aasaipriya intrapred: updated asm code for intra_ang_4x4 mode 2 to 34
    • 8 hours ago Dnyaneshwar G intrapred: remove deprecated all_angs_pred and planar_pred functions and unit test code
    • 6 days ago Dnyaneshwar G intrapred: updated asm code for all_ang_pred_32x32
    • 6 days ago Dnyaneshwar G intrapred: cleanup all_angs_pred_4x4, 8x8 and 16x16 older asm code
    • 6 days ago Dnyaneshwar G intrapred: updated asm code for all_ang_pred_16x16
    • 6 days ago Dnyaneshwar G intrapred: updated asm code for all_ang_pred_8x8
    • 7 days ago Praveen Tiwari all_angs_pred_new_4x4: asm code
    • 7 days ago Dnyaneshwar G 16bpp: cleanup intra_ang8x8, intra_ang16x16 and intra_ang32x32 older asm code
    • 7 hours ago Dnyaneshwar G 16bpp: updated asm and unit test code for intrapred_ang_8x8, 16x16 and 32x32 mode 2 to 34
    • 11 days ago Dnyaneshwar G intrapred: cleanup intra_ang8x8, intra_ang16x16 and intra_ang32x32 older asm code
    • 11 days ago Dnyaneshwar G intrapred: updated asm and unit test code for intra_ang_32x32 mode 2 to 34
    • 13 days ago Dnyaneshwar G intrapred: updated asm and unit test code for intra_ang_16x16 mode 2 to 34
    • 13 days ago Dnyaneshwar G intrapred: fix 32-bit build crash due to stack broken for intra_ang8x8
    • 30 hours ago Dnyaneshwar G intrapred: updated asm and unit test code for intra_ang_8x8 mode 2 to 34
    • 2 weeks ago Dnyaneshwar G intrapred: cleanup intra_pred_dc and intra_pred_planar older code
    • 3 weeks ago Dnyaneshwar G intrapred: updated HIGH_BIT_DEPTH asm and unit test code for intra_planar 4x4, 8x8, 16x16 and 32x32
    • 3 weeks ago Dnyaneshwar G intrapred: updated asm and unit test code for intra_planar 4x4, 8x8, 16x16 and 32x32
    • 3 weeks ago Dnyaneshwar G intra_pred_dc: fix 32-build error, invalid register used
    • 4 weeks ago Dnyaneshwar G intra_pred_dc: cleanup, remove unused asm code
    • 30 hours ago Dnyaneshwar G intra_pred_dc: cleanup, remove unused functions
    • 4 weeks ago Dnyaneshwar G high_bit_depth: intra_pred_dc_new updated asm and unit test code
    • 4 weeks ago Dnyaneshwar G intra_pred_dc_new: updated asm and unit test code
    • 30 hours ago Praveen Tiwari scale1D_128to64_new: cleanup
    • 30 hours ago Praveen Tiwari scale1D_128to64: cleanup depricated primitive
    • 31 hours ago Praveen Tiwari scale1D_128to64_new primitive: ASM code and unit test code
    • 22 hours ago Ashok Kumar Mishra intra refactoring: intra prediction with modified planar, angular prediction functions
    • 28 hours ago Ashok Kumar Mishra intra refactoring: removed storing left neighbouring pixels in 144 bytes stride - new
    • 23 hours ago Deepthi Nandakumar rc: remove unused variable
    • 27 hours ago Deepthi Nandakumar rc: For strict cbr short segment encodes, set lookahead depth equal to frame count.
    • 28 hours ago Deepthi Nandakumar x265: remove rateTolerance from cli/param interface
    • 27 hours ago Deepthi Nandakumar ratecontrol: reorder strictCbr condition checks
    • 6 days ago Aarthi Thirumalai rc: control bitrate overshoots during midframe encodes in strict-cbr
    • 3 weeks ago Aarthi Thirumalai rc: restrict low tolerance for bitrate fluctuations to strict-cbr option
    • 3 weeks ago Aarthi Thirumalai cli: remove tune cbr and introduce strict-cbr as a param option instead
    • 29 hours ago Steve Borho primitives: move luma_sub_ps, luma_add_ps from PU to CU
Version 1.4.296
  • Release Date: Jan 12, 2015
    • 6 hours ago Steve Borho Merge
    • 19 hours ago Gopu Govindaswamy analysis load/save: dump skip mode info for reuse
    • 2 weeks ago Gopu Govindaswamy analysis: simplify inter analysis structure to share more inter analysis data
    • 21 hours ago Deepthi Nandakumar intrapred: clarify angle/mode.
    • 2 days ago Steve Borho primitives: move extendPicBorder funcdef to common.h
    • 2 days ago Steve Borho primitives: white-space and comment cleanpus
    • 2 days ago Divya Manivannan fix bug in sa8d_8x8 for psyCost_ss
    • 3 days ago Divya Manivannan add testbench for psyCost_ss and asm for psyCost_ss_4x4: improve 1989c->515c
    • 3 days ago Divya Manivannan change data type in satd_4x4 for psyCost_ss
    • 3 days ago Kevin Wu test: Move dst/idst tests out of DctConf struct
    • 4 days ago Kevin Wu Refactor EncoderPrimitives under test.
    • 5 days ago Kevin Wu Fix index to dct primitive when using dst.
    • 3 days ago Kevin Wu Refactor EncoderPrimitives under encoder.
    • 3 days ago Kevin Wu Refactor EncoderPrimitives under common.
    • 3 days ago Steve Borho vtune: add comma to prevent string concatenation - fixes task profiling
Version 1.4.288
  • Release Date: Jan 12, 2015
    • 10 hours ago Steve Borho vtune: sprintf needs stdio.h
    • 10 hours ago Steve Borho prof: instrument file reads
    • 10 hours ago Steve Borho prof: expose vtune pause/resume
    • 10 hours ago Steve Borho vtune: according to their docs, there is no need to check for NULL domain
    • 11 hours ago Steve Borho prof: add support for thread names
    • 12 hours ago Steve Borho rc: partial backout of 53e722adccfb, bad qrefresh
    • 12 hours ago Steve Borho prof: cleanup profile integrations
    • 13 hours ago Steve Borho prof: add new scoped events for pmode and pme
    • 31 hours ago Divya Manivannan fix bug in satd_4x4 for psyCost_ss
    • 29 hours ago Satoshi Nakagawa slicetype: fix flush
    • 29 hours ago Steve Borho add support for VTune profiling with task indicators
Version 1.4.285
  • Release Date: Jan 10, 2015
    • 10 hours ago Steve Borho vtune: sprintf needs stdio.h
    • 10 hours ago Steve Borho prof: instrument file reads
    • 10 hours ago Steve Borho prof: expose vtune pause/resume
    • 10 hours ago Steve Borho vtune: according to their docs, there is no need to check for NULL domain
    • 11 hours ago Steve Borho prof: add support for thread names
    • 12 hours ago Steve Borho rc: partial backout of 53e722adccfb, bad qrefresh
    • 12 hours ago Steve Borho prof: cleanup profile integrations
    • 13 hours ago Steve Borho prof: add new scoped events for pmode and pme
    • 31 hours ago Divya Manivannan fix bug in satd_4x4 for psyCost_ss
    • 29 hours ago Satoshi Nakagawa slicetype: fix flush
    • 29 hours ago Steve Borho add support for VTune profiling with task indicators

Version 1.4.257
  • Release Date: Jan 8, 2015
    • 3 hours ago Nabajit Deka sao: merge saoCuOrgE3 asm with encoder along with sign asm code integration
    • 3 hours ago Nabajit Deka asm: saoCuOrgE3 asm code
    • 3 hours ago Nabajit Deka asm: saoCuOrgE1 asm code
    • 105 minutes ago Ashok Kumar Mishra Quant: modified rate cost calculation of last significant coefficient
    • 2 hours ago Steve Borho primitives: avoid alias chain, direcly alias base primitive
    • 2 hours ago Steve Borho asm: remove redundant alias, this is handled by Setup_Alias_Primitives()
    • 26 hours ago Ashok Kumar Mishra entropy: modified last coefficient position encoding in codeCoeffNxN()
    • 25 hours ago Praveen Tiwari saoCuOrgE2: asm code
    • 26 hours ago Steve Borho slicetype: allow queue to fill past full to prevent bottlenecks
    • 29 hours ago Steve Borho frameencoder: skip active/busy row checks if WPP is disabled
    • 29 hours ago Steve Borho encoder: disable WPP if not enough columns
    • 33 hours ago Satoshi Nakagawa fix weightCost() [CHANGES OUTPUT]
    • 46 hours ago Praveen Tiwari sao.cpp: fixed shadow warnings
    • 2 days ago Praveen Tiwari sao.cpp: fixed compiler warnings
    • 2 days ago Praveen Tiwari saoCuOrgB0: asm code
    • 2 days ago Gopu Govindaswamy encoder: initialize analysis data to null
    • 2 days ago Min Chen testbench: fix bug in generate weight input data
    • 2 days ago Divya Manivannan asm: psyCost_pp_64x64 in sse4: improve 417824c->56347c
    • 2 days ago Divya Manivannan asm: psyCost_pp_32x32 in sse4: improve 110849c->13373c
    • 4 days ago Praveen Tiwari calcSaoStatsCu, SAO_EO_1: sign asm code integration
    • 5 days ago Praveen Tiwari SAO_EO_2: sign asm code integration
    • 4 days ago Praveen Tiwari SAO_EO_1: sign asm code integration
    • 4 days ago Steve Borho asm: fix denoise assembly following int32->int16 coeff change, re-enable it
    • 13 days ago Praveen Tiwari _upBuff1: scale down fron int32_t to int8_t
    • 5 days ago Praveen Tiwari calSign: clarify constness
    • 5 days ago Steve Borho intrapred: nits
    • 5 days ago Steve Borho common: unify clip templates, no output changes
    • 7 days ago Divya Manivannan asm: psyCost_pp_16x16 in sse4: improve 27086c->3566c
    • 7 days ago Min Chen update restrict in weight_pp_c reference code
    • 7 days ago Divya Manivannan asm: fix error in psyCost_pp_8x8
    • 7 days ago Praveen Tiwari added calSign primitive, improved 2316.99 -> 233.63 (9.92x) over C code
    • 2 weeks ago Praveen Tiwari _upBuff1: scale down from int32_t to int8_t
    • 2 weeks ago Praveen Tiwari saoCuOrgE0 asm code, improved 500.43 -> 466.58
    • 8 days ago Divya Manivannan asm: psyCost_pp_8x8 for HIGH_BIT_DEPTH in sse4: improve 6995c->1070c
    • 8 days ago Divya Manivannan asm: psyCost_pp_8x8 in sse4: improve 6425c->928c
    • 9 days ago Divya Manivannan asm & testbench: psyCost_pp_4x4 in sse4: improve 2088c->337c
    • 8 days ago David T Yuen Added cmake support to pass along build flags to yasm.
    • 13 days ago Satoshi Nakagawa refine intra neighbors
    • 9 days ago Deepthi Nandakumar sei: m_lastBPSEI is overwritten each time
    • 9 days ago Deepthi Nandakumar nal: VPS startCodeprefix needs 4 bytes
    • 2 weeks ago Ashok Kumar Mishra entropy: inline codeTransformSkipFlags()
    • 2 weeks ago Satoshi Nakagawa rdcost: unify scaleChromaDist*()
    • 2 weeks ago Gopu Govindaswamy encoder: allocate memory for inter and intra analysis data based on slicetype
    • 2 weeks ago Gopu Govindaswamy analysis: remove redundant argument in compressIntraCU
Version 1.4.244
  • Release Date: Jan 5, 2015
Version 1.4.241
  • Release Date: Jan 4, 2015
Version 1.4.226
  • Release Date: Dec 30, 2014
    • 3 days ago Satoshi Nakagawa fix 4:4:4 rd<=1
    • 4 days ago David T Yuenasm: idct[8x8] sse2 12232 -> 3500 over c code 3550 -> 3500 over intrinsic
Version 1.4.222
  • Release Date: Dec 24, 2014
    • 3 days ago Satoshi Nakagawa fix 4:4:4 rd<=1
    • 4 days ago David T Yuenasm: idct[8x8] sse2 12232 -> 3500 over c code 3550 -> 3500 over intrinsic

Version 1.4.219
  • Release Date: Dec 21, 2014
    • 47 hours ago Steve Borho ppa: emit one event per CTU for more clarity, disable frame threads events
    • 2 days ago Steve Borho ppa: refine event names
    • 2 days ago Steve Borho ppa: simplify interfaces, enforce coding style
    • 2 days ago Steve Borho ppa: minimize code foot-print of profiling events
    • 2 days ago Steve Borho doc: improve documentation for --stats and multi-pass in general
    • 2 days ago Min Chen fix: output wrong WppEntryOffset when emulating start code at end of WPP row
    • 3 days ago Aasaipriya Chandran asm: chroma_hpp[16x16] for colorspace i420 in avx2 improve 1540c->969c
    • 3 days ago Aasaipriya Chandran asm: chroma_hpp[32x32] for colorspace i420 in avx2 improve 6189c->3537c
    • 6 days ago Steve Borho encoder: combine create() and init() functions
Version 1.4.217
  • Release Date: Dec 20, 2014
    • 47 hours ago Steve Borho ppa: emit one event per CTU for more clarity, disable frame threads events
    • 2 days ago Steve Borho ppa: refine event names
    • 2 days ago Steve Borho ppa: simplify interfaces, enforce coding style
    • 2 days ago Steve Borho ppa: minimize code foot-print of profiling events
    • 2 days ago Steve Borho doc: improve documentation for --stats and multi-pass in general
    • 2 days ago Min Chen fix: output wrong WppEntryOffset when emulating start code at end of WPP row
    • 3 days ago Aasaipriya Chandran asm: chroma_hpp[16x16] for colorspace i420 in avx2 improve 1540c->969c
    • 3 days ago Aasaipriya Chandran asm: chroma_hpp[32x32] for colorspace i420 in avx2 improve 6189c->3537c
    • 6 days ago Steve Borho encoder: combine create() and init() functions
Version 1.4.209
  • Release Date: Dec 15, 2014
    • 23 hours ago Steve Borho doc: describe what happens when psy-rd is too high for bitrate
    • 35 hours ago Divya Manivannan asm: chroma_vpp[32x32] for colorspace i420 in avx2: improve 3881c->1933c
    • 25 hours ago Steve Borho analysis: avoid redundant MC work
    • 25 hours ago Steve Borho analysis: fix chroma predictions for 2Nx2N bidir at zero mv
    • 26 hours ago Steve Borho api: add some blank lines
    • 2 days ago Divya Manivannan asm: chroma_vpp[4x4] for colorspace i422 in avx2: improve 228c->184c
    • 39 hours ago Steve Borho reference: weight chroma planes of reference pictures if using chroma satd
    • 2 days ago Steve Borho motion: chroma ME [CHANGES OUTPUTS]
    • 2 days ago Steve Borho primitives: use NULL chroma satd func pointers for blocks not capable of satd
    • 2 days ago Steve Borho primitives: use luma satd functions for chroma, where applicable
    • 2 days ago Steve Borho primitives: add a chroma satd table that is indexed by luma partition
    • 2 days ago Steve Borho reference: move reconPic pointer to base class so it is available to ME
    • 3 days ago Steve Borho motion: sync argument names between the header and the cpp file
    • 3 days ago Steve Borho yuv: fix size check in copyFromYuv
    • 2 days ago Steve Borho search: rename index variable to puIdx for consistency
    • 3 days ago Steve Borho motion: add a version of setSourcePU which can accept fenc from another Yuv
    • 3 days ago Steve Borho motion: use Yuv instance to hold fenc PU pixels (preparing for chroma ME)
    • 3 days ago Steve Borho slicetype: cleanups - use bufSATD method where applicable
    • 3 days ago Steve Borho yuv: plumb in support for mono-chrome YUV buffers
    • 2 days ago Aasaipriya Chandran asm: chroma_hpp[8x8] for colorspace i420 in avx2 improve 530c->373c
    • 2 days ago Steve Borho asm: fix x86 link errors
    • 2 days ago Deepthi Nandakumar asm: chroma_vpp[16x16] for colorspace i420 in avx2: improve 998c->978c
    • 6 days ago Divya Manivannan asm: chroma_vpp[8x8] for colorspace i420 in avx2: improve 338c->269c
    • 5 days ago Satoshi Nakagawa refine tuDepth related
Version 1.4.208
  • Release Date: Dec 14, 2014
    • 23 hours ago Steve Borho doc: describe what happens when psy-rd is too high for bitrate
    • 35 hours ago Divya Manivannan asm: chroma_vpp[32x32] for colorspace i420 in avx2: improve 3881c->1933c
    • 25 hours ago Steve Borho analysis: avoid redundant MC work
    • 25 hours ago Steve Borho analysis: fix chroma predictions for 2Nx2N bidir at zero mv
    • 26 hours ago Steve Borho api: add some blank lines
    • 2 days ago Divya Manivannan asm: chroma_vpp[4x4] for colorspace i422 in avx2: improve 228c->184c
    • 39 hours ago Steve Borho reference: weight chroma planes of reference pictures if using chroma satd
    • 2 days ago Steve Borho motion: chroma ME [CHANGES OUTPUTS]
    • 2 days ago Steve Borho primitives: use NULL chroma satd func pointers for blocks not capable of satd
    • 2 days ago Steve Borho primitives: use luma satd functions for chroma, where applicable
    • 2 days ago Steve Borho primitives: add a chroma satd table that is indexed by luma partition
    • 2 days ago Steve Borho reference: move reconPic pointer to base class so it is available to ME
    • 3 days ago Steve Borho motion: sync argument names between the header and the cpp file
    • 3 days ago Steve Borho yuv: fix size check in copyFromYuv
    • 2 days ago Steve Borho search: rename index variable to puIdx for consistency
    • 3 days ago Steve Borho motion: add a version of setSourcePU which can accept fenc from another Yuv
    • 3 days ago Steve Borho motion: use Yuv instance to hold fenc PU pixels (preparing for chroma ME)
    • 3 days ago Steve Borho slicetype: cleanups - use bufSATD method where applicable
    • 3 days ago Steve Borho yuv: plumb in support for mono-chrome YUV buffers
    • 2 days ago Aasaipriya Chandran asm: chroma_hpp[8x8] for colorspace i420 in avx2 improve 530c->373c
    • 2 days ago Steve Borho asm: fix x86 link errors
    • 2 days ago Deepthi Nandakumar asm: chroma_vpp[16x16] for colorspace i420 in avx2: improve 998c->978c
    • 6 days ago Divya Manivannan asm: chroma_vpp[8x8] for colorspace i420 in avx2: improve 338c->269c
    • 5 days ago Satoshi Nakagawa refine tuDepth related
Version 1.4.200
  • Release Date: Dec 12, 2014
    • 23 hours ago Steve Borho doc: describe what happens when psy-rd is too high for bitrate
    • 35 hours ago Divya Manivannan asm: chroma_vpp[32x32] for colorspace i420 in avx2: improve 3881c->1933c
    • 25 hours ago Steve Borho analysis: avoid redundant MC work
    • 25 hours ago Steve Borho analysis: fix chroma predictions for 2Nx2N bidir at zero mv
    • 26 hours ago Steve Borho api: add some blank lines
    • 2 days ago Divya Manivannan asm: chroma_vpp[4x4] for colorspace i422 in avx2: improve 228c->184c
    • 39 hours ago Steve Borho reference: weight chroma planes of reference pictures if using chroma satd
    • 2 days ago Steve Borho motion: chroma ME [CHANGES OUTPUTS]
    • 2 days ago Steve Borho primitives: use NULL chroma satd func pointers for blocks not capable of satd
    • 2 days ago Steve Borho primitives: use luma satd functions for chroma, where applicable
    • 2 days ago Steve Borho primitives: add a chroma satd table that is indexed by luma partition
    • 2 days ago Steve Borho reference: move reconPic pointer to base class so it is available to ME
    • 3 days ago Steve Borho motion: sync argument names between the header and the cpp file
    • 3 days ago Steve Borho yuv: fix size check in copyFromYuv
    • 2 days ago Steve Borho search: rename index variable to puIdx for consistency
    • 3 days ago Steve Borho motion: add a version of setSourcePU which can accept fenc from another Yuv
    • 3 days ago Steve Borho motion: use Yuv instance to hold fenc PU pixels (preparing for chroma ME)
    • 3 days ago Steve Borho slicetype: cleanups - use bufSATD method where applicable
    • 3 days ago Steve Borho yuv: plumb in support for mono-chrome YUV buffers
    • 2 days ago Aasaipriya Chandran asm: chroma_hpp[8x8] for colorspace i420 in avx2 improve 530c->373c
    • 2 days ago Steve Borho asm: fix x86 link errors
    • 2 days ago Deepthi Nandakumar asm: chroma_vpp[16x16] for colorspace i420 in avx2: improve 998c->978c
    • 6 days ago Divya Manivannan asm: chroma_vpp[8x8] for colorspace i420 in avx2: improve 338c->269c
    • 5 days ago Satoshi Nakagawa refine tuDepth related
Version 1.4.179
  • Release Date: Dec 10, 2014
    • 26 hours ago Steve Borho cmake: do not use a cache string for version found in hg_archive.txt (refs #84)
    • 2 days ago Aarthi Thirumalai rc : fix bug in deciding qp for first frame in CRF
    • 2 days ago Aarthi Thirumalai rc: fix chroma qp and chroma lambda derivations.
    • 37 hours ago Deepthi Nandakumar analysis: comments
    • 39 hours ago Satoshi Nakagawa fix chroma distortion for 4:2:2
    • 39 hours ago Steve Borho cmake: disable idiotic uninitialized local variable warnings from VC
Version 1.4.174
  • Release Date: Dec 7, 2014
    • 26 hours ago Steve Borho cmake: do not use a cache string for version found in hg_archive.txt (refs #84)
    • 2 days ago Aarthi Thirumalai rc : fix bug in deciding qp for first frame in CRF
    • 2 days ago Aarthi Thirumalai rc: fix chroma qp and chroma lambda derivations.
    • 37 hours ago Deepthi Nandakumar analysis: comments
    • 39 hours ago Satoshi Nakagawa fix chroma distortion for 4:2:2
    • 39 hours ago Steve Borho cmake: disable idiotic uninitialized local variable warnings from VC

Version 1.4.164
  • Release Date: Dec 5, 2014
    • 26 hours ago Divya Manivannan asm: chroma_vpp[4x4] for colorspace i420 in avx2: improve 228c->184c
    • 15 hours ago Steve Borho analysis: cache m_bChromaSa8d and reduce redundant work
    • 21 hours ago Deepthi Nandakumar analysis: add chroma distortion to rdLevels 3 and 4
    • 27 hours ago Deepthi Nandakumar noiseReduction: allow separate strengths to be specified for intra and inter CUs
    • 25 hours ago Aarthi Thirumalai primitives: fix build error in refactor of chroma p2s primitive.
    • 27 hours ago Steve Borho primitives: cleanup EncoderPrimitives, refactor chroma p2s primitive
    • 28 hours ago Steve Borho primitives: remove unused chroma lowres primitive
    • 28 hours ago Steve Borho search: avoid AMVP selection if both MVs are the same
    • 30 hours ago Steve Borho cmake: remove buggy workarounds for partial SIMD support (fixes #92)
    • 38 hours ago Steve Borho doc: fix typo (closes #83)
    • 38 hours ago Steve Borho param: allow NR values from 1..99, clarify docs (closes #87)
    • 38 hours ago Steve Borho doc: add support for reST generated man-pages (closes #89)
    • 39 hours ago Steve Borho constants: adjust lambda tabels for 10bit encodes (fixes #55)
    • 2 days ago Gopu Govindaswamy encoder: fix binary mismatch for analysis load vs save with same bitrate
Version 1.4.154
  • Release Date: Dec 3, 2014
    • 13 hours ago Steve Borho Merged
    • 25 hours ago Deepthi Nandakumar x265: add ratetol to command line help
    • 44 hours ago Deepthi Nandakumar encoder: free csv file name
Version 1.4.151
  • Release Date: Dec 2, 2014
    • 2 days ago Divya Manivannan asm: luma_vpp[8x16, 8x32] in avx2: improve 1139c->774c, 1968c->1452c
    • 2 days ago Divya Manivannan asm: luma_vpp[12x16] in avx2: improve 1977c->1418c
    • 2 days ago Divya Manivannan asm: luma_vpp[24x32] in avx2: improve 5637c->3695c
    • 2 days ago Divya Manivannan asm: luma_vpp[48x64] in avx2: improve 21298c->14696c
    • 2 days ago Satoshi Nakagawa primitives: refactor tskip related
    • 26 hours ago Satoshi Nakagawa nits
    • 40 hours ago Deepthi Nandakumar param: disable b-intra in B frames when tune grain is true.
    • 4 days ago Gopu Govindaswamy encoder: make all member fields public
Version 1.4.146
  • Release Date: Nov 30, 2014
    • 2 days ago Divya Manivannan asm: luma_vpp[8x16, 8x32] in avx2: improve 1139c->774c, 1968c->1452c
    • 2 days ago Divya Manivannan asm: luma_vpp[12x16] in avx2: improve 1977c->1418c
    • 2 days ago Divya Manivannan asm: luma_vpp[24x32] in avx2: improve 5637c->3695c
    • 2 days ago Divya Manivannan asm: luma_vpp[48x64] in avx2: improve 21298c->14696c
    • 2 days ago Satoshi Nakagawa primitives: refactor tskip related
    • 26 hours ago Satoshi Nakagawa nits
    • 40 hours ago Deepthi Nandakumar param: disable b-intra in B frames when tune grain is true.
    • 4 days ago Gopu Govindaswamy encoder: make all member fields public

Version 1.4.144
  • Release Date: Nov 29, 2014
    • 39 hours ago Steve Borho doc: restructure documentation with better grouping, improve cross-refs
    • 39 hours ago Steve Borho doc: fix a sphinx build warning
    • 42 hours ago Steve Borho doc: improve readability of film grain section
    • 42 hours ago Steve Borho doc: add cbr to the list of tunings, add helpful cross-refs
    • 43 hours ago Aarthi Thirumalai rc: introduce cli option to tune for cbr.
    • 3 days ago Aarthi Thirumalai rc: improve the frame size planning with respect to vbv buffer occupancy and the lookahead window.
    • 3 days ago Aarthi Thirumalai rc: adjust qp for B frames from ABR feedback in case of CBR.
    • 3 days ago Aarthi Thirumalai rc: limit bit amortization in ABR to longer sequences
    • 43 hours ago Steve Borho rc: use c-style typecasts
    • 9 days ago Aarthi Thirumalai rc: tune midframe vbv logic for B frames
    • 7 days ago Aarthi Thirumalai slicetype: fix vbv lookahead data collection for all frames within the lookahead window.
    • 2 days ago Divya Manivannan asm: luma_hpp[8x8, 8x16, 8x32] in avx2: improve 623c->523c, 1384c->1083c, 2555c->2058c
    • 2 days ago Aasaipriya Chandran Luma_hpp[48x64] avx2 asm code : improved 25053c->17882c
    • 2 days ago Divya Manivannan asm: luma_hpp[8x4] in avx2: improve 357c->261c
    • 2 days ago Aasaipriya Chandran Luma_hpp[32x8 , 32x16 , 32x24 , 32x32 , 32x64] avx2 asm code: improved 2032c->1556c, 4238c->3014c, 6696c->4801c, 8697c->6433c, 16823c->12297c
    • 2 days ago Divya Manivannan asm: luma_vpp[64x16] in avx2: improve 7245c->4910c
    • 2 days ago Divya Manivannan asm: luma_vpp[64x32, 64x48, 64x64] in avx2: improve 14150c->9810c, 21132c->14684c, 28663c->19616c
Version 1.4.125
  • Release Date: Nov 27, 2014
    • 32 hours ago Divya Manivannan asm: luma_vpp[32x8] in avx2: improve 2047c->1472c
    • 32 hours ago Divya Manivannan asm: luma_vpp[32x24] in avx2: improve 5562c->3899c
    • 32 hours ago Divya Manivannan asm: luma_vpp[32x16] in avx2: improve 3808c->2491c
    • 33 hours ago Divya Manivannan asm: luma_vpp[32x32, 32x64] in avx2: improve 7247c->4909c, 14365c->9774c
    • 33 hours ago Divya Manivannan asm: luma_vpp[16x32, 16x64] in avx2: improve 3875c->2463c, 7499c->4894c
    • 30 hours ago Aasaipriya Chandran asm: avx2 for Luma_hpp[16x4, 16x8, 16x12, 16x16 , 16x32, 16x64]
    • 32 hours ago Aasaipriya Chandran asm: avx2 for luma_hpp[64x64, 64x48, 64x32, 64x16]
    • 3 days ago Steve Borho frameencoder: do not use bitmaps for framefilter if not WPP
    • 41 hours ago Steve Borho frameencoder: release row lock while waiting during VBV restarts
    • 5 days ago Divya Manivannan asm: luma_vpp[16x4] in avx2: improve 734c->497c
    • 5 days ago Divya Manivannan asm: luma_vpp[16x8] in avx2: improve 1195c->745c
    • 5 days ago Divya Manivannan asm: luma_vpp[16x12] in avx2: improve 1644c->1018c

Version 1.4.116
  • Release Date: Nov 24, 2014
    • 2 days ago Praveen Tiwari idct32_c: C code optimization
    • 2 days ago Praveen Tiwari idct16_c: optimization
    • 2 days ago Praveen Tiwari idct8_c: optimization
    • 2 days ago Praveen Tiwari idct4_c: optimization
    • 2 days ago Praveen Tiwari dct32_c: optimization
    • 2 days ago Praveen Tiwari dct16_c: optimization
    • 2 days ago Praveen Tiwari dct8_c: optimization
    • 2 days ago Praveen Tiwari dct4_c: C code optimization
    • 2 days ago Praveen Tiwari idst4_c: optimization
    • 2 days ago Praveen Tiwari dst4_c: optimization
    • 2 days ago Satoshi Nakagawa fix copy16to16_shl
    • 2 days ago Steve Borho analysis: explicit locking for pmode and pme parameters
    • 2 days ago David T Yuen Updated intrinsic of idct8 sse3 for new input format
    • 3 days ago Divya Manivannan asm: luma_vpp[16x16] in avx2: improve 2141c->1284c
    • 3 days ago Divya Manivannan asm: luma_vpp[8x4] in avx2: improve 498c->257c
    • 3 days ago Divya Manivannan asm: luma_vpp[8x8] in avx2: improve 701c->387c
    • 2 days ago Steve Borho encoder: nits and alloc zero fix
    • 3 days ago Praveen Tiwari Fix for C code mismatch
    • 3 days ago Satoshi Nakagawa fix for old gcc
    • 3 days ago Deepthi Nandakumar build: remove icl32 and icl64 scripts
Version 1.4.95
  • Release Date: Nov 21, 2014
    • 8 hours ago Praveen Tiwari luma_hpp[4x4]: AVX2 asm code bug fix
    • 8 hours ago Gopu Govindaswamy encoder: fix analysis file read
    • 9 hours ago Satoshi Nakagawa fix for rd=0
    • 12 hours ago Satoshi Nakagawa replace char to int8_t, where it should be signed char
    • 25 hours ago Praveen Tiwari disable denoiseDct asm code until fixed for Mac OS
    • 4 days ago Satoshi Nakagawa primitives: clarify constness
    • 44 hours ago Steve Borho dct: fix gcc warnings
    • 2 days ago Praveen Tiwari refactorizaton of the transform/quant path.
    • 32 hours ago Satoshi Nakagawa fseeko for mingw32
    • 31 hours ago Steve Borho threadind: fixes for VC11 Win32 includes, prune two unused functions
    • 35 hours ago Steve Borho wavefront: fix msvc warning
    • 36 hours ago Steve Borho threading: use 32bit atomic integer operations exclusively
    • 37 hours ago Steve Borho constants: remove init/destroyROM functions
Version 1.4.80
  • Release Date: Nov 20, 2014
    • 22 hours ago Steve Borho api: fix range limit docs for RQT limit params
    • 23 hours ago Steve Borho frameencoder: white-space nits
    • 23 hours ago Steve Borho analysis: drop MATCH_NON_PMODE macro
    • 23 hours ago Steve Borho threading: don't use this->
    • 23 hours ago Steve Borho threading: copyright comment format nits
    • 23 hours ago Steve Borho param: use strdup() on input strings uniformly
    • 37 hours ago Deepthi Nandakumar encoder: init filename to NULL
    • 2 days ago Gopu Govindaswamy search: fix binary mismatch and inconsistent crash for share inter information
Version 1.4.74
  • Release Date: Nov 19, 2014
    • 30 hours ago Gopu Govindaswamy encoder: force slicetype using analysis file
    • 30 hours ago Satoshi Nakagawa modify MV default constructor to do nothing
    • 45 hours ago Aarthi Thirumalai vbv: tune vbv predictors for better mapping of predicted bits to encoded bits
    • 47 hours ago Deepthi Nandakumar analysis: cleanups, init pointers, variable names are made self-explanatory
    • 6 days ago Gopu Govindaswamy analysis: fix binary mismatch for share intra save and load mode with same cli
Version 1.4.67
  • Release Date: Nov 16, 2014
    • 23 hours ago Steve Borho cli: fix analysis filename argument
    • 27 hours ago Steve Borho encoder: add prefix to FREAD and FWRITE macros to avoid MacOSX macro conflict
    • 27 hours ago Steve Borho common: move analysis reuse structs to common.h
    • 37 hours ago Satoshi Nakagawa analysis: encodeResidue() directly write to reconPic
    • 33 hours ago Deepthi Nandakumar analysis save/load: refactor full implementation
    • 2 days ago Satoshi Nakagawa analysis: don't add the cost of picture boundary CU to avgCost [CHANGES OUTPUT]
    • 46 hours ago Steve Borho cmake: hack to avoid escaping problems in cmake 3.1 parser
    • 2 days ago Satoshi Nakagawa nits
Version 1.4.59
  • Release Date: Nov 14, 2014
  • 8 hours ago Steve Borho rdcost: lower the psy-rd scale factor for I slices to 96/256
    45 hours ago Satoshi Nakagawa refine initializeGeoms()
    36 hours ago Steve Borho analysis: fix bidir non-determinism in --pmode --rd 5
    37 hours ago Steve Borho Merge
    46 hours ago Deepthi Nandakumar x265: more meaningful error messages in analysis
    47 hours ago Deepthi Nandakumar api: cleanup
    47 hours ago Deepthi Nandakumar api: replace analysis data with pre defined constant
    47 hours ago Deepthi Nandakumar x265: create and initialise recon object if analysis mode is enabled
    47 hours ago Deepthi Nandakumar param: add default value to analysis mode
    2 days ago Gopu Govindaswamy x265: remove redundant variables from intra and inter analysis structure
    2 days ago Gopu Govindaswamy analysis: Dump best MV statistics and re-use this for analysis load mode
    2 days ago Steve Borho quant: allow --nr in all slice types evenly
    6 days ago Deepthi Nandakumar noiseReduction: apply only for I and P, move NoiseReduction to quant.h
    2 days ago Steve Borho param: raise --nr limit to 2000
    2 days ago Steve Borho param: remove --b-intra from --tune grain, document rdoq restriction
    3 days ago Steve Borho rdcost: experimental slice-type based psy-rd scale factor
    4 days ago Steve Borho analysis: RDO based BIDIR decisions
Version 1.4.55
  • Release Date: Nov 12, 2014
    • 7 hours ago Satoshi Nakagawa cleanup SIZE_NONE. empty CU has MODE_NONE.
    • 38 hours ago Steve Borho search: fixup
    • 2 days ago Steve Borho reference: add methods for querying CU/PU pointers
    • 2 days ago Steve Borho analysis: delay initialization of prediction cu in RD 5 and 6
    • 43 hours ago Satoshi Nakagawa fix typo
    • 2 days ago Steve Borho analysis: delay initialization of prediction cu until just before use
    • 2 days ago Steve Borho search: keep AMVP candidates in mode structure
    • 2 days ago Steve Borho motion: remove trivial set methods; make some members public
    • 2 days ago Steve Borho nr: fix denoise offset memcopy size
    • 3 days ago Steve Borho entropy: pass context model (state) to bitsCodeBin as uint32_t
Version 1.4.47
  • Release Date: Nov 11, 2014
    • 38 hours ago Steve Borho search: fixup
    • 2 days ago Steve Borho reference: add methods for querying CU/PU pointers
    • 2 days ago Steve Borho analysis: delay initialization of prediction cu in RD 5 and 6
    • 43 hours ago Satoshi Nakagawa fix typo
    • 2 days ago Steve Borho analysis: delay initialization of prediction cu until just before use
    • 2 days ago Steve Borho search: keep AMVP candidates in mode structure
    • 2 days ago Steve Borho motion: remove trivial set methods; make some members public
    • 2 days ago Steve Borho nr: fix denoise offset memcopy size
    • 3 days ago Steve Borho entropy: pass context model (state) to bitsCodeBin as uint32_t

Version 1.4.31
  • Release Date: Nov 9, 2014
    • 13 hours ago Steve Borho entropy: nit
    • 13 hours ago Steve Borho entropy: ensure X265_CHECK() has braces
    • 13 hours ago Steve Borho entropy: inline methods which mapped to encodeBin() calls
    • 14 hours ago Steve Borho entropy: inline bit counting functions
    • 14 hours ago Steve Borho entropy: use bitsCodeBin in intra mode bit estimate functions
    • 14 hours ago Steve Borho entropy: rename encodeBinContext to bitsCodeBin, make const
    • 14 hours ago Steve Borho entropy: white-space nits
    • 21 hours ago Satoshi Nakagawa fix bug in 522baf03fbbd
    • 20 hours ago Deepthi Nandakumar search: fix warnings
    • 23 hours ago Satoshi Nakagawa fix typo
    • 2 days ago Ashok Kumar Mishra [REVIEW PATCH/OUTPUT CHANGED]search: removed multiple encode Coefficients from estimateResidualQT()
    • 2 days ago Ashok Kumar Mishra search: made a function for null cost calculation in xEstimateResidualQT()
    • 2 days ago Ashok Kumar Mishra search: made separate functions for encoding cbfs in xEstimateResidualQT()
    • 29 hours ago Satoshi Nakagawa cudata: remove default argument
    • 30 hours ago Steve Borho doc: fix sub-title depth
    • 35 hours ago Steve Borho param: add --tune grain
    • 31 hours ago Steve Borho search: ugly bias hack for bidir with psy-rd
    • 32 hours ago Steve Borho docs: document RC params, at least minimally
    • 32 hours ago Steve Borho api: cleanup comments
    • 32 hours ago Steve Borho cli: cleanup CLI help, add 'verbose' tier
    • 32 hours ago Steve Borho api: expose rate control params via x265_param_parse() and CLI
    • 2 days ago Steve Borho deblock: fix type conversion warnings
    • 2 days ago Satoshi Nakagawa refine deblocking filter
    • 4 days ago Gopu Govindaswamy cli: bug fix for validatefanout param analysis-mode=save and load
    • 4 days ago gopi jayaraman encoder: use 6 frameNumThreads for cpucount 32 and above
Version 1.3.794
  • Release Date: Oct 28, 2014
    • 2 hours ago Steve Borho search: leave a helpful comment to avoid unfortunate reordering
    • 2 hours ago Steve Borho predict: cache color space dimension shifts, use in search.cpp
    • 2 hours ago Steve Borho analysis: in RD 5/6 ignore redundant merge candidates
    • 3 hours ago Steve Borho analysis: use same MV range check in merge functions
    • 3 hours ago Steve Borho analysis: defer broadcast sets of merge mv field data until final selection
    • 3 hours ago Steve Borho search: use member variable from Predict for csp
    • 4 hours ago Steve Borho entropy: nit
    • 4 hours ago Steve Borho search: simplify handling of TU size edge case in extractIntraResultChromaQT
    • 4 hours ago Steve Borho search: fixes for chroma tskip coding with placebo preset
    • 6 hours ago Steve Borho search: trMode -> tuDepth, improve clarity
    • 6 hours ago Steve Borho cudata: remove unused setDepthSubParts
    • 6 hours ago Steve Borho cudata: setTrIdxSubParts() -> setTUDepthSubParts()
    • 6 hours ago Steve Borho cudata: renames (m_depth, m_trIdx) -> (m_cuDepth, m_tuDepth)
    • 7 hours ago Steve Borho docs: update --tu-intra-depth / --tu-inter-depth docs (refs #71)
    • 7 hours ago Steve Borho entropy: improve check strings and comments
    • 7 hours ago Steve Borho search: further --rd-penalty 2 fixes (refs #71)
    • 8 hours ago Steve Borho analysis: invalidate slave entropy contexts, to help find bugs
    • 8 hours ago Steve Borho analysis: checkIntraInInter_rd0_4() needs the entropy context to be initialized
    • 41 hours ago Steve Borho analysis: further work on --pmode for --rd 5/6
    • 3 days ago gopi jayaraman analysis: adding support for --rd 5/6 in compressInterCU_dist() and parallelModeAnalysis()
    • 23 hours ago Steve Borho entropy: readability nits
    • 23 hours ago Steve Borho common: add support for debug breaks in Xcode
    • 30 hours ago Steve Borho param: fix w-s spacing in log messages for --rdpenalty and chroma offsets
    • 10 hours ago Deepthi Nandakumar analysis: motionCompensation(chroma) needs to be performed for asymmetric merge also
    • 11 hours ago Deepthi Nandakumar motion: this emms is required
    • 13 hours ago Deepthi Nandakumar analysis: cleanup
    • 17 hours ago Satoshi Nakagawa cugeom: fix uninitialized reported by valgrind

Version 1.3.787
  • Release Date: Oct 27, 2014
    • 15 hours ago Steve Borho cudata: validate subCU size against current analysis depth
    • 15 hours ago Steve Borho cudata: make sure per-part buffers are allocate for g_maxCUSize
    • 18 hours ago Steve Borho search: improve comments in mergeEstimation()
    • 18 hours ago Steve Borho search: turn some redundant clears of tskip flags into runtime checks
    • 18 hours ago Steve Borho search: cleanup residualQTIntraChroma
    • 20 hours ago Steve Borho docs: update --tskip and --cu-lossless docs
    • 21 hours ago Steve Borho encoder: issue warning and disable --pmode if rdlevel < 2
    • 21 hours ago Steve Borho search: remove resiYuv from Mode, keep tmpResiYuv in m_rqt[]
    • 21 hours ago Steve Borho search: updateCandList() can be a static method
    • 21 hours ago Steve Borho search: inline updateModeCost
    • 21 hours ago Steve Borho encoder: issue warnings and explicitly disable tskip or culossless if rd < 3
    • 21 hours ago Steve Borho search: prevent warnings about unused bCheckSplit value
    • 21 hours ago Steve Borho primitives: remove unused calcrecon primitive (assembly needs cleanup)
    • 22 hours ago Steve Borho trim x265_emms(), try to only use prior to floating point operations
    • 22 hours ago Steve Borho search: avoid a context save at the last recursion depth
    • 24 hours ago Steve Borho search: improve comments and readability of residualTransformQuantIntra
    • 24 hours ago Steve Borho search: keep recon QT in pixels, instead of shorts
    • 25 hours ago Steve Borho search: remove tskip analysis out of luma chroma normal path
    • 26 hours ago Steve Borho search: nit. splitted is not a word
    • 26 hours ago Steve Borho search: rename methods that read coeff and recon from RQT struct at final depths
    • 26 hours ago Steve Borho search: give offsetSubTUCBFs a basic comment
    • 39 hours ago Steve Borho search: simplify RDO chroma intra coding, changes tskip outputs
    • 27 hours ago Steve Borho search: simplify initTrDepth
    • 27 hours ago Steve Borho search: reconYuv as ref
    • 27 hours ago Steve Borho search: improve a variable name
    • 27 hours ago Steve Borho search: rename a couple chroma intra helper methods
    • 2 days ago Steve Borho search: rename tmpCoeff to coeffRQT, tmpShortYuv to reconQtYuv / resiQtYuv
    • 46 hours ago Steve Borho search: fix 4:2:2 chroma tskip bit-cost estimation
    • 2 days ago Steve Borho yuv: add copyPartToPart* methods for recon RQT finalization
    • 2 days ago Steve Borho shortyuv: use absPartIdx for CU/TU part offset like everywhere else
    • 2 days ago Praveen Tiwari weight_sp: pshufd to handle width 6 for SSE version of asm code
    • 2 days ago Ashok Kumar Mishra search: refactored xEstimateResidualQT() to remove cbf flag settings
    • 2 days ago Ashok Kumar Mishra search: remove redundant cbf flags setting in xEstimateResidualQT()
    • 2 days ago Praveen Tiwari quant.cpp: nits
    • 2 days ago Ashok Kumar Mishra search: remove unnecessary set of cbf flags in xEstimateResidualQT()
    • 2 days ago Steve Borho cmake: add -Wno-array-bounds if the compiler supports it
    • 3 days ago Ashok Kumar Mishra [OUTPUT CHANGED for 444] : considering cbf bits for best cost estimation for 444 format
    • 2 days ago Steve Borho analysis: cleanup checkInter functions
    • 2 days ago Steve Borho analysis: remove unnecessary set of skip flags in checkInter_rd5_6()
    • 2 days ago Steve Borho search: large mostly mechanical change to pass cu by reference
    • 2 days ago Steve Borho cudata: remove unused method
    • 2 days ago Steve Borho predict: enforce calling conventions, fix wrong side-effects
    • 2 days ago Steve Borho predict: rename members for clarity, save work in singleMotionEstimation()
    • 2 days ago Steve Borho search: fix a change of outputs from f3bd6e5a880a, always zero unused refs
    • 3 days ago Steve Borho search: re-combine --pme with --no-pme code paths
    • 3 days ago Steve Borho cudata: cache numPartInCUSize as a class static
    • 3 days ago Steve Borho entropy: drop last use of g_winUnitX, g_winUnitY
    • 3 days ago Steve Borho slice: move numPartitions and numPartInCUSize from FrameData to SPS
    • 3 days ago Steve Borho cudata: remove default arguments for getPUAboveRightAdi(), getPUBelowLeftAdi()
    • 3 days ago Steve Borho cudata: simplify allocation / initialization interfaces
    • 3 days ago Steve Borho search: use intptr_t for picture stride variables
    • 3 days ago Steve Borho cudata: push more data type casts out to callers
    • 3 days ago Steve Borho cudata: use static array of absolute depth broadcast set functions
    • 3 days ago Steve Borho analysis: encodeResidue rewrite, much improved --rd 0
Version 1.3.733
  • Release Date: Oct 24, 2014
    • 7 hours ago Steve Borho nr: move noise reduction arrays to Quant, simplify its upkeep
    • 8 hours ago Steve Borho cleanup enough of the context tables to bring them into common/
    • 8 hours ago Steve Borho pull TComRom into common/constants. bring CommonDef and TypeDefs into various
    • 9 hours ago Steve Borho cudata: nits, use m_encData directly
    • 9 hours ago Steve Borho search: use proper allocation size for m_rqt, fix --preset placebo
    • 13 hours ago Steve Borho analysis: fix --rd 1,2 behavior (broken by 84933c3136ec)
    • 13 hours ago Steve Borho cudata: avoid 'char subscript' warnings from GCC
    • 13 hours ago Steve Borho framedata: param is now unused
    • 15 hours ago Steve Borho slice: move numCUsInFrame from FrameData to SPS
    • 15 hours ago Steve Borho framefilter: optimize row-height operations
    • 15 hours ago Steve Borho slice: move numCuInWidth, numCuInHeight to SPS, remove Frame pointer
    • 16 hours ago Steve Borho cudata: keep a FrameData pointer instead of a Frame pointer
    • 16 hours ago Steve Borho cudata: pass picWidth and picHeight to calcCTUGeoms()
    • 16 hours ago Steve Borho cudata: simplify setQPSubCUs()
    • 16 hours ago Steve Borho cudata: consistent names for consistent functionality
    • 16 hours ago Steve Borho cudata: no need to pass part size to setAll*, it can look it up itself
    • 17 hours ago Steve Borho search: use helper function to set transform depth
    • 17 hours ago Steve Borho cudata: simplify clearCbf()
    • 17 hours ago Steve Borho cudata: hoist a number of trivial functions to cudata.h for inlining
    • 17 hours ago Steve Borho cudata: clean up a couple defines
    • 17 hours ago Steve Borho search: remove unnecessary parens
    • 17 hours ago Steve Borho cudata: remove depth argument to setInterDirSubParts()
    • 18 hours ago Steve Borho cudata: merge motion field data directly into the CUData
    • 18 hours ago Steve Borho mvfield: move TComMvField into cudata as MVField
    • 28 hours ago Steve Borho mvfield: combine mv buffers into a single allocation
    • 28 hours ago Steve Borho mvfield: remove unnecessary depth arguments to set functions
    • 28 hours ago Steve Borho search: nits
    • 28 hours ago Steve Borho mvfield: remove setAllMvField wrapper
    • 28 hours ago Steve Borho mvfield: remove setMvField() method
    • 28 hours ago Steve Borho mvfield: pass objects by reference, const when possible
    • 28 hours ago Steve Borho cudata: replace clearMvField() with m_partSet calls
    • 28 hours ago Steve Borho cudata: copy mvfield data from CTU in copyFromPic (more --rd 0 fixes)
    • 28 hours ago Steve Borho entropy: remove only reference to TComCUMvField outside of cudata.h
    • 29 hours ago Steve Borho search: avoid unnecessary memcopies in inter prediction
    • 29 hours ago Steve Borho defs: remove last traces of REF_PIC_LIST enums, NOT_VALID -> REF_NOT_VALID
    • 29 hours ago Steve Borho mvfield: remove getRefIdx and getMvd access methods
    • 29 hours ago Steve Borho mvfield: remove getMv access method
    • 29 hours ago Steve Borho mvfield: remove default arguments, use consistent variable names, cleanup
    • 30 hours ago Steve Borho mvfield: class to struct, remove setMvd method
Version 1.3.694
  • Release Date: Oct 23, 2014
    • 6 hours ago Steve Borho docs: improve documentation for tskip options
    • 6 hours ago Steve Borho search: fix --rdpenalty 2, make logic explicit (closes #71)
    • 24 hours ago Murugan Vairavel asm: avx2 asm code for 8bpp and 16bpp vesion of scale1D_128to64 module
    • 23 hours ago Praveen Tiwari weight_sp: sse version of asm code optimization
    • 29 hours ago Praveen Tiwari weight_pp: avx2 asm code as per new interface
    • 8 hours ago Steve Borho fix checked build errors
    • 8 hours ago Steve Borho picyuv: cache offset arrays in the top-level encoder
    • 9 hours ago Steve Borho cudata: add a helper function for not-present CUs
    • 9 hours ago Steve Borho cudata: comment nits
    • 9 hours ago Steve Borho search: move auto var initialization to avoid goto warning
    • 9 hours ago Steve Borho cudata: consistent naming rules for part data (singular, no b prefix for flags)
    • 10 hours ago Steve Borho cudata: cleanup copy methods, fill in missing copies in copyFromPic()
    • 10 hours ago Steve Borho cudata: reorder and white-space nits
    • 10 hours ago Steve Borho cudata: inline single caller of getPartPosition
    • 10 hours ago Steve Borho cudata: consistent use of absPartIdx
    • 10 hours ago Steve Borho cudata: remove unused setCUTransquantBypassSubParts
    • 10 hours ago Steve Borho cudata: m_cuTransquantBypass will not be set without lossless being enabled
    • 10 hours ago Steve Borho cudata: nits
    • 10 hours ago Steve Borho cudata: define copy and broadcast set methods
    • 12 hours ago Steve Borho cudata: simplify setPartSizeSubParts and friends
    • 12 hours ago Steve Borho bring TComDataCU into common/ as CUData
    • 13 hours ago Steve Borho rebrand CU/cuData as CUGeom/cuGeom
    • 13 hours ago Steve Borho encoder: improve comment, frame encoders to not use the worker pool
    • 13 hours ago Steve Borho frameencoder: pre-calculate the set of unique geoms for the picture size
    • 15 hours ago Steve Borho frameencoder: group pointer members together for better alignment
    • 15 hours ago Steve Borho cu: nits
    • 15 hours ago Steve Borho cu: use memset in setQPSubParts()
    • 15 hours ago Steve Borho cu: inline single use of setSubPart into setInterDirSubParts
    • 15 hours ago Steve Borho cu: remove unused enum NDBFBlockBorderTag
    • 15 hours ago Steve Borho cu: repair original intent of getLastCodedQP(), fix 73c6c9086577
    • 15 hours ago Steve Borho cu: remove m_cuLocalData from CU, use child offsets instead of absolute indices
    • 16 hours ago Steve Borho cu: pass m_cuLocalData to loadCTUData()
    • 16 hours ago Steve Borho cu: style nits, no behavior change
    • 16 hours ago Steve Borho search: m_qtTempCoeff[ttype][qtLayer] -> m_rqt[qtLayer].tmpCoeff[ttype]
    • 16 hours ago Steve Borho search: m_qtTempShortYuv[qtlayer] -> m_rqt[layer].tmpShortYuv
    • 16 hours ago Steve Borho search: move inter search temp Yuv buffers into RQTData
    • 17 hours ago Steve Borho search: rename m_rdContexts to m_rqt since its purpose has expanded
    • 25 hours ago gopi jayaraman search: --pme bug fixes with slave threads tie up case
    • 31 hours ago Steve Borho search: move the destroy() method to after initSearch()
    • 30 hours ago Steve Borho frame: initialize m_reconPicYuv pointer
    • 29 hours ago Deepthi Nandakumar search: make split choice logic cleaner
    • 29 hours ago Deepthi Nandakumar search: rename variables to mightSplit and mightNotSplit
    • 32 hours ago Steve Borho search: use small struct to accumulate costs
    • 33 hours ago Steve Borho search: remove shadow variable
    • 34 hours ago Steve Borho search: improve comments
    • 34 hours ago Steve Borho search: avoid redundant work in typical path
    • 34 hours ago Steve Borho search: inline single caller of xGetIntraBitsLuma, avoid extra copies
    • 34 hours ago Steve Borho search: inline single call of calcIntraLumaRecon
    • 35 hours ago Steve Borho search: rename xRecurIntraCodingQT to codeIntraLumaQT, save a bit of work
    • 35 hours ago Steve Borho search: combine xRecurIntraCodingQT() bCheckFull sections
    • 35 hours ago Steve Borho search: add tmpBits to xRecurIntraCodingQT
    • 36 hours ago Steve Borho search: remove remnants of tqbypass from xRecurIntraCodingQT, improve var names
    • 36 hours ago Steve Borho search: create a per-depth temp buffer for temporary recon blocks
    • 36 hours ago Steve Borho search: fix --rdpenalty=2 at --rd 0,1
    • 36 hours ago Steve Borho search: use member variable access to current slice and frame, not the cu's
    • 36 hours ago Steve Borho search: clarify --rdpenalty
    • 37 hours ago Steve Borho search: simplify checkTransformSkip logic
    • 37 hours ago Steve Borho encoder: make assignments to Frame::m_intraData and m_interData unconditional
    • 37 hours ago Steve Borho search: remove redundant clearCbf call
    • 37 hours ago Steve Borho search: move checkIntra from Analysis to Search
    • 37 hours ago Steve Borho search: rename xSetResidualQTData to saveResidualQTData and simplify
    • 38 hours ago Steve Borho search: do not return CBF=0 distortion from xEstimateResidualQT()
    • 39 hours ago Steve Borho analysis: nit
    • 40 hours ago Steve Borho doc: update description of --cu-lossless
    • 40 hours ago Steve Borho analysis: do not try lossless if best mode had no distortion
    • 41 hours ago Steve Borho analysis: copy inter prediction when evaluating lossless
    • 42 hours ago Aarthi Thirumalai rc: fix error in vbv due to access of unreferenced refFrame data
Version 1.3.654
  • Release Date: Oct 21, 2014
Version 1.3.612
  • Release Date: Oct 20, 2014
Version 1.3.582
  • Release Date: Oct 18, 2014
Version 1.3.386
  • Release Date: Oct 17, 2014
    • 14 hours ago Steve Borho analysis: fix CABAC context state handling after splits [CHANGES OUTPUTS]
    • 20 hours ago Steve Borho entropy: add a mechanism to detect reads without writes in checked builds
    • 21 hours ago Steve Borho entropy: give each Search instance its own set of RD contexts
    • 15 hours ago Steve Borho threadpool: nit
    • 16 hours ago Steve Borho threading: nits
    • 16 hours ago Steve Borho cmake: nit
    • 21 hours ago Steve Borho entropy: make copy methods properly const
    • 20 hours ago Steve Borho cmake: bump X265_BUILD for new parallelism params
    • 33 hours ago Steve Borho encoder: prevent broken combinations of options
    • 35 hours ago Steve Borho analysis: encodeIntraInInter() must write directly to given cabac context
    • 2 days ago Steve Borho analysis: use enums to enumerate prediction buffers
    • 2 days ago Steve Borho analysis: add a hack to try and match --pmode with --no-pmode
    • 2 days ago Steve Borho analysis: cleanups
    • 2 days ago Steve Borho analysis: measure best pmode intra RD cost in worker thread
    • 2 days ago Steve Borho analysis: do not pass member vars to checkMerge2Nx2N_rd0_4 as pointer references
    • 2 days ago Steve Borho param: show when pmode and pme are enabled
    • 2 days ago Steve Borho analysis: remove default argument for compressInterCU_rd5_6
    • 2 days ago Steve Borho TComDataCU: make most get methods const, remove some trivial access methods
    • 2 days ago Steve Borho analysis: use slave instance for MVP eval for --pme
    • 2 days ago Steve Borho analysis: replace prepMotionCompensation() calls in parallelInterSearch()
    • 2 days ago Steve Borho analysis: --pme workers do not need m_origYuv in slave instance
    • 3 days ago Steve Borho analysis: cleanup variable names
    • 3 days ago Steve Borho analysis: prevent race hazard in parallel ME state variables
    • 3 days ago Steve Borho analysis: parallel ME can get partsize and depth from ME CU
    • 3 days ago Steve Borho analysis: support --pme for all rd levels
    • 3 days ago Steve Borho api: add --pme to enable parallel motion estimation
    • 4 days ago Steve Borho analysis: further parallelME progress
    • 4 days ago Steve Borho api: add --pmode to enable parallel mode decision
    • 3 days ago Steve Borho encoder: correct logging of number of WPP streams
Version 1.3.362
  • Release Date: Oct 7, 2014
    • 14 hours ago Steve Borho analysis: fix CABAC context state handling after splits [CHANGES OUTPUTS]
    • 20 hours ago Steve Borho entropy: add a mechanism to detect reads without writes in checked builds
    • 21 hours ago Steve Borho entropy: give each Search instance its own set of RD contexts
    • 15 hours ago Steve Borho threadpool: nit
    • 16 hours ago Steve Borho threading: nits
    • 16 hours ago Steve Borho cmake: nit
    • 21 hours ago Steve Borho entropy: make copy methods properly const
    • 20 hours ago Steve Borho cmake: bump X265_BUILD for new parallelism params
    • 33 hours ago Steve Borho encoder: prevent broken combinations of options
    • 35 hours ago Steve Borho analysis: encodeIntraInInter() must write directly to given cabac context
    • 2 days ago Steve Borho analysis: use enums to enumerate prediction buffers
    • 2 days ago Steve Borho analysis: add a hack to try and match --pmode with --no-pmode
    • 2 days ago Steve Borho analysis: cleanups
    • 2 days ago Steve Borho analysis: measure best pmode intra RD cost in worker thread
    • 2 days ago Steve Borho analysis: do not pass member vars to checkMerge2Nx2N_rd0_4 as pointer references
    • 2 days ago Steve Borho param: show when pmode and pme are enabled
    • 2 days ago Steve Borho analysis: remove default argument for compressInterCU_rd5_6
    • 2 days ago Steve Borho TComDataCU: make most get methods const, remove some trivial access methods
    • 2 days ago Steve Borho analysis: use slave instance for MVP eval for --pme
    • 2 days ago Steve Borho analysis: replace prepMotionCompensation() calls in parallelInterSearch()
    • 2 days ago Steve Borho analysis: --pme workers do not need m_origYuv in slave instance
    • 3 days ago Steve Borho analysis: cleanup variable names
    • 3 days ago Steve Borho analysis: prevent race hazard in parallel ME state variables
    • 3 days ago Steve Borho analysis: parallel ME can get partsize and depth from ME CU
    • 3 days ago Steve Borho analysis: support --pme for all rd levels
    • 3 days ago Steve Borho api: add --pme to enable parallel motion estimation
    • 4 days ago Steve Borho analysis: further parallelME progress
    • 4 days ago Steve Borho api: add --pmode to enable parallel mode decision
    • 3 days ago Steve Borho encoder: correct logging of number of WPP streams
Version 1.3.331
  • Release Date: Oct 6, 2014
    • 5 hours ago Steve Borho search: remove redundant calls to prepMotionCompensation()
    • 6 hours ago Steve Borho search: use sad cost directly to pick MVP
    • 6 hours ago Steve Borho search: make some helper methods const
    • 7 hours ago Steve Borho analysis: remove unused variables, fixes warnings
    • 16 hours ago Aarthi Thirumalai rc: correct the threshold for resetABR function
    • 16 hours ago Aarthi Thirumalai rc : correct max AU size for first frame
    • 11 hours ago Satoshi Nakagawa fix bug in 73c6c9086577 for rdLevel=0
    • 29 hours ago Steve Borho analysis: further work on parallel ME
    • 16 hours ago Steve Borho analysis: nit, remove obviously wrong comment
    • 16 hours ago Steve Borho analysis: use source buffer for source stride
    • 16 hours ago Steve Borho analysis: initialize job counters
    • 15 hours ago Steve Borho analysis: remove bMergeOnly argument to checkInter_rd0_4, always false
    • 16 hours ago Steve Borho analysis: remove PartSize argument to checkIntraInInter_rd0_4
    • 17 hours ago Deepthi Nandakumar ratecontrol: fix float absolute check
    • 17 hours ago Deepthi Nandakumar ratecontrol: replace an imprecise comparison with a more precise check to ensure
    • 31 hours ago Deepthi Nandakumar slice: better structure packing
    • 32 hours ago Steve Borho analysis: fixup
    • 35 hours ago Steve Borho analysis: move non-distributed path into else clause
    • 4 days ago Steve Borho stub in framework for parallel mode analysis and parallel ME
    • 35 hours ago Steve Borho analysis: remove default arguments to checkInter_rd5_6 and checkInter_rd0_4
    • 36 hours ago Steve Borho cli: display param->bSaoNonDeblocked as bool in CLI help
Version 1.3.330
  • Release Date: Oct 3, 2014
    • 5 hours ago Steve Borho search: remove redundant calls to prepMotionCompensation()
    • 6 hours ago Steve Borho search: use sad cost directly to pick MVP
    • 6 hours ago Steve Borho search: make some helper methods const
    • 7 hours ago Steve Borho analysis: remove unused variables, fixes warnings
    • 16 hours ago Aarthi Thirumalai rc: correct the threshold for resetABR function
    • 16 hours ago Aarthi Thirumalai rc : correct max AU size for first frame
    • 11 hours ago Satoshi Nakagawa fix bug in 73c6c9086577 for rdLevel=0
    • 29 hours ago Steve Borho analysis: further work on parallel ME
    • 16 hours ago Steve Borho analysis: nit, remove obviously wrong comment
    • 16 hours ago Steve Borho analysis: use source buffer for source stride
    • 16 hours ago Steve Borho analysis: initialize job counters
    • 15 hours ago Steve Borho analysis: remove bMergeOnly argument to checkInter_rd0_4, always false
    • 16 hours ago Steve Borho analysis: remove PartSize argument to checkIntraInInter_rd0_4
    • 17 hours ago Deepthi Nandakumar ratecontrol: fix float absolute check
    • 17 hours ago Deepthi Nandakumar ratecontrol: replace an imprecise comparison with a more precise check to ensure
    • 31 hours ago Deepthi Nandakumar slice: better structure packing
    • 32 hours ago Steve Borho analysis: fixup
    • 35 hours ago Steve Borho analysis: move non-distributed path into else clause
    • 4 days ago Steve Borho stub in framework for parallel mode analysis and parallel ME
    • 35 hours ago Steve Borho analysis: remove default arguments to checkInter_rd5_6 and checkInter_rd0_4
    • 36 hours ago Steve Borho cli: display param->bSaoNonDeblocked as bool in CLI help
Version 1.3.315
  • Release Date: Oct 1, 2014
    • 12 hours ago Steve Borho analysis: nit
    • 12 hours ago Steve Borho replace lcu with ctu in variable names
    • 12 hours ago Steve Borho replace LCU with CTU globally in comments
    • 12 hours ago Steve Borho sao: nits
    • 12 hours ago Steve Borho sao: rename resetLcuPart to resetCtuPart
    • 12 hours ago Steve Borho sao: rename saoLcuParam to ctuParam
    • 12 hours ago Steve Borho sao: rename SaoLcuParam to SaoCtuParam
    • 12 hours ago Steve Borho api: rename --sao-lcu-bounds to --sao-non-deblock
    • 13 hours ago Steve Borho search: make bidir temp YUVs Search members
    • 13 hours ago Steve Borho datacu: coding style rename of m_DataCUMemPool and m_CULocalData
    • 13 hours ago Steve Borho motioninfo: coding style rename of m_MVFieldMemPool
    • 14 hours ago Steve Borho common: break into debugger when check fails in debug build
    • 14 hours ago Steve Borho TComData: do not leave m_tqBypassOrigYuv uninitialized
    • 24 hours ago Murugan Vairavel asm: avx2 assembly code for idct4x4
    • 29 hours ago Praveen Tiwari blockfill_s_32x32 avx2 asm code: performance improved from 1354.05 cycles to 705.81 cycles, over sse version of asm code
    • 29 hours ago Praveen Tiwari blockfill_s_16x16 avx2 asm code: performance improved from 389.21 cycles to 204.38 cycles, over sse version of asm code
    • 2 days ago Aarthi Thirumalai rc: apply maxAU size restrictions while encoding each frame
    • 31 hours ago Santhoshini Sekar TComDataCU: replace getTotalNumPart() with CU structure details
    • 31 hours ago Santhoshini Sekar TComDataCU: replace getZorderIdxInCU() with encodeIdx of CU structure
    • 27 hours ago Steve Borho rd: move lambda and analysis qp init to rdcost.h
    • 34 hours ago Satoshi Nakagawa sao: remove frame-based SAO
    • 36 hours ago Steve Borho TComMvField: class to struct, white-space cleanups
    • 5 days ago Steve Borho search: reorder nits, no effect
    • 5 days ago Steve Borho quant: pass entropy instance through init function
    • 5 days ago Steve Borho search: white-space nits
    • 5 days ago Steve Borho search: give each Search instance an Entropy encoder (no output changes)
    • 4 days ago David T Yuen Changes for loadCTUData
    • 4 days ago Steve Borho nits
Version 1.3.288
  • Release Date: Oct 1, 2014
    • 3 days ago Murugan Vairavel asm: avx2 assembly code for idct32x32
    • 3 days ago David T Yuen Removed unnecessary call to loadCTUData
    • 3 days ago Steve Borho analysis: more style nits, code simplifications. no behavior change
    • 3 days ago Steve Borho analysis: remove #define conditionals for control flow
    • 3 days ago Steve Borho analysis: coding style and comment nits
    • 3 days ago Steve Borho analysis: remove unused LAMBDA_PARTITION_SELECT
    • 3 days ago Steve Borho analysis: hoist local function into anonymous namespace (file local)
    • 4 days ago Yuvaraj Venkatesh asm: avx2 assembly code for idct8x8
    • 4 days ago Santhoshini Sekar remove getNumPartInCU() and replace it with macro
    • 3 days ago Satoshi Nakagawa refine deblocking filter

Version 1.3.282
  • Release Date: Sep 30, 2014
    • 3 days ago Murugan Vairavel asm: avx2 assembly code for idct32x32
    • 3 days ago David T Yuen Removed unnecessary call to loadCTUData
    • 3 days ago Steve Borho analysis: more style nits, code simplifications. no behavior change
    • 3 days ago Steve Borho analysis: remove #define conditionals for control flow
    • 3 days ago Steve Borho analysis: coding style and comment nits
    • 3 days ago Steve Borho analysis: remove unused LAMBDA_PARTITION_SELECT
    • 3 days ago Steve Borho analysis: hoist local function into anonymous namespace (file local)
    • 4 days ago Yuvaraj Venkatesh asm: avx2 assembly code for idct8x8
    • 4 days ago Santhoshini Sekar remove getNumPartInCU() and replace it with macro
    • 3 days ago Satoshi Nakagawa refine deblocking filter

Version 1.3.271
  • Release Date: Sep 27, 2014
    • 36 hours ago Steve Borho predict: split weighted prediction values from WeightParam
    • 36 hours ago Steve Borho predict: combine and check allocations and return failures
    • 36 hours ago Steve Borho predict: remove check for reallocations, comment nits
    • 36 hours ago Steve Borho predict: inline predInterUni(), getWpScaling() and simplify motionCompensation()
    • 37 hours ago Steve Borho predict: inline single call of predInterBi()
    • 38 hours ago Steve Borho encoder: rename cuCoder to analysis for better clarity

Version 1.3.265
  • Release Date: Sep 26, 2014
    • 9 hours ago Steve Borho Backed out changeset: eb011fa1d2d8
    • 16 hours ago David T Yuen Changed FrameEncoder::m_tld to a pointer and set it to one of Encoder's ThreadLocalData instances.
    • 15 hours ago Steve Borho vec: make a note for why we keep some of the remaining vector routines
    • 16 hours ago Steve Borho vec: remove idct8, we have SSSE3 assembly for it
    • 16 hours ago Steve Borho cmake: remove blockcopy-sse3.cpp
    • 16 hours ago Steve Borho primitives: remove unused block copy primitives
    • 22 hours ago Praveen Tiwari bloccopy_pp avx asm code: 32x32, 32x48, 32x64 improved by 803.69 -> 514.90, 1126.36 -> 655.24, 1454.09 -> 835.76 cycles
    • 24 hours ago Praveen Tiwari blockcopy_pp_32x24: avx asm code, improved 621.84 cycles -> 371.94
    • 24 hours ago Praveen Tiwari blockcopy_pp_32x16: avx asm code, improved 477.74 cycles -> 309.99
    • 26 hours ago Praveen Tiwari blockcopy_pp_32x8: avx asm code, improved 281.20 cycles -> 165.47
    • 27 hours ago Satoshi Nakagawa refine deblocking filter
    • 4 days ago Steve Borho predict: remove checkIdenticalMotion()
    • 4 days ago Steve Borho predict: remove list argument from motionCompensation(), always REF_PIC_LIST_X
    • 4 days ago Steve Borho predict: streamline getWpScaling()
    • 4 days ago Steve Borho predict: use faster unidir prediction for B frames when weighting not enabled
    • 4 days ago Steve Borho predict: combine redundant logic paths in predInterBi()
    • 30 hours ago Deepthi Nandakumar Backed out changeset: fa2f1aa1456e
    • 47 hours ago Sagar Kotecha blockcopy_ss: 64x16, 64x32, 64x48, 64x64 AVX version of asm code, approx double speedup comapre to SSE
    • 2 days ago Yuvaraj Venkatesh asm: avx2 code for dct8x8
    • 41 hours ago Min Chen asm: replace mova by movu to avoid AVX2 testbench crash in dct16, dct32, denoise_dct, its same speed on Haswell
    • 2 days ago Sagar Kotecha add avx version for chroma_copy_ss 16x4, 16x8, 16x12, 16x16, 16x24, 16x32, 16x64 based on csp, approx 1.5x-2x speedup over SSE
    • 2 days ago Satoshi Nakagawa simplify intra filter (with fix for da61cf406f16)
    • 2 days ago Deepthi Nandakumar Backed out changeset: 25dde1ffab66
    • 3 days ago Deepthi Nandakumar nits: use parantheses to improve readability in shifts
    • 4 days ago Steve Borho predict: don't bother keeping refidx as an array
    • 4 days ago Steve Borho predict: merge TComWeightPrediction functions into Predict
    • 4 days ago Steve Borho TComWeightPrediction: combine duplicate inline functions (refs #80)
    • 4 days ago Steve Borho encoder: use %u to sprintf unsigned ints (refs #80)
    • 4 days ago Steve Borho entropy: fix SAO enable detection (refs #80)
    • 4 days ago Steve Borho bitstream: add paren to avoid ambiguous precedence in X265_CHECK
    • 4 days ago Steve Borho nits: do not check for NULL from new operations
    • 4 days ago Steve Borho bitcost: use enums for special constants rather than static const ints
    • 4 days ago Steve Borho motion: avoid extra iterations when no subpel motion found
    • 3 days ago Deepthi Nandakumar search: clean xRecurIntraCodingQT
    • 3 days ago Deepthi Nandakumar analysis: nits
    • 3 days ago Deepthi Nandakumar psy-rd: fix bug in chroma psyEnergy for intra 4x4
    • 3 days ago Deepthi Nandakumar analysis: remove CheckBestMode from CheckIntra
    • 3 days ago Deepthi Nandakumar analysis: remove redundant variables, cleanup variable names
Version 1.3.248
  • Release Date: Sep 25, 2014
    • 18 hours ago Steve Borho param: do not allow VBV without WPP
    • 2 days ago Gopu Govindaswamy search: simplify and remove redundant variables in getBestIntraModeChroma
    • 2 days ago Gopu Govindaswamy search: remove redundant loacal variables in encodeResAndCalcRdSkipCU
    • 3 days ago Gopu Govindaswamy search: cleanup and remove redundant variable in checkintra
    • 2 days ago Satoshi Nakagawa primitives: intra_pred[4][35] => intra_pred[35][4] (avoid *35)
    • 2 days ago Satoshi Nakagawa inline simple functions
    • 3 days ago Steve Borho testbench: allocate test harnesses on heap, for better valgrind coverage
    • 2 days ago Praveen Tiwari denoiseDct: align performance data while reporting speedup
    • 2 days ago Praveen Tiwari copy_cnt_32: avx2 asm code, improved 1521.17 cycles -> 934.46 cycles
    • 2 days ago Praveen Tiwari copy_cnt_16: avx2 asm code, improved 514.32 cycles -> 313.66 cycles
    • 2 days ago Praveen Tiwari denoise_dct: avx2 asm code
    • 2 days ago Praveen Tiwari denoise_dct asm code: SSE version
    • 3 days ago Murugan Vairavel asm: avx2 assembly code for idct16x16
    • 2 days ago Aarthi Thirumalai rc: improvements for cbr
    • 3 days ago Praveen Tiwari denoiseDct: nit unused asm function declarations
    • 3 days ago Praveen Tiwari denoiseDct asm code: nit faulty code, need a new SSE version
    • 3 days ago Praveen Tiwari denoiseDct unit test code: fixed bound value problem
    • 9 days ago Praveen Tiwari copy_cnt_4 avx2 asm code: nit, same speedup by sse version
    • 3 days ago Aarthi Thirumalai rc: fix bugs in using boundary condition for cu while encoding each frame.
    • 3 days ago Steve Borho frameencoder: fix VBV row resets when SAO is disabled
Version 1.3.243
  • Release Date: Sep 23, 2014
    • 18 hours ago Steve Borho param: do not allow VBV without WPP
    • 2 days ago Gopu Govindaswamy search: simplify and remove redundant variables in getBestIntraModeChroma
    • 2 days ago Gopu Govindaswamy search: remove redundant loacal variables in encodeResAndCalcRdSkipCU
    • 3 days ago Gopu Govindaswamy search: cleanup and remove redundant variable in checkintra
    • 2 days ago Satoshi Nakagawa primitives: intra_pred[4][35] => intra_pred[35][4] (avoid *35)
    • 2 days ago Satoshi Nakagawa inline simple functions
    • 3 days ago Steve Borho testbench: allocate test harnesses on heap, for better valgrind coverage
    • 2 days ago Praveen Tiwari denoiseDct: align performance data while reporting speedup
    • 2 days ago Praveen Tiwari copy_cnt_32: avx2 asm code, improved 1521.17 cycles -> 934.46 cycles
    • 2 days ago Praveen Tiwari copy_cnt_16: avx2 asm code, improved 514.32 cycles -> 313.66 cycles
    • 2 days ago Praveen Tiwari denoise_dct: avx2 asm code
    • 2 days ago Praveen Tiwari denoise_dct asm code: SSE version
    • 3 days ago Murugan Vairavel asm: avx2 assembly code for idct16x16
    • 2 days ago Aarthi Thirumalai rc: improvements for cbr
    • 3 days ago Praveen Tiwari denoiseDct: nit unused asm function declarations
    • 3 days ago Praveen Tiwari denoiseDct asm code: nit faulty code, need a new SSE version
    • 3 days ago Praveen Tiwari denoiseDct unit test code: fixed bound value problem
    • 9 days ago Praveen Tiwari copy_cnt_4 avx2 asm code: nit, same speedup by sse version
    • 3 days ago Aarthi Thirumalai rc: fix bugs in using boundary condition for cu while encoding each frame.
    • 3 days ago Steve Borho frameencoder: fix VBV row resets when SAO is disabled
Version 1.3.226
  • Release Date: Sep 22, 2014
    • 18 hours ago Steve Borho param: do not allow VBV without WPP
    • 2 days ago Gopu Govindaswamy search: simplify and remove redundant variables in getBestIntraModeChroma
    • 2 days ago Gopu Govindaswamy search: remove redundant loacal variables in encodeResAndCalcRdSkipCU
    • 3 days ago Gopu Govindaswamy search: cleanup and remove redundant variable in checkintra
    • 2 days ago Satoshi Nakagawa primitives: intra_pred[4][35] => intra_pred[35][4] (avoid *35)
    • 2 days ago Satoshi Nakagawa inline simple functions
    • 3 days ago Steve Borho testbench: allocate test harnesses on heap, for better valgrind coverage
    • 2 days ago Praveen Tiwari denoiseDct: align performance data while reporting speedup
    • 2 days ago Praveen Tiwari copy_cnt_32: avx2 asm code, improved 1521.17 cycles -> 934.46 cycles
    • 2 days ago Praveen Tiwari copy_cnt_16: avx2 asm code, improved 514.32 cycles -> 313.66 cycles
    • 2 days ago Praveen Tiwari denoise_dct: avx2 asm code
    • 2 days ago Praveen Tiwari denoise_dct asm code: SSE version
    • 3 days ago Murugan Vairavel asm: avx2 assembly code for idct16x16
    • 2 days ago Aarthi Thirumalai rc: improvements for cbr
    • 3 days ago Praveen Tiwari denoiseDct: nit unused asm function declarations
    • 3 days ago Praveen Tiwari denoiseDct asm code: nit faulty code, need a new SSE version
    • 3 days ago Praveen Tiwari denoiseDct unit test code: fixed bound value problem
    • 9 days ago Praveen Tiwari copy_cnt_4 avx2 asm code: nit, same speedup by sse version
    • 3 days ago Aarthi Thirumalai rc: fix bugs in using boundary condition for cu while encoding each frame.
    • 3 days ago Steve Borho frameencoder: fix VBV row resets when SAO is disabled
Version 1.3.199
  • Release Date: Sep 18, 2014
    • 20 hours ago Sagar Kotecha add fanout validation module to check param compatibility
    • 21 hours ago Gopu Govindaswamy api: do not reuse the analysisData buffer for more then one picture, set it NULL
    • 26 hours ago Santhoshini Sekar analysis: add CU specific details to encodeCU()
    • 21 hours ago Steve Borho analysis: nits
    • 21 hours ago Gopu Govindaswamy analysis: intra picture estimation (mode and split decision)information sharing
    • 26 hours ago Praveen Tiwari denoiseDct: test bench code
    • 41 hours ago Steve Borho search: save a few cycles
    • 2 days ago Aarthi Thirumalai rc: fixes for 2 pass +- vbv to calculate frameSizePlanned accurately.
Version 1.3.198
  • Release Date: Sep 16, 2014
    • 20 hours ago Steve Borho param: preset tuning changes
    • 20 hours ago Steve Borho doc: describe performance impact of SAO
    • 20 hours ago Steve Borho doc: fix typo and nit in threading page
    • 4 days ago Satoshi Nakagawa sao: some cleanups
    • 21 hours ago Steve Borho search: header cleanups, no functional change
    • 5 days ago Steve Borho search: measure RDO of intra modes within 12% of least cost [CHANGES OUTPUTS]
    • 21 hours ago Steve Borho search: comment nits
    • 21 hours ago Steve Borho Merge with stable
    • 22 hours ago Aarthi Thirumalai rc: bug fix for 2 pass when bframes = 0. fixes Issue #77stable
    • 22 hours ago Aarthi Thirumalai rc: check for changes in scenecut input between multiple passes.
    • 22 hours ago Aarthi Thirumalai rc: bug fix for 2 pass when bframes = 0. fixes Issue #77
    • 5 days ago Ashok Kumar Mishra Search: remove redundant encode coefficients in intra for performance
    • 25 hours ago Murugan Vairavel asm: fix mismatch due to dct32 avx2 assembly code
    • 4 days ago Murugan Vairavel asm: avx2 assembly code for dct32x32
    • 2 days ago Deepthi Nandakumar x265: add missing typedefs
    • 4 days ago David T Yuen Resolved gcc compiler error of mismatched type
    • 3 days ago Deepthi Nandakumar Merge with stable
    • 3 days ago Deepthi Nandakumar asm: disable buggy denoise primitives until the bugs are fixedstable
    • 4 days ago Sagar Kotecha cli: add cli options analysis-mode and analysis-file
    • 4 days ago Sagar Kotecha store analysis information in buffers
    • 4 days ago Sagar Kotecha api: introduce methods to allocate and free analysis buffers
    • 4 days ago Sagar Kotecha api: add analysis data structures and param options
    • 4 days ago Steve Borho analysis: minor comment and code cleanups, no behavior change
    • 4 days ago Steve Borho search: remove x prefixes from ME helper functions
    • 4 days ago Steve Borho dpb: does not need to include frameencoder.h
Version 1.3.186
  • Release Date: Sep 15, 2014
    • 28 hours ago Steve Borho analysis: nit
    • 28 hours ago Steve Borho search: don't pass top-level encoder to initSearch()
    • 33 hours ago Steve Borho search: store rd costs in first pass through intra modes
    • 29 hours ago Steve Borho cleanup header dependencies; dpb should not need encoder.h
    • 13 hours ago Deepthi Nandakumar asm: enable copy_cnt8
    • 38 hours ago Yuvaraj Venkatesh asm: avx2 assembly code for dct16
    • 37 hours ago Praveen Tiwari copy_cnt_8, AVX2 asm code as per new interface, performance improved from 5.13x to 7.59x on HASWELL-I5
    • 34 hours ago Steve Borho encoder: nits
    • 2 days ago Steve Borho search: re-enable chroma tskip
    • 2 days ago Aarthi Thirumalai rc: use m_frameDuration instead of rce->frameDuration to derive complexity for each frame in 2nd pass.
    • 38 hours ago Deepthi Nandakumar param: apply missing default values to param, mostly zero and reorder
    • 47 hours ago Steve Borho search: reverse meaning of bCheckFirst to bAllowRQTSplit
    • 2 days ago Steve Borho search: !a ? b : c; -> a ? c : b;
    • 2 days ago Steve Borho search: rename some helper functions without x prefixes
    • 2 days ago Steve Borho backout 0dc2cbc36ee5 to 331ef5121676
    • 2 days ago Ashok Kumar Mishra analysis: modified compressInterCU_rd0_4() with CU-specific information
    • 2 days ago Steve Borho search: fix camel case of residualQTIntraChroma
    • 2 days ago Steve Borho search: don't pass a zeroDistortion pointer if you don't want the answer
    • 2 days ago Steve Borho search: return distortion from xEstimateResidualQT
    • 2 days ago Steve Borho search: pass depthRange uniformly as uint32_t depthRange[2]
    • 2 days ago Steve Borho search: return distortion from xRecurIntraCodingQT
    • 2 days ago Steve Borho search: return distortion from xIntraCodingChromaBlk, do not pass by ref
    • 2 days ago Steve Borho search: return distortion from xRecurIntraChromaCodingQT, do not pass by ref
    • 2 days ago Steve Borho search: return distortion from xIntraCodingLumaBlk, do not pass by reference
    • 2 days ago Praveen Tiwari copy_cnt_8 AVX2 asm code, as per new interface
    • 2 days ago Praveen Tiwari copy_cnt_4: faster AVX2 code
    • 2 days ago Praveen Tiwari copy_cnt_4: combine mova and paddb to reduce code size, same speedup
    • 2 days ago Praveen Tiwari copy_cnt_4: enable fast non zero coefficient count path
    • 2 days ago Steve Borho frameencoder: use simple shifts to scale 2-pass CU type counters
    • 2 days ago Steve Borho frameencoder: use x265_emms() prior to double QP clipping for VBV
    • 2 days ago Deepthi Nandakumar search: remove warning from MS compiler
    • 2 days ago Min Chen asm: avx2 version of quant, improve 16.6k cycles -> 8.4k cycles
    • 2 days ago Min Chen asm: improve quant by replace variant shift to fixed shift, 19k cycles -> 16.6k cycles
    • 2 days ago Min Chen testbench(quant): the qBits value must be more than or equal to 8
    • 2 days ago Min Chen testbench(quant): the Round value must be less than (2 ^ qbits)
Version 1.3.175
  • Release Date: Sep 12, 2014
    • 28 hours ago Steve Borho analysis: nit
    • 28 hours ago Steve Borho search: don't pass top-level encoder to initSearch()
    • 33 hours ago Steve Borho search: store rd costs in first pass through intra modes
    • 29 hours ago Steve Borho cleanup header dependencies; dpb should not need encoder.h
    • 13 hours ago Deepthi Nandakumar asm: enable copy_cnt8
    • 38 hours ago Yuvaraj Venkatesh asm: avx2 assembly code for dct16
    • 37 hours ago Praveen Tiwari copy_cnt_8, AVX2 asm code as per new interface, performance improved from 5.13x to 7.59x on HASWELL-I5
    • 34 hours ago Steve Borho encoder: nits
    • 2 days ago Steve Borho search: re-enable chroma tskip
    • 2 days ago Aarthi Thirumalai rc: use m_frameDuration instead of rce->frameDuration to derive complexity for each frame in 2nd pass.
    • 38 hours ago Deepthi Nandakumar param: apply missing default values to param, mostly zero and reorder
    • 47 hours ago Steve Borho search: reverse meaning of bCheckFirst to bAllowRQTSplit
    • 2 days ago Steve Borho search: !a ? b : c; -> a ? c : b;
    • 2 days ago Steve Borho search: rename some helper functions without x prefixes
    • 2 days ago Steve Borho backout 0dc2cbc36ee5 to 331ef5121676
    • 2 days ago Ashok Kumar Mishra analysis: modified compressInterCU_rd0_4() with CU-specific information
    • 2 days ago Steve Borho search: fix camel case of residualQTIntraChroma
    • 2 days ago Steve Borho search: don't pass a zeroDistortion pointer if you don't want the answer
    • 2 days ago Steve Borho search: return distortion from xEstimateResidualQT
    • 2 days ago Steve Borho search: pass depthRange uniformly as uint32_t depthRange[2]
    • 2 days ago Steve Borho search: return distortion from xRecurIntraCodingQT
    • 2 days ago Steve Borho search: return distortion from xIntraCodingChromaBlk, do not pass by ref
    • 2 days ago Steve Borho search: return distortion from xRecurIntraChromaCodingQT, do not pass by ref
    • 2 days ago Steve Borho search: return distortion from xIntraCodingLumaBlk, do not pass by reference
    • 2 days ago Praveen Tiwari copy_cnt_8 AVX2 asm code, as per new interface
    • 2 days ago Praveen Tiwari copy_cnt_4: faster AVX2 code
    • 2 days ago Praveen Tiwari copy_cnt_4: combine mova and paddb to reduce code size, same speedup
    • 2 days ago Praveen Tiwari copy_cnt_4: enable fast non zero coefficient count path
    • 2 days ago Steve Borho frameencoder: use simple shifts to scale 2-pass CU type counters
    • 2 days ago Steve Borho frameencoder: use x265_emms() prior to double QP clipping for VBV
    • 2 days ago Deepthi Nandakumar search: remove warning from MS compiler
    • 2 days ago Min Chen asm: avx2 version of quant, improve 16.6k cycles -> 8.4k cycles
    • 2 days ago Min Chen asm: improve quant by replace variant shift to fixed shift, 19k cycles -> 16.6k cycles
    • 2 days ago Min Chen testbench(quant): the qBits value must be more than or equal to 8
    • 2 days ago Min Chen testbench(quant): the Round value must be less than (2 ^ qbits)
Version 1.3.145
  • Release Date: Sep 10, 2014
    • 16 hours ago Steve Borho search: prune more unnecessary work from estIntraPredQT()
    • 18 hours ago Steve Borho search: remove some redundant work from estIntraPredQT
    • 18 hours ago Steve Borho search: remove m_qtTempTrIdx
    • 20 hours ago Steve Borho pull search class into encoder/ (TLibEncoder is no more)
    • 18 hours ago Steve Borho api: remove include of frameencoder
    • 20 hours ago Steve Borho nits
    • 26 hours ago Ashok Kumar Mishra analysis: modified compressInterCU_rd5_6() with CU-specific information
    • 29 hours ago Ashok Kumar Mishra Analysis: compressIntraCU clean up
    • 21 hours ago Steve Borho sao: move frame/slice initialization into SAO::startSlice
    • 22 hours ago Steve Borho sao: minor cleanups, no behavior change
    • 24 hours ago Steve Borho frameencoder: avoid another call to resetEntropy(), they are expensive
    • 24 hours ago Steve Borho frameencoder: combine some conditional expressions
    • 24 hours ago Steve Borho rc: move FrameStats to ratecontrol.h
    • 24 hours ago Steve Borho frameencoder: rename percent fields for clarity
    • 24 hours ago Steve Borho frameencoder: do more CU stat math as integer
    • 26 hours ago Steve Borho frameencoder: merge more of encodeSlice() into processCU
    • 3 days ago Steve Borho frameencoder: remove second encodeCU() pass over CTUs when SAO is disabled
    • 25 hours ago Steve Borho frameencoder: remove redundant clear of frame stats
    • 25 hours ago Steve Borho nits
    • 27 hours ago Steve Borho Merge with correct x86inc.asm patch
    • 30 hours ago Deepthi Nandakumar entropy: change top-level encode to encodeCTU
    • 3 days ago Min Chen x86inc.asm: fix vpbroadcastd bug on Mac platform
    • 2 days ago Satoshi Nakagawa fix sao
    • 2 days ago Satoshi Nakagawa fix CHECKED_BUILD
    • 3 days ago Min Chen asm: AVX2 version of dequant_normal, improve 9.3k Cycles -> 4.2k Cycles
    • 3 days ago Min Chen asm: reduce number of movd in dequant_normal
    • 3 days ago Min Chen x86inc.asm: fix vpbroadcastd bug on Mac platform
    • 3 days ago Steve Borho frameencoder: remove unnecessary call to resetBits(), improve comment
    • 4 days ago Steve Borho frameencoder: cleanup CTURow::init
    • 4 days ago Steve Borho frameencoder: CTURow class -> struct, remove m_ prefixes
    • 4 days ago Steve Borho tld: remove m_ prefixes from struct members
    • 4 days ago Steve Borho cturow: merge CTURow and TLD objects into frameencoder.h
    • 4 days ago Steve Borho frameencoder: remove unused variable, nit
    • 4 days ago Steve Borho cturow: inline processCU()
    • 4 days ago Steve Borho cmake: enable MACOSX_RPATH on Mac shared libraries
    • 4 days ago Praveen Tiwari copy_cnt 4x4 AVX2 asm code, as per new interface
    • 4 days ago Min Chen asm: reenable IACA support, it remove by 'inappropriate instruction...' patch
    • 4 days ago Min Chen asm: avx2 version of nquant(), improve 9.8k cycles -> 5.3k cycles
    • 4 days ago Min Chen asm: optimize nquant by PSIGND, improve 11k cycles -> 9.8k cycles
    • 4 days ago Praveen Tiwari count_nonzero asm code, reduceded code size by combining mova and packsswb
    • 4 days ago Praveen Tiwari copy_cnt: nits
    • 4 days ago Praveen Tiwari copy_cnt 4x4, eliminated move instructions, +-1x improvement
    • 4 days ago Satoshi Nakagawa fix getQuadtreeTULog2MinSizeInCU()
    • 4 days ago Min Chen asm: enable SSE2 version of pixel_ssd_ss[]
    • 4 days ago Min Chen asm: fix output mistake in pixel_ssd_ss_4xN
    • 4 days ago Satoshi Nakagawa fix cbf context
    • 4 days ago Min Chen asm: replace ssse3 instruction in pixel_ssd_ss_*_sse2
    • 4 days ago Steve Borho asm: fix SSSE3 in SSE2 build warning in unused primitive
    • 4 days ago Steve Borho quant: fix check macro
    • 4 days ago Steve Borho quant: nits
    • 6 days ago Praveen Tiwari quant path cleanup
    • 7 days ago Praveen Tiwari quant_c optimization, downscaling qCoef from int32_t* to int16_t*
    • 7 days ago Praveen Tiwari optimize cvt32to16_shl by replacing copy_shl
    • 7 days ago Praveen Tiwari added copy_shl primitive
    • 7 days ago Praveen Tiwari added copy_shr primitive
    • 7 days ago Praveen Tiwari temporarily disable avx2 version of copy_cnt primitive, need to update as per new interface
    • 7 days ago Praveen Tiwari conv16to32_count renamed to copy_count as per new interface
    • 7 days ago Praveen Tiwari conv16to32_count C interface modification, downscaling coeff from int32_t* to int16_t*
    • 8 days ago Praveen Tiwari cvt16to32_cnt optimization
    • 2 weeks ago Praveen Tiwari signBitHidingHDQ optimization, downscaling coeff from int32_t* to int16_t*
    • 2 weeks ago Praveen Tiwari rdoQuant optimization, downscaling dstCoeff fron int32_t* to int16_t*
    • 7 days ago Praveen Tiwari nquant optimization, downscaling qCoef from int32_t* to int16_t*
    • 2 weeks ago Praveen Tiwari quant.cpp, cleaned redundant code
    • 7 days ago Praveen Tiwari dequant_scaling optimization, downscaling quantCoef from int32_t* to int16_t*
    • 7 days ago Praveen Tiwari dequant_normal asm code optimization as per new interface
    • 7 days ago Praveen Tiwari dequant_normal optimization, downscaling quantCoef from int32_t* to int16_t*
    • 2 weeks ago Praveen Tiwari count_nonzero primitive optimization, downscaling quantCoef from int32_t* to int16_t*
    • 5 days ago Anton Mitrofanov x86asm: warn when inappropriate instruction used in function with specified cpuflags
    • 6 days ago Ashok Kumar Mishra TComDataCU: Reduced repeated function call to calculate depth range
    • 6 days ago Satoshi Nakagawa change index of m_buOffsetY[] from raster to zscan
    • 5 days ago David T Yuen Cleaned up TComDataCU::getQuadtreeTULog2MinSizeInCU for clarity and a bit of performance
    • 8 days ago Dnyaneshwar G asm: avx2 asm code for dct4
    • 5 days ago Deepthi Nandakumar Merge
    • 5 days ago Deepthi Nandakumar entropy: cleanup codeQtRootCbf
    • 5 days ago Deepthi Nandakumar quant: use table for ctxCbf
    • 5 days ago Deepthi Nandakumar entropy: cleanup codePredMode
    • 5 days ago Deepthi Nandakumar entropy: cleanup codeCUTransQuantBypassFlag
    • 8 days ago Ashok Kumar Mishra Entropy: Replaced getCtxQtCbf() with table
    • 5 days ago Steve Borho Merge with stable
    • 5 days ago Steve Borho version: detect OpenBSD for version string (closes #76)stable
    • 5 days ago Steve Borho cmake: fix BSD link, only link with librt if it exists (closes #75)stable
    • 6 days ago Murugan Vairavel asm: enable 16bpp primitives of cvt32to16 and cvt16to32 for all block sizes
    • 5 days ago Min Chen testbench(nquant): the Round value must be less than (2 ^ qbits)
    • 5 days ago David T Yuen Resolve gcc warnings
    • 5 days ago Ashok Kumar Mishra fix: hash/binary mismatch for new CU structure holds CU-specific info
    • 12 days ago Ashok Kumar Mishra analysis: CU structure now holds CU-specific information,
    • 8 days ago Deepthi Nandakumar Merge with stable
    • 8 days ago Deepthi Nandakumar Backed out changeset: 35b2d9e774c8stable
Version 1.3.59
  • Release Date: Sep 3, 2014
    • 2 hours ago Steve Borho Merge with stable
    • 4 days ago Steve Borho encoder: re-enable --cu-losslessstable
    • 4 days ago Min Chen search: fix decoder intra crash with --cu-losslessstable
    • 4 days ago Deepthi Nandakumar cu-lossless: fix inter hash mistakestable
    • 5 hours ago Satoshi Nakagawa asm: fix dequant_normalstable
    • 26 hours ago Steve Borho sao: fix signed loop bounds bug
    • 4 days ago Steve Borho encoder: re-enable --cu-lossless
    • 4 days ago Deepthi Nandakumar cu-lossless: fix inter hash mistake
    • 2 days ago Steve Borho mbdst: cleanup dequant test
    • 2 days ago Steve Borho mbdst: make buffers nice even size
    • 2 days ago Steve Borho test: fix MSVC warnings
    • 2 days ago Steve Borho mbdst: align buffers to 32byte boundary
    • 2 days ago Steve Borho mbdst: fix memset lengths, change MEM_CMP_SIZE to MAX_TU_SIZE
    • 2 days ago Steve Borho mbdst: reshuffle varnames
    • 2 days ago Steve Borho mbdst: remove redundant mintbuf buffers
    • 2 days ago Steve Borho mbdst: remove unused mbuf4, rename other short output buffers
    • 2 days ago Steve Borho mbdstharness: remove completely unused arrays
    • 2 days ago Steve Borho pixel: use fixed buffer allocations
    • 2 days ago Steve Borho ipfilter: use fixed buffer allocations
    • 2 days ago Steve Borho testbench: move test benches from stack allocations to global
    • 2 days ago Steve Borho intra: use fixed buffer allocations
Version 1.3.57
  • Release Date: Aug 31, 2014
    • 2 hours ago Steve Borho Merge with stable
    • 4 days ago Steve Borho encoder: re-enable --cu-losslessstable
    • 4 days ago Min Chen search: fix decoder intra crash with --cu-losslessstable
    • 4 days ago Deepthi Nandakumar cu-lossless: fix inter hash mistakestable
    • 5 hours ago Satoshi Nakagawa asm: fix dequant_normalstable
    • 26 hours ago Steve Borho sao: fix signed loop bounds bug
    • 4 days ago Steve Borho encoder: re-enable --cu-lossless
    • 4 days ago Deepthi Nandakumar cu-lossless: fix inter hash mistake
    • 2 days ago Steve Borho mbdst: cleanup dequant test
    • 2 days ago Steve Borho mbdst: make buffers nice even size
    • 2 days ago Steve Borho test: fix MSVC warnings
    • 2 days ago Steve Borho mbdst: align buffers to 32byte boundary
    • 2 days ago Steve Borho mbdst: fix memset lengths, change MEM_CMP_SIZE to MAX_TU_SIZE
    • 2 days ago Steve Borho mbdst: reshuffle varnames
    • 2 days ago Steve Borho mbdst: remove redundant mintbuf buffers
    • 2 days ago Steve Borho mbdst: remove unused mbuf4, rename other short output buffers
    • 2 days ago Steve Borho mbdstharness: remove completely unused arrays
    • 2 days ago Steve Borho pixel: use fixed buffer allocations
    • 2 days ago Steve Borho ipfilter: use fixed buffer allocations
    • 2 days ago Steve Borho testbench: move test benches from stack allocations to global
    • 2 days ago Steve Borho intra: use fixed buffer allocations
Version 1.3.55
  • Release Date: Aug 28, 2014
    • 16 hours ago Steve Borho Merge with stable
    • 16 hours ago Steve Borho docs: fix a typostable
    • 16 hours ago Steve Borho mbdst: add missing test of dequant_scaling
    • 16 hours ago Steve Borho mbdst: use fixed buffer allocations, cleanup some tests
    • 22 hours ago Ashok Kumar Mishra TEncSearch: remove redundant function call
    • 23 hours ago Aarthi Thirumalai rc: fix bug in predicting B frame bits in vbvstable
    • 27 hours ago Deepthi Nandakumar Merge with stable
    • 30 hours ago Aarthi Thirumalai rc: fix bitrate accuracy for vbv in fast presets.stable
    • 30 hours ago Aarthi Thirumalai rc: rename m_bframes to m_leadingBframesstable
Version 1.3.36
  • Release Date: Aug 28, 2014
    • 18 hours ago Steve Borho common: rename QP range macros to be consistent with x264
    • 18 hours ago Steve Borho types: remove generic TEXT_CHROMA enum, no longer used
    • 18 hours ago Steve Borho pattern: use isLuma instead of restricted TextType range
    • 39 hours ago Steve Borho sao: don't pass member variables to functions
    • 20 hours ago Steve Borho framefilter: move SAO init logic into the frame filter
    • 19 hours ago Min Chen x86inc: Make INIT_CPUFLAGS support an arbitrary number of cpuflags
    • 19 hours ago Min Chen asm: Minor pixel_ssim_end4 improvements
    • 30 hours ago Satoshi Nakagawa fix m_initSliceContext (uninitialised m_sliceQp)
    • 40 hours ago Min Chen search: fix decoder intra crash with --cu-lossless

Version 1.3.36
  • Release Date: Aug 28, 2014
    • 18 hours ago Steve Borho common: rename QP range macros to be consistent with x264
    • 18 hours ago Steve Borho types: remove generic TEXT_CHROMA enum, no longer used
    • 18 hours ago Steve Borho pattern: use isLuma instead of restricted TextType range
    • 39 hours ago Steve Borho sao: don't pass member variables to functions
    • 20 hours ago Steve Borho framefilter: move SAO init logic into the frame filter
    • 19 hours ago Min Chen x86inc: Make INIT_CPUFLAGS support an arbitrary number of cpuflags
    • 19 hours ago Min Chen asm: Minor pixel_ssim_end4 improvements
    • 30 hours ago Satoshi Nakagawa fix m_initSliceContext (uninitialised m_sliceQp)
    • 40 hours ago Min Chen search: fix decoder intra crash with --cu-lossless

Version 1.3.28
  • Release Date: Aug 27, 2014
    • 31 hours ago Satoshi Nakagawa replace g_rasterToPelX[g_zscanToRaster[idx]] by g_zscanToPelX[idx]
    • 23 hours ago Steve Borho Merge with stable
    • 33 hours ago Aarthi Thirumalai rc: don't read slicetypes from qpfile in 2nd pass.stable
    • 34 hours ago Aarthi Thirumalai rc: write I/i slice in stats file based on whether openGop is enabled or not.stable
    • 3 days ago Satoshi Nakagawa fix losslessstable
    • 3 days ago Steve Borho sao: nits
    • 3 days ago Steve Borho sao: remove unused m_qp member
    • 4 days ago Steve Borho sao: nit
    • 4 days ago Steve Borho sao: simplify entropy coder initialization [CHANGES OUTPUTS]
    • 3 days ago Steve Borho sao: fix inf loop bounds
    • 3 days ago Steve Borho sao: fix MSVC warnings
Version 1.3.20
  • Release Date: Aug 24, 2014
    • 33 hours ago Steve Borho sao: minor logic simplifications and cleanups
    • 34 hours ago Steve Borho sao: use typedefs to simplify multi-dimensional mallocs
    • 35 hours ago Steve Borho sao: sChroma should have been isChroma
    • 35 hours ago Steve Borho sao: handle malloc failures more cleanly
    • 35 hours ago Steve Borho sao: rename shared table to m_tableBo
    • 35 hours ago Steve Borho sao: share m_lumaTableBo with chroma
    • 36 hours ago Steve Borho sao: use checked mallocs, disable SAO if any mallocs fail
    • 36 hours ago Steve Borho common: define a CHECKED_MALLOC_ZERO method to auto-reset allocated memory
    • 36 hours ago Steve Borho sao: move getPicYuvAddr() to TComPicYuv and simplify
    • 36 hours ago Steve Borho sao: remove unused m_tmpYuv and an obsolete comment
    • 37 hours ago Steve Borho sao: luma and chroma can share a clip table
    • 37 hours ago Steve Borho sao: unify signOf
    • 37 hours ago Steve Borho sao: use more compile-time values
    • 37 hours ago Steve Borho sao: we don't support different bit depths between luma and chroma
    • 37 hours ago Steve Borho sao: remove vars which are dups of params
    • 38 hours ago Steve Borho sao: fix warnings
    • 39 hours ago Steve Borho Merge with stable
    • 39 hours ago Steve Borho search: nit
    • 40 hours ago Steve Borho nits
    • 40 hours ago Steve Borho sao: convertLevelRowCol2Idx did not need to be a member function
    • 40 hours ago Steve Borho sao: merge create/destroy methods
    • 40 hours ago Steve Borho sao: use m_param directly for user-configurables
    • 40 hours ago Steve Borho sao: cleanup member prefixes, remove createPicSaoInfo and endSaoEnc
    • 40 hours ago Steve Borho pull SAO code into encoder/ and begin cleanup
    • 39 hours ago Steve Borho Added tag 1.3 for changeset c1e4fc0162c1stable
    • 39 hours ago Steve Borho encoder: disable --cu-lossless until coding errors can be fixed (refs #73)stable 1.3
    • 2 days ago Satoshi Nakagawa simplify AMVPInfo to MV[2]

Version 1.2.580
  • Release Date: Aug 22, 2014
    • 17 hours ago Steve Borho encoder: nit, remove obsolete comment
    • 17 hours ago Steve Borho Merge with stable
    • 17 hours ago Steve Borho encoder: do not allow scaling lists to be used with 4:4:4 color space (refs #72)stable
    • 18 hours ago Steve Borho search: nit
    • 18 hours ago Steve Borho Merge with stable
    • 18 hours ago Steve Borho search: do not skip intra tu 32x32 unless TU intra depth is deep (refs #71)stable
    • 40 hours ago David T Yuen Removed code that will never be executed.
    • 42 hours ago Min Chen fix single thread encode issue
    • 42 hours ago Min Chen cleanup: remove reduce condition check
    • 44 hours ago Steve Borho Merge with stable
    • 2 days ago Steve Borho frameencoder: do not allocate noise reduction struct unless NR is enabledstable
    • 2 days ago Deepthi Nandakumar search: nit
    • 2 days ago Deepthi Nandakumar search: cleanup variable names and lossless comments
    • 2 days ago Deepthi Nandakumar Merge with stable
    • 2 days ago Deepthi Nandakumar psy-rdoq: eliminate dependency between psy-rd, psy-rdoq and rdoq.stable
    • 2 days ago Steve Borho param: correctly detect adaptive quant usage when checking --psnrstable

Version 1.2.575
  • Release Date: Aug 21, 2014
    • 14 hours ago Steve Borho Merge with stable
    • 14 hours ago Min Chen sao: move m_upBuff* to stack, fixes (#69)stable
    • 14 hours ago Min Chen cleanup: remove unused m_upBuff2stable
    • 15 hours ago Steve Borho encoder: fix deadlocks/crashes when picture is a single CTU rowstable
    • 16 hours ago Steve Borho Merge with stable
    • 16 hours ago Steve Borho param: typostable
    • 2 days ago Steve Borho quant: combine psy-rdoq scale and lambda into one variable
    • 2 days ago Steve Borho entropy: pass bIsLuma to estBit() instead of limited TextType range
    • 2 days ago Steve Borho entropy: simplify EstBitsSbac
    • 16 hours ago Steve Borho Merge with stable
    • 17 hours ago Steve Borho encoder: remove m_maxCuDQPDepth, simplify dqp logic
    • 27 hours ago Satoshi Nakagawa refine depth related.
    • 18 hours ago Steve Borho cli: fix help syntaxstable
    • 19 hours ago Steve Borho param: add psy-rd and psy-rdoq to param2string (for info SEI and stats file)stable
    • 19 hours ago Steve Borho cli: update psy-rdoq rangw in CLI helpstable
    • 2 days ago Steve Borho ctu: optimize away per-row bitstream coders, no output changes
    • 2 days ago Steve Borho analysis: move encodeCU() methods and friends to Entropy
    • 24 hours ago Deepthi Nandakumar Merge with stable
    • 24 hours ago Deepthi Nandakumar analysis: missing braces caused a bug in CABAC context save for rd=2stable
    • 29 hours ago Tom Vaughan introduction.rst edited online with Bitbucket
    • 29 hours ago Tom Vaughan introduction.rst edited online with Bitbucket - added link to MPEG-LA
    • 2 days ago Steve Borho rest: fix typo, document default psy-rdoq valuestable
    • 2 days ago Steve Borho merge with default, feature freeze for 1.3stable
    • 2 days ago Steve Borho param: enable --fast-intra for all fast modes and turbo first-pass, document
    • 2 days ago Steve Borho slicetype: use fast intra scan unconditionally in lookahead
    • 3 days ago Steve Borho rest: improve documentation for psycho-visual options
    • 2 days ago Steve Borho analysis: nits
    • 2 days ago Steve Borho quant: set lambdas at the same time quant QP is configured
    • 3 days ago Steve Borho param: re-enable psy-rdoq range check for [0..10], print with two digits
    • 3 days ago Steve Borho rdcost: add arbitray scale to make --psy-rd 0..2 range give acceptable quality
    • 4 days ago Steve Borho quant: scale psy-rdoq by lambda

Version 1.2.570
  • Release Date: Aug 19, 2014
    • 14 hours ago Steve Borho Merge with stable
    • 14 hours ago Min Chen sao: move m_upBuff* to stack, fixes (#69)stable
    • 14 hours ago Min Chen cleanup: remove unused m_upBuff2stable
    • 15 hours ago Steve Borho encoder: fix deadlocks/crashes when picture is a single CTU rowstable
    • 16 hours ago Steve Borho Merge with stable
    • 16 hours ago Steve Borho param: typostable
    • 2 days ago Steve Borho quant: combine psy-rdoq scale and lambda into one variable
    • 2 days ago Steve Borho entropy: pass bIsLuma to estBit() instead of limited TextType range
    • 2 days ago Steve Borho entropy: simplify EstBitsSbac
    • 16 hours ago Steve Borho Merge with stable
    • 17 hours ago Steve Borho encoder: remove m_maxCuDQPDepth, simplify dqp logic
    • 27 hours ago Satoshi Nakagawa refine depth related.
    • 18 hours ago Steve Borho cli: fix help syntaxstable
    • 19 hours ago Steve Borho param: add psy-rd and psy-rdoq to param2string (for info SEI and stats file)stable
    • 19 hours ago Steve Borho cli: update psy-rdoq rangw in CLI helpstable
    • 2 days ago Steve Borho ctu: optimize away per-row bitstream coders, no output changes
    • 2 days ago Steve Borho analysis: move encodeCU() methods and friends to Entropy
    • 24 hours ago Deepthi Nandakumar Merge with stable
    • 24 hours ago Deepthi Nandakumar analysis: missing braces caused a bug in CABAC context save for rd=2stable
    • 29 hours ago Tom Vaughan introduction.rst edited online with Bitbucket
    • 29 hours ago Tom Vaughan introduction.rst edited online with Bitbucket - added link to MPEG-LA
    • 2 days ago Steve Borho rest: fix typo, document default psy-rdoq valuestable
    • 2 days ago Steve Borho merge with default, feature freeze for 1.3stable
    • 2 days ago Steve Borho param: enable --fast-intra for all fast modes and turbo first-pass, document
    • 2 days ago Steve Borho slicetype: use fast intra scan unconditionally in lookahead
    • 3 days ago Steve Borho rest: improve documentation for psycho-visual options
    • 2 days ago Steve Borho analysis: nits
    • 2 days ago Steve Borho quant: set lambdas at the same time quant QP is configured
    • 3 days ago Steve Borho param: re-enable psy-rdoq range check for [0..10], print with two digits
    • 3 days ago Steve Borho rdcost: add arbitray scale to make --psy-rd 0..2 range give acceptable quality
    • 4 days ago Steve Borho quant: scale psy-rdoq by lambda

Version 1.2.547
  • Release Date: Aug 16, 2014
    • 17 hours ago Steve Borho analysis: avoid MSVC uninitialized variable warning
    • 22 hours ago Steve Borho quant: use optimized primitive for transquant bypass in invTransformNxN
    • 19 hours ago Steve Borho param: don't allow turbo mode to increase rd-level, improve docs
    • 23 hours ago Steve Borho analysis: use macro and for-loop to simplify fast-intra
    • 26 hours ago David T Yuen Added fast intra search option to Analysis::checkIntraInInter_rd0_4
Version 1.2.536
  • Release Date: Aug 15, 2014
  • 19 minutes ago Steve Borho nit
    19 minutes ago Steve Borho asm: fix HIGH_BIT_DEPTH build again
    2 hours ago Steve Borho analysis: white-space nits, possibly exposing a logic bug
    2 hours ago Steve Borho remove predict.h includes from TLibCommon/
    2 hours ago Steve Borho quant: +- has higher precedence than ?; use parens to fix unquantShift logic
    18 hours ago Satoshi Nakagawa replace IntraFilterType[][] to IntraFilterFlags[] (aboid *35)
    13 hours ago Deepthi Nandakumar slicetype: initialise lowmode to 4.
    15 hours ago Steve Borho fix HIGH_BIT_DEPTH compile on Windows
    16 hours ago Steve Borho quant: typo
    16 hours ago Steve Borho quant: update comment for UNQUANT macro
    24 hours ago Steve Borho quant: simplify UNQUANT() variables
    36 hours ago Deepthi Nandakumar quant: remove extra upshift by 4 in s_invQuantScales
    20 hours ago Min Chen move m_estBitsSbac from Quant to Entropy
    20 hours ago Min Chen asm: integrate denoise_dct, malloc NoiseReduction for alignment
    22 hours ago Steve Borho scaling: remove array of rarely used trivially calculated values
    22 hours ago Steve Borho scaling: remove s_numListsAtSize[], it is 6 at every size
    23 hours ago Steve Borho predict: nits
    23 hours ago Steve Borho predict: prevent MSVC warning about int to bool conversion
    34 hours ago Aarthi Thirumalai refactor: cleanup setting up of lambda
    34 hours ago Aarthi Thirumalai cli: nits. fix typo
    28 hours ago Aarthi Thirumalai rc: bug fix - m_avgQpAq can be from 0-69.
    24 hours ago Steve Borho cli: free param structures on early-outs
    25 hours ago Steve Borho asm: fix HIGH_BIT_DEPTH link problems
    25 hours ago Steve Borho slicetype: adjust variable scoping to avoid GCC warnings
    39 hours ago Steve Borho slicetype: nit
    39 hours ago Steve Borho cmake: bump build number
    46 hours ago David T Yuen Added fast intra search option
    2 days ago David T Yuen Moved IntraFilterType from intrapred.cpp to predict.h to make it available
    2 days ago Sumalatha Polureddy analysis: check for proper cost
Version 1.2.518
  • Release Date: Aug 14, 2014
  • 6 hours ago Steve Borho quant: initialize qp to impossible value so first set always works
    9 hours ago Satoshi Nakagawa replace g_convertToBit[] to g_log2Size[] const table
    13 hours ago Min Chen asm: AVX2 version cvt32to16_shl
    2 days ago Satoshi Nakagawa quant: remove scaledCoeff from nquant()
    43 hours ago Aarthi Thirumalai rc: nits; add LF in frame stats of 2 pass
    39 hours ago Steve Borho level: signal profile/level NONE if non-compliant configuration is found
Version 1.2.513
  • Release Date: Aug 12, 2014
    • 6 hours ago Steve Borho quant: initialize qp to impossible value so first set always works
    • 9 hours ago Satoshi Nakagawa replace g_convertToBit[] to g_log2Size[] const table
    • 13 hours ago Min Chen asm: AVX2 version cvt32to16_shl
    • 2 days ago Satoshi Nakagawa quant: remove scaledCoeff from nquant()
    • 43 hours ago Aarthi Thirumalai rc: nits; add LF in frame stats of 2 pass
    • 39 hours ago Steve Borho level: signal profile/level NONE if non-compliant configuration is found
Version 1.2.496
  • Release Date: Aug 10, 2014
    • 6 hours ago Steve Borho quant: improve flow and comments for last non-zero refinement
    • 7 hours ago Steve Borho quant: comment improvements
    • 11 hours ago Steve Borho quant: remove floating point operations from RDOQ [CHANGES OUTPUTS]
    • 11 hours ago Steve Borho quant: cleanup chroma QP function
    • 11 hours ago Steve Borho quant: header cleanups, no functional change
    • 24 hours ago Steve Borho quant: use standard rd cost formula for sign-hiding [CHANGES OUTPUTS]
    • 25 hours ago Steve Borho quant: do not check CG bitmap for implied-present coeff groups
    • 25 hours ago Steve Borho quant: reduce conditional expression depths (mostly for readability)
    • 26 hours ago Steve Borho quant: avoid an extra shift by adjusting the unquant coeff shift
    • 35 hours ago Steve Borho quant: improve variable names and comments (no behavior change)
    • 35 hours ago Steve Borho entropy: simplify sign hide flag
    • 2 days ago Min Chen asm: cvt32to16_shl[*] for TSkip
    • 2 days ago Min Chen asm: cvt16to32_shr[*] for TSkip
    • 2 days ago Satoshi Nakagawa denoise: fix numCoeff (bug from 42b1d7c17510)
    • 2 days ago Steve Borho main10: create a hybrid all-angs primitve for 16bpp compiles
    • 2 days ago Steve Borho entropy: remove implicit memset from constructor
    • 2 days ago Steve Borho entropy: pad size of context array to 32 * 5 bytes
    • 2 days ago Steve Borho entropy: remove ContextModel structure, use uint8_t directly
    • 2 days ago Steve Borho entropy: remove bBinsCoded from ContextModel (no more users)
    • 2 days ago Steve Borho entropy: disable signaling of CABAC init state
    • 2 days ago Steve Borho frameencoder: nit
    • 2 days ago Steve Borho frameencoder: avoid redundant calls to resetEntropy()
    • 2 days ago Steve Borho quant: avoid runtime check of transform shift size
    • 2 days ago Ashok Kumar Mishraanalysis: cleanup
    • 2 days ago Aarthi Thirumalai rc: set rdlevel to 2 in fast first pass for multipass encode.
    • 2 days ago Satoshi Nakagawa asm: cleanup unused sub_ps and add_ps
    • 2 days ago Steve Borho entropy: remove m_ prefix from ContextModel.state, other nits
    • 3 days ago Steve Borho param: log noise reduction use/strength
    • 3 days ago Steve Borho refactor: move motion reference pointers from TEncSearch to Slice
    • 3 days ago Steve Borho denoise: further cleanups
    • 3 days ago Praveen Tiwari noise reduction, cleaned unused buffer space
    • 3 days ago Praveen Tiwari noise reduction, use all eight categories of transform for inter blocks
    • 3 days ago Aarthi Thirumalai rc : nit
    • 4 days ago Min Chen asm: cvt16to32_cnt[32x32] for TSkip
    • 4 days ago Satoshi Nakagawa primitives for RExt
    • 4 days ago Santhoshini Sekar TComPattern: remove redundant functions used for CIP
Version 1.2.473
  • Release Date: Aug 6, 2014
    • 14 hours ago Steve Borho me: clip motion search area to signaled motion vector length limits
    • 16 hours ago Steve Borho asm: asm header updates
    • 18 hours ago Min Chen asm: cvt16to32_cnt[16x16] for TSkip
    • 19 hours ago Steve Borho quant: change how RDOQ measures distortion [CHANGES OUTPUTS]
    • 23 hours ago Steve Borho psy-rdoq: fix unquant shift factors
    • 25 hours ago Steve Borho quant: remove TODO comment
    • 4 days ago Aarthi Thirumalai rc: add slow first pass as option. set turbo first pass as default in 2 pass
    • 38 hours ago Santhoshini Sekar bugfix: disable aq only when both aq-strength and cu-tree is off
    • 38 hours ago Steve Borho psy-rdoq: include psy-cost in uncoded coefficient distortion
    • 44 hours ago Steve Borho pixelharness: fix MSVC warning about intptr_t conversion to int
    • 45 hours ago Steve Borho quant: nit
    • 45 hours ago Steve Borho entropy: cleanup calcCost, add EMMS for float safety
    • 45 hours ago Steve Borho quant: assign a signal cost of zero to zero coeff after lastNZ
    • 46 hours ago Steve Borho quant: rdStats -> cgRdStats
    • 46 hours ago Steve Borho quant: cleanup coeff group zero-check code
    • 46 hours ago Steve Borho quant: sign hide nits
    • 47 hours ago Steve Borho quant: denoise nits
    • 47 hours ago Steve Borho quant: don't pass member variable to class method
    • 2 days ago Steve Borho quant: inline the quant method, only called from transformNxN
    • 2 days ago Steve Borho quant: disallow psy-rdoq if transformSkip was used
    • 2 days ago Steve Borho quant: comment nits
    • 2 days ago Steve Borho quant: remove curUseRDOQ argument from transformNxN, always true
    • 2 days ago Steve Borho search: remove curUseRDOQ arguments from two more methods
    • 2 days ago Steve Borho search: hard-code curUseRDOQ
    • 2 days ago Steve Borho search: remove curUseRDOQ argument from encodeResAndCalcRdInterCU, always true
    • 2 days ago Steve Borho quant: hoist IEP_RATE out of getICRateCost
    • 3 days ago Min Chen asm: cvt16to32_cnt[8x8] for TSkip
    • 3 days ago Min Chen asm: cvt16to32_cnt[4x4] for TSkip
    • 3 days ago Min Chen update header and support Intel IACA marker
    • 3 days ago Steve Borho quant: change lastCG into a bool, use isOne flag to avoid abs() calls
    • 3 days ago Steve Borho quant: consistent comment style, improve comments
    • 3 days ago Steve Borho quant: levelDouble -> levelScaled
    • 3 days ago Steve Borho quant: rename sigCost to codedSigBits, comment nit
    • 3 days ago Steve Borho quant: RDO_CODED_LEVEL macro can now be inlined for easier debugging
    • 3 days ago Steve Borho quant: re-order rdoq logic so only one RDO_CODED_LEVEL() call is required
    • 3 days ago Steve Borho quant: more readability nits - no output changes
    • 3 days ago Steve Borho quant: improve comments for trailing zero coeff
    • 3 days ago Steve Borho quant: remove redundant level intialization
    • 3 days ago Steve Borho quant: blockUncodedCost -> totalUncodedCost, improve comments
    • 3 days ago Steve Borho quant: correct rounding factor for unquant
    • 3 days ago Steve Borho quant: clarify last-nz optimization loop
    • 3 days ago Steve Borho quant: rename costCoeff0 to costUncoded, add docs
    • 3 days ago Steve Borho quant: support scaling lists in psy-rdoq
    • 3 days ago Steve Borho quant: make IEP_RATE an anonymous enum, it doesn't need storage
    • 3 days ago Steve Borho quant: m_lambda2 no longer needs to be a member variable
    • 4 days ago Steve Borho quant: convert getCodedLevel() into a macro, remove m_transformShift hack
    • 4 days ago Steve Borho quant: simplify minAbsLevel
    • 4 days ago Steve Borho quant: hoist some calculations out of the loop
    • 4 days ago Steve Borho quant: delay err3, err4 calculation until/if necessary
    • 4 days ago Steve Borho quant: apply scale factor in just one place
    • 4 days ago Steve Borho dpb: cleanup decodingRefreshMarking()
    • 4 days ago Steve Borho dpb: remove checks for slice types we do not emit
    • 4 days ago Steve Borho dpb: style nits
    • 4 days ago Steve Borho dpb: getNalUnitType() cannot return NAL_UNIT_CODED_SLICE_IDR_N_LP
    • 4 days ago Steve Borho dpb: remove redundant call to getNalUnitType(), output will not change
    • 4 days ago Aarthi Thirumalai rc: enable abr reset in the first pass of two pass encode.
    • 4 days ago Santhoshini Sekar cleanup: move m_predYuv and m_predTempYuv from predict to TEncSearch
    • 4 days ago Deepthi Nandakumar predict: whitespace nits
Version 1.2.436
  • Release Date: Aug 4, 2014
    • 15 hours ago Santhoshini Sekar rc: update vbv for all b frames
    • 9 hours ago Deepthi Nandakumar predict: nits, cleanup, add TODO comments
    • 10 hours ago Deepthi Nandakumar predict: save clipped MVs
    • 20 hours ago Deepthi Nandakumar predict: prepare motionCompensation sets base class fields
    • 14 hours ago Steve Borho dpb: fix picture re-use bug introduced in bc53f6dcda7f
    • 23 hours ago Steve Borho entropy: remove unused flush function
    • 17 hours ago Steve Borho entropy: remove redundant !isIntra() check
    • 18 hours ago Steve Borho quant: use bisLuma
    • 18 hours ago Steve Borho quant: use x264 style lambda naming scheme
    • 15 hours ago Steve Borho dpb: cleanup prepareEncode()
    • 18 hours ago Steve Borho rdcost: settle on x264 style lambda naming scheme
    • 18 hours ago Steve Borho rdcost: change input type of setC*DistortionWeight to uint16_t
    • 18 hours ago Steve Borho rdcost: move zeroPel buffer to TEncSearch, its only user
    • 18 hours ago Steve Borho rdcost: keep psy-rd scale factor as an int (0..512), remove psyRdEnabled()
    • 15 hours ago Steve Borho common: remove unused SCALING_LIST_PARAMETER enum
    • 15 hours ago Steve Borho TComRom: nits
    • 19 hours ago Steve Borho entropy: nits
    • 19 hours ago Steve Borho entropy: use bIsLuma in codeCoeffNxN()
    • 17 hours ago Steve Borho slice: allow loop filter boundary flag to be analyzed
    • 19 hours ago Steve Borho quant: use bIsLuma args instead of using a subset of TextType
    • 19 hours ago Steve Borho TComTrQuant to Quant
    • 25 hours ago Steve Borho scaling: use anonymous enums for class constants
    • 34 hours ago Deepthi Nandakumar predict: predict.cpp/.h defined, hungarian function names cleaned up, warnings removed
Version 1.2.414
  • Release Date: Aug 2, 2014
    • 15 hours ago Santhoshini Sekar rc: update vbv for all b frames
    • 9 hours ago Deepthi Nandakumar predict: nits, cleanup, add TODO comments
    • 10 hours ago Deepthi Nandakumar predict: save clipped MVs
    • 20 hours ago Deepthi Nandakumar predict: prepare motionCompensation sets base class fields
    • 14 hours ago Steve Borho dpb: fix picture re-use bug introduced in bc53f6dcda7f
    • 23 hours ago Steve Borho entropy: remove unused flush function
    • 17 hours ago Steve Borho entropy: remove redundant !isIntra() check
    • 18 hours ago Steve Borho quant: use bisLuma
    • 18 hours ago Steve Borho quant: use x264 style lambda naming scheme
    • 15 hours ago Steve Borho dpb: cleanup prepareEncode()
    • 18 hours ago Steve Borho rdcost: settle on x264 style lambda naming scheme
    • 18 hours ago Steve Borho rdcost: change input type of setC*DistortionWeight to uint16_t
    • 18 hours ago Steve Borho rdcost: move zeroPel buffer to TEncSearch, its only user
    • 18 hours ago Steve Borho rdcost: keep psy-rd scale factor as an int (0..512), remove psyRdEnabled()
    • 15 hours ago Steve Borho common: remove unused SCALING_LIST_PARAMETER enum
    • 15 hours ago Steve Borho TComRom: nits
    • 19 hours ago Steve Borho entropy: nits
    • 19 hours ago Steve Borho entropy: use bIsLuma in codeCoeffNxN()
    • 17 hours ago Steve Borho slice: allow loop filter boundary flag to be analyzed
    • 19 hours ago Steve Borho quant: use bIsLuma args instead of using a subset of TextType
    • 19 hours ago Steve Borho TComTrQuant to Quant
    • 25 hours ago Steve Borho scaling: use anonymous enums for class constants
    • 34 hours ago Deepthi Nandakumar predict: predict.cpp/.h defined, hungarian function names cleaned up, warnings removed
Version 1.2.412
  • Release Date: Aug 2, 2014
    • 15 hours ago Santhoshini Sekar rc: update vbv for all b frames
    • 9 hours ago Deepthi Nandakumar predict: nits, cleanup, add TODO comments
    • 10 hours ago Deepthi Nandakumar predict: save clipped MVs
    • 20 hours ago Deepthi Nandakumar predict: prepare motionCompensation sets base class fields
    • 14 hours ago Steve Borho dpb: fix picture re-use bug introduced in bc53f6dcda7f
    • 23 hours ago Steve Borho entropy: remove unused flush function
    • 17 hours ago Steve Borho entropy: remove redundant !isIntra() check
    • 18 hours ago Steve Borho quant: use bisLuma
    • 18 hours ago Steve Borho quant: use x264 style lambda naming scheme
    • 15 hours ago Steve Borho dpb: cleanup prepareEncode()
    • 18 hours ago Steve Borho rdcost: settle on x264 style lambda naming scheme
    • 18 hours ago Steve Borho rdcost: change input type of setC*DistortionWeight to uint16_t
    • 18 hours ago Steve Borho rdcost: move zeroPel buffer to TEncSearch, its only user
    • 18 hours ago Steve Borho rdcost: keep psy-rd scale factor as an int (0..512), remove psyRdEnabled()
    • 15 hours ago Steve Borho common: remove unused SCALING_LIST_PARAMETER enum
    • 15 hours ago Steve Borho TComRom: nits
    • 19 hours ago Steve Borho entropy: nits
    • 19 hours ago Steve Borho entropy: use bIsLuma in codeCoeffNxN()
    • 17 hours ago Steve Borho slice: allow loop filter boundary flag to be analyzed
    • 19 hours ago Steve Borho quant: use bIsLuma args instead of using a subset of TextType
    • 19 hours ago Steve Borho TComTrQuant to Quant
    • 25 hours ago Steve Borho scaling: use anonymous enums for class constants
    • 34 hours ago Deepthi Nandakumar predict: predict.cpp/.h defined, hungarian function names cleaned up, warnings removed
Version 1.2.391
  • Release Date: Jul 31, 2014
    • 9 hours ago Steve Borho param: disable range checks for psy-rdoq while we tune it
    • 10 hours ago Steve Borho defs: remove DISTORTION_PRECISION_ADJUSTMENT, fix bug in 2e22ea6ec4bc
    • 5 days ago Sumalatha Polureddy psy-rdoq: implementation of psy-rdoq (highly experimental)
    • 10 hours ago Steve Borho quant: minor cleanups
    • 12 hours ago Steve Borho quant: rename absLevel to level, remove diffLevel
    • 12 hours ago Steve Borho quant: readability nit
    • 12 hours ago Steve Borho quant: pass curCostSig to getCodedLevel as an integer
    • 13 hours ago Steve Borho quant: return signal cost from getRateLast(), do not include lambda
    • 13 hours ago Steve Borho quant: don't bother with extra temp variables
    • 16 hours ago Min Chen TEncSearch: fix output non-determination bug
    • 16 hours ago Steve Borho quant: getICRate() and getICRateCost() do not need to be class methods
    • 16 hours ago Steve Borho quant: hoist lambda multiplication out of getICRateCost()
    • 17 hours ago Steve Borho quant: inline getRateSigCoef(), getRateSigCoeffGroup() and getICost()
    • 19 hours ago Steve Borho quant: improve documentation and code clarity, no logic change
    • 21 hours ago Steve Borho scaling: factor prec into scalingBits
    • 22 hours ago Steve Borho scaling: cleanup
    • 22 hours ago Steve Borho move quant defines from TComRom.h to CommonDef.h; they are not table sizes
    • 22 hours ago Steve Borho typedef: remove redundant ChromaFormat enums
    • 22 hours ago Steve Borho defs: cleanup CommonDef.h and TypeDef.h again
    • 22 hours ago Steve Borho scaling: further simplify scaling list calculation
    • 22 hours ago Steve Borho defs: remove FULL_NBIT, no change to logic
    • 22 hours ago Steve Borho scaling: replace pow(2.0, -2.0 * transformShift) with shift adjustment
    • 23 hours ago Steve Borho quant: use ScanType enum as type for scanType, other cleanups
    • 34 hours ago Steve Borho quant: move enums for scan order to TComRom together with the tables
    • 24 hours ago Aarthi Thirumalai rc: fix crash in 2 pass for faster preset (ultrafast - faster)
    • 24 hours ago Steve Borho hrd: add a comment about when an APS is required, update SEI
    • 24 hours ago Steve Borho sei: fix byte alignment following HRD SEI
    • 36 hours ago Steve Borho cmake: enable full warnings for remaining TLibEncoder files
    • 38 hours ago Steve Borho quant: delay getTUEntropyCodingParameters() until known necessary
    • 38 hours ago Albert Wang frameencoder: fix for crash in frameencoder when AccessUnitDelimiter is on
    • 36 hours ago Steve Borho remove sign compare warnings from TLibEncoder, cleanup SAO
    • 37 hours ago Steve Borho rc,weight: nits
    • 37 hours ago Steve Borho rc: call getEstimatedPictureCost() from API thread, prevent race hazards
    • 39 hours ago Steve Borho dpb: cleanup
    • 39 hours ago Steve Borho slice: move m_bReferenced to picsym and rename for clarity
    • 44 hours ago Steve Borho frameencoder: nits
    • 44 hours ago Steve Borho frameencoder: re-order compressFrame() to avoid dependency problem with HRD
    • 2 days ago Satoshi Nakagawa search: separate bSkipRes == true path
    • 2 days ago Aarthi Thirumalai slicetype: compute frame satd cost for zero latency
    • 2 days ago Steve Borho scaling: remove m_refMatrixId; it was only used to return an int
    • 2 days ago Steve Borho scaling: use simple ints for small counters and int32_t for scales and coeffs
    • 2 days ago Steve Borho cmake: ignore msvc warnings about forcing value to bool 'true' or 'false'
    • 2 days ago Steve Borho quant: fix for tskip
Version 1.2.363
  • Release Date: Jul 30, 2014
    • 41 hours ago Steve Borho quant: nits
    • 41 hours ago Steve Borho quant: inline transformSkip
    • 41 hours ago Steve Borho quant: inline invTransformSkip
    • 41 hours ago Steve Borho quant: simplify dequant, remove redundant invQuantScales array
    • 40 hours ago Steve Borho level: signal level None if lossless coding - we have no bitrate control
    • 46 hours ago Steve Borho scaling: remove duplicate square-block size enums
    • 46 hours ago Steve Borho scaling: move global arrays into class statics, rename
    • 2 days ago Steve Borho quant: move quant/dequant/errScale buffer maintenance to scaling list class
    • 2 days ago Steve Borho quant: remove x prefixes from remaining methods; no logic change
    • 2 days ago Steve Borho quant: replace xGetIEPRate() with a constant
    • 2 days ago Steve Borho quant: remove x prefixes from scaling list methods
    • 2 days ago Steve Borho quant: use rem consistently for variables holding qp % 6
    • 2 days ago Steve Borho quant: remove default curUseRDOQ argument value
    • 2 days ago Steve Borho quant: remove default useTransformSkip argument value
    • 2 days ago Steve Borho quant: more coding style cleanup, update argument list in xGetCodedLevel comment
    • 2 days ago Steve Borho quant: QpParam class to struct
    • 2 days ago Steve Borho quant: re-order header and improve comments
    • 2 days ago Steve Borho quant: remove get/set methods for scaling list enable flag
    • 2 days ago Steve Borho quant: remove access methods for the class's own members
    • 2 days ago Steve Borho api: expose scaling lists and psy-rdoq as a CLI/API options
Version 1.2.327
  • Release Date: Jul 28, 2014
    • 13 hours ago Steve Borho quant: don't pass m_resiDctCoeff to member methods
    • 39 hours ago Steve Borho param: move all consistency param tweaks into Encoder::configure()
    • 13 hours ago Steve Borho quant: cleanup initialization; catch malloc failures. rename m_tmpCoeff
    • 36 hours ago Steve Borho quant: nits
    • 36 hours ago Steve Borho quant: cleanups
    • 38 hours ago Steve Borho quant: cleanup signBitHidingHDQ()
    • 39 hours ago Steve Borho quant: simplify setQPforQuant
    • 39 hours ago Steve Borho quant: remove dead comment
    • 46 hours ago Steve Borho quant: do not malloc the EstBitsSbac structure, pass by reference
    • 47 hours ago Steve Borho white-space and brace nits
    • 2 days ago Steve Borho search: simplify verbage, fix white-space, no logic change
    • 2 days ago Steve Borho search: move all init code into TEncSearch::initSearch()
    • 44 hours ago Steve Borho framefilter: initialize saoRowDelay just once
    • 44 hours ago Steve Borho frameencoder: remove getSAO() access method; delay allocation of SAOParam
    • 45 hours ago Steve Borho move slice setup together into one function, reorder functions for clarity
    • 45 hours ago Steve Borho slice: remove redundant initSlice(), m_numRefIdx[] is set by DPB unconditionally
    • 46 hours ago Steve Borho slice: move coloc/tmvp flag setting all to one place
    • 39 hours ago Steve Borho api: improve documentation of x265_encoder_encode
    • 14 hours ago Steve Borho analysis: cleanup early-skip
    • 18 hours ago Steve Borho cudata: nits
    • 19 hours ago Steve Borho entropy: nit
    • 19 hours ago Steve Borho encoder: remove unused m_maxRefPicNum and m_maxNumOffsetsPerPic
    • 46 hours ago Steve Borho rest: better document configuration options of --vbv-init
    • 2 days ago Aarthi Thirumalai rc: fix crashes in vbv with 2 pass

Version 1.2.305
  • Release Date: Jul 26, 2014
    • 24 hours ago Steve Borho encoder: remove redundant m_pad array
    • 24 hours ago Steve Borho encoder: fix uninitialized variable use reported by valgrind
    • 24 hours ago Steve Borho nits
    • 25 hours ago Steve Borho level: move all profile set/determine logic into one file, fixes for MSP
    • 26 hours ago Steve Borho cli: call x265_param_apply_profile() after applying file-reader params
    • 26 hours ago Steve Borho cli: nit
    • 2 days ago Satoshi Nakagawa qtLayer in reverse order
    • 3 days ago Satoshi Nakagawa refine partition size related
    • 27 hours ago Steve Borho param: log multi-pass state
    • 31 hours ago Steve Borho rc: nits
    • 31 hours ago Steve Borho rc: instrument all error conditions in initPass2()
    • 31 hours ago Steve Borho bitcost: nits
    • 31 hours ago Steve Borho rest: better document that -I N -i N does not disable scenecut detection
    • 33 hours ago Steve Borho Merge with stable
    • 33 hours ago Steve Borho bitcost: increase size of cost arrays, for pessimal situationsstable
    • 33 hours ago Steve Borho bitcost: fix bitcost() function, broken 13 months ago by 07015bbe306bstable
    • 36 hours ago Steve Borho cli: reorder argument help for clarity, add more sections
    • 38 hours ago Aarthi Thirumalai rc: add cli options for multi-pass rate control
    • 45 hours ago Gopu Govindaswamy port TComLoopFilter Functions into /common/deblock
Version 1.2.288
  • Release Date: Jul 25, 2014
    • 2 hours ago Steve Borho entropy: reorder methods in general encode order for clarity, no logic change
    • 2 hours ago Steve Borho entropy: remove ambiguity about timing info signaling
    • 3 hours ago Steve Borho entropy: nits
    • 3 hours ago Steve Borho entropy: do not code a scaling list in PPS
    • 11 hours ago Aarthi Thirumalai rc: calculate mvBits, coefBits per Cu for 2 pass in skip mode as well
    • 11 hours ago Aarthi Thirumalai rc: compute average of qpaRc, qpAq per frame even in 2 pass
    • 8 hours ago Steve Borho pred: remove default motionCompensation argument values
    • 8 hours ago Steve Borho analysis: eoln fixup
    • 21 hours ago Deepthi Nandakumar TComPrediction: fix MC error caused by previous patch
    • 22 hours ago Deepthi Nandakumar TComPrediction: remove unused parameter warning
    • 24 hours ago Deepthi Nandakumar analysis: whitespace nit
    • 24 hours ago Steve Borho rest: tpyo
Version 1.2.267
  • Release Date: Jul 23, 2014
    • 53 minutes ago Steve Borho rest: move profile, level, and tier into their own section
    • 57 minutes ago Deepthi Nandakumar analysis: add in extra X265_CHECK
    • 81 minutes ago Gopu Govindaswamy analysis: setQPforQuant in checkIntraInter to fix the hash mismatch at rd=5&6
    • 2 hours ago Deepthi Nandakumar TComPrediction: white space nits, remove isPSlice (next series)
    • 16 hours ago Deepthi Nandakumar TComPrediction: remove redundant colorspace information
    • 19 hours ago Deepthi Nandakumar TComPrediction: save CU data (partAddr, width, height) as member fields
    • 44 hours ago Deepthi Nandakumar TComPrediction: remove TComDataCU as pointer to private functions
    • 12 hours ago Steve Borho param: re-order code to avoid unreachable code warnings in 10bit builds
    • 12 hours ago Steve Borho remove very obsolete regression batch files
    • 12 hours ago Steve Borho level: fix GCC signed/unsigned comparison
    • 28 hours ago Steve Borho rest: update --profile, --level, --high-tier docs
    • 30 hours ago Steve Borho level: add --high-tier and auto-configure VBV if --crf N --level M (refs #61)
    • 19 hours ago Aarthi Thirumalai rc: bug fixes in 2 pass ABR
    • 27 hours ago Steve Borho rc: prevent gcc shadow warning about int terminate and terminate()
    • 27 hours ago Steve Borho analysis: don't initialize members of base class in constructor
    • 30 hours ago Steve Borho param: fix profile warnings in x265_param_apply_profile()
    • 30 hours ago Steve Borho cli: expose x265_param_apply_profile() via --profile
    • 31 hours ago Steve Borho level: nit
    • 31 hours ago Steve Borho level: move m_vps.maxDecPicBuffering logic into one function
    • 32 hours ago Steve Borho ptl: move enums from TypeDef.h to slice.h
    • 32 hours ago Steve Borho slice: add ProfileTierLevel instance to VPS
    • 32 hours ago Steve Borho level: fix typo
    • 5 days ago Steve Borho fix for GCC warning about loop bounds
    • 32 hours ago Steve Borho entropy: settle on Entropy class name and member names
    • 32 hours ago Steve Borho cmake: tabs to spaces
    • 33 hours ago Steve Borho cmake: only create SONAME shared lib on POSIX systems (closes #62)
    • 34 hours ago Steve Borho rc: prevent deadlock on encoder flush or close
    • 2 days ago Aarthi Thirumalai rc: fix warnings in ratecontrol
    • 5 days ago Aarthi Thirumalai rc: generate cutree offsets for 2nd pass from stats
    • 4 days ago Aarthi Thirumalai rc: fix sliceType of 2nd pass from prev pass stats
    • 4 days ago Aarthi Thirumalai rc: add 2 pass logic in rateEstimateQscale
    • 2 days ago Satoshi Nakagawa trquant: store QpParam for each component
    • 3 days ago Satoshi Nakagawa scan order tables
    • 38 hours ago Deepthi Nandakumar TComWeightPrediction: initialising pointers to NULL
    • 38 hours ago Deepthi Nandakumar TComWeightPrediction: add braces to prevent errors for an optional else statement
Version 1.2.239
  • Release Date: Jul 21, 2014
    • 38 hours ago Michel Zou cmake: fix dll installation directory on Windows
    • 26 hours ago Min Chen use macro HADDD to improve AMD performance
    • 8 hours ago Ashok Kumar Mishra TComDataCU: replaced functions with tables
    • 12 hours ago Ashok Kumar Mishra TComDataCU: remove redundant functions
    • 16 hours ago Santhoshini Sekar rc: update ratecontrol stats in every frame, avoid frame parallelism lag in abr
    • 14 hours ago Steve Borho split ScalingList class into its own header and source file
    • 15 hours ago Steve Borho a copy of POC is cached in the Frame struct
    • 15 hours ago Steve Borho frame,cu: remove getSlice()
    • 15 hours ago Steve Borho picsym: remove getSlice()
    • 15 hours ago Steve Borho move PPS, SPS, Slice and ScalingList to common/
    • 17 hours ago Steve Borho slice: remove unused initWpScaling()
    • 17 hours ago Steve Borho slice: remove getWpScaling()
    • 17 hours ago Steve Borho slice: make xGetRefPic a PicList method
    • 17 hours ago Steve Borho slice: do not initialize fields that are always written unconditionally
    • 17 hours ago Steve Borho encoder: do some slice initializations only once
    • 17 hours ago Steve Borho entropy: slice->m_colFromL0Flag is a bool
    • 18 hours ago Steve Borho dpb: simplify check for m_bCheckLDC; make note for future optimization
    • 18 hours ago Steve Borho dpb: remove check for B frame without L1
    • 19 hours ago Steve Borho slice: remove m_bLMvdL1Zero
    • 19 hours ago Steve Borho entropy: fix warning reported by llvm
    • 20 hours ago Steve Borho wpp: generate stream size array as coded (in bytes) and pre-calculate max offset
    • 20 hours ago Steve Borho slice: remove substream array, move to frame encoder
    • 20 hours ago Steve Borho nit
    • 20 hours ago Steve Borho replace slice->m_refPicList[][]->getPOC() with slice->m_refPOCList[][]
    • 20 hours ago Steve Borho slice: merge setRefPOCList() into setRefPicList()
    • 21 hours ago Steve Borho slice: remove getRefPOC
    • 21 hours ago Steve Borho dpb: perform setRefPOCList() directly after setRefPicList()
    • 21 hours ago Steve Borho slice: remove getRefPic(l,r)
    • 21 hours ago Steve Borho slice: remove setNumRefIdx
    • 21 hours ago Steve Borho slice: remove set/get for m_colRefIdx and m_sliceCurEndCUAddr
    • 21 hours ago Steve Borho slice: remove getMvdL1ZeroFlag/setMvdL1ZeroFlag
    • 21 hours ago Steve Borho slice: remove getColFromL0Flag()
    • 22 hours ago Steve Borho slice: remove getNumRefIdx() access method
    • 22 hours ago Steve Borho slice: removed unused setRefPic() and setRefPOC() methods
    • 22 hours ago Steve Borho slice: remove unused getNumRpsCurrTempList() method
    • 22 hours ago Steve Borho slice: remove get/set methods for slice QP and lastIDR
    • 23 hours ago Steve Borho slice: remove get/set methods for m_bReferenced
    • 23 hours ago Steve Borho slice: remove get/set methods for m_sliceType
    • 23 hours ago Steve Borho nit
    • 23 hours ago Steve Borho slice: remove m_cabacInitFlag, it was write-only
    • 24 hours ago Steve Borho slice: remove m_numEntryPointOffsets, it was write-only
    • 24 hours ago Steve Borho slice: remove slice bit counter
    • 24 hours ago Steve Borho slice: remove get/set methods for m_poc
    • 24 hours ago Steve Borho nits
    • 24 hours ago Steve Borho slice: remove m_sliceSegmentBits and get/set methods for max-merge
    • 25 hours ago Steve Borho slice: remove m_saoEnabledFlag, m_saoEnabledFlagChroma - use saoParams
    • 26 hours ago Steve Borho sao: combine if(){} expressions
    • 26 hours ago Steve Borho sao: (bool == 1? ? true : false -> wat?
    • 26 hours ago Steve Borho rc: fix vbvBufferSize calculation
    • 26 hours ago Steve Borho slice: nits
    • 26 hours ago Steve Borho slice: cleanup m_nalUnitType and helper methods
    • 26 hours ago Steve Borho slice: remove get/set methods for m_pic
    • 26 hours ago Steve Borho slice: temporal MVP was never disabled
    • 26 hours ago Steve Borho rc: HRD must be initialized prior to calling init()
    • 27 hours ago Steve Borho slice: remove unused deblock data, directly use status/offsets from PPS
    • 27 hours ago Steve Borho slice: remove unused slice QP deltas
    • 27 hours ago Steve Borho nits
    • 27 hours ago Steve Borho slice: we only use a slice-local RPS (none signaled in SPS)
    • 27 hours ago Steve Borho slice: TComReferencePictureSet -> RPS
    • 27 hours ago Steve Borho slice: hoist constructor to header (it will be removed in later commits)
    • 27 hours ago Steve Borho slice: remove get/set methods for SPS and PPS, make pointers const
    • 28 hours ago Steve Borho nit
    • 28 hours ago Steve Borho slice: rename wpScalingParam -> WeightParam
    • 28 hours ago Steve Borho pps: simplify deblocking signaling, disable slice override
    • 38 hours ago Steve Borho nits

Version 1.2.174
  • Release Date: Jul 18, 2014
    • 15 hours ago Steve Borho encoder: make SPS and PPS top-level encoder singletons
    • 16 hours ago Steve Borho pps: we never very chroma QP per picture, so remove bSliceChromaQpFlag
    • 16 hours ago Steve Borho pps: move deblocking logic together, fix a couple signaling bugs
    • 17 hours ago Steve Borho pps: class to struct, remove m_ prefix, unify naming scheme
    • 18 hours ago Steve Borho pps: remove set/get methods and hard-coded members
    • 19 hours ago Steve Borho slice: white-space nits
    • 19 hours ago Steve Borho vps: remove constructor, those values are always initialized
    • 16 hours ago Steve Borho sps: fix check statement
    • 19 hours ago Steve Borho sps: class to struct, remove m_ prefixes
    • 19 hours ago Steve Borho sps: remove set/get methods, remove members with hard-coded values
    • 21 hours ago Steve Borho vps: class to struct and other nits
    • 21 hours ago Steve Borho pps: remove hard-coded getLog2ParallelMergeLevelMinus2(0)
    • 21 hours ago Steve Borho remove hard-coded pps->getSliceHeaderExtensionPresentFlag()
    • 21 hours ago Steve Borho nits
    • 21 hours ago Steve Borho slice: remove redundant/unused scaling lists from SPS, PPS, and Slice
    • 22 hours ago Steve Borho slice: nits
    • 22 hours ago Steve Borho slice: remove trivial set/get methods, move all relavent functions to the class
    • 22 hours ago Steve Borho slice: remove unsignaled m_temporalLayerNonReferenceFlag
    • 24 hours ago Steve Borho rest: fix crf min/max option names
    • 26 hours ago Steve Borho slice: start to clean up scaling list struct
    • 27 hours ago Steve Borho slice: convert Window to struct, simplify names
    • 28 hours ago Steve Borho rest: fix documented default AQ mode
    • 38 hours ago Steve Borho bitstream: change writeFlag() to take a bool to avoid int conversions
    • 39 hours ago Steve Borho vui: prune set/get methods and unsignaled fields
    • 39 hours ago Steve Borho slice: finish cleanup of TComReferencePictureSet
    • 39 hours ago Steve Borho slice: remove unused PPS ID
    • 40 hours ago Steve Borho brace nits
    • 40 hours ago Steve Borho slice: Cleanup TComReferencePictureSet, prune the notion of long-term-pictures
    • 41 hours ago Steve Borho entropy: nits
    • 2 days ago Deepthi Nandakumar port TEncCU functions into analysis structure
Version 1.2.152
  • Release Date: Jul 17, 2014
    • 12 hours ago Steve Borho search: assert to X265_CHECK
    • 13 hours ago Min Chen replace sse_sp(residual, ZERO) by ssd_s(residual)
    • 29 hours ago Santhoshini Sekar rc: define function for updating rate control statistics in processRowEncoder
    • 31 hours ago Sumalatha Polureddy fix for crash when scalingList feature is enabled
    • 14 hours ago Steve Borho entropy: remove obsolete check macro
    • 14 hours ago Steve Borho slice: fix warning
    • 14 hours ago Steve Borho slice: remove unused m_useDF
    • 14 hours ago Steve Borho slice: remove more hard-coded fields
    • 14 hours ago Steve Borho defs: remove more dead definitions
    • 15 hours ago Steve Borho hrd: remove *Minus1 from member names and semantics, remove m_ prefixes
    • 15 hours ago Steve Borho hrd: remove set/get methods of TComHRD, remove unused members
    • 16 hours ago Steve Borho hrd: remove four unused fields
    • 16 hours ago Steve Borho hrd: CPB count is always 1
    • 16 hours ago Steve Borho hrd: inline HrdLayerInfo, remove second array dimension (was for nal/vcl)
    • 16 hours ago Steve Borho hrd: we always signal NAL HRD but not VCL and not sub-pic
    • 16 hours ago Steve Borho slice: simplify TimingInfo to a struct
    • 16 hours ago Steve Borho slice: nits
    • 17 hours ago Steve Borho change ProfileTierLevel to a struct, no m_ prefixes
    • 17 hours ago Steve Borho ptl: remove hard-coded m_profileSpace
    • 17 hours ago Steve Borho entropy: de-hungarian codePredWeightTable(), fix comments (remove redundants)
    • 17 hours ago Steve Borho entropy: nits
    • 18 hours ago Steve Borho vps: remove m_maxLatencyIncrease, it was always 0
    • 18 hours ago Steve Borho vps: remove remaining set/get methods
    • 18 hours ago Steve Borho vps: remove now unused m_layerIdIncludedFlag
    • 18 hours ago Steve Borho vps: remove unused m_maxNuhReservedZeroLayerId, m_numOpSets
    • 18 hours ago Steve Borho sps: hard-code an SPS ID of 0, we only ever use one SPS
    • 18 hours ago Steve Borho vps: remove m_VPSId, signal hard-coded 0
    • 18 hours ago Steve Borho entropy: remove broken X265_CHECK
    • 18 hours ago Steve Borho frameencoder: nits
    • 18 hours ago Steve Borho remove unused multiple HRD parameter sets from VPS class
    • 18 hours ago Steve Borho defs: remove some dead definitions, cleanup a comment
    • 18 hours ago Steve Borho slice: remove unused m_interRPSPrediction from TComReferencePictureSet
    • 19 hours ago Steve Borho slice: simplify max AMP level determination
    • 19 hours ago Steve Borho encoder: remove redundant member variables
    • 19 hours ago Steve Borho Remove MAX_TLAYER and stop pretending we can code multiple temporal layers
    • 20 hours ago Steve Borho entropy: nits
    • 2 days ago Sumalatha Polureddy compress: fix for sa8dcost storage in rdlevel 1, 0
    • 31 hours ago Deepthi Nandakumar TEncCU: remove useless field
Version 1.2.76
  • Release Date: Jul 13, 2014
    • 23 hours ago Steve Borho encoder: nits
    • 23 hours ago Steve Borho encoder: remove unused variables
    • 29 hours ago Steve Borho sei: emit SEI describing the encoder and encode options
    • 29 hours ago Steve Borho frameencoder: nit
    • 32 hours ago Steve Borho param: fix typo in d1d5b377294e
    • 32 hours ago Steve Borho rc: nits
    • 32 hours ago Steve Borho cmake: nit - about target arch strings
    • 44 hours ago Santhoshini Sekar frameencoder: store reflagrows as a member variable
    • 43 hours ago Aarthi Thirumalai rc: init RC data for second pass in a multi pass encode
    • 43 hours ago Aarthi Thirumalai param: keep the total frame count of the input video in param.
    • 43 hours ago Aarthi Thirumalai rc: store final cu counts in frameStats as a percentage of total # cus
    • 34 hours ago Steve Borho param: use dashes consistently in param2string
    • 43 hours ago Aarthi Thirumalai param: add more param options to print as string
    • 43 hours ago Aarthi Thirumalai common: introduce x265_slurp_file, reads data from file to string buffer.
    • 35 hours ago Steve Borho level: in CRF/CQP encodes, allow user specifed level to increase stream level
    • 44 hours ago Deepthi Nandakumar cu: fixing int-to-bool compile warnings
    • 47 hours ago Steve Borho cu: directly use param fields for lossless coding options
    • 47 hours ago Steve Borho cu: nits
    • 2 days ago Steve Borho cli: retrieve params from the encoder, in case logging flags have changed
    • 2 days ago Satoshi Nakagawa zero stride for zeroPel[]
    • 2 days ago Steve Borho search: make RDCost and TComTrQuant member instances of TEncSearch
    • 2 days ago Steve Borho derive TEncCu from TEncSearch
    • 2 days ago Deepthi Nandakumar compress: save CABAC context for rd = 2 after inter/intra encodes
    • 2 days ago Deepthi Nandakumar compress: save CABAC context after intra decision.
    • 2 days ago Deepthi Nandakumar compress: save CABAC context after inter/merge decision
    • 2 days ago Deepthi Nandakumar compress: save CABAC context in merge modes for lower RD levels.
    • 2 days ago Steve Borho entropy: clarify the SBac's bit counting mode
    • 2 days ago Steve Borho framefilter: rename row0 coder
    • 2 days ago Steve Borho cturow: rename m_rdGoOnSbacCoder to m_sbacCoder; there is only one
    • 2 days ago Steve Borho frameencoder: remove unnecessary include of TEncSearch.h
    • 2 days ago Steve Borho ppa: cleanup event names and remove uninteresting events
    • 2 days ago Steve Borho cu, search: remove redundant m_rdGoOnSbacCoder pointer
    • 2 days ago Steve Borho bitstream: hoist trivial methods to the header so they may be easier inlined
    • 2 days ago Steve Borho bitstream: convert macros into inline members for better compile-time checking
    • 2 days ago Steve Borho bring SyntaxElementWriter into common/, remove trace file macros
Version 1.2.45
  • Release Date: Jul 11, 2014
    • 13 hours ago Steve Borho entropy: nits
    • 13 hours ago Steve Borho entropy: re-order methods again for better clarity, make many private
    • 14 hours ago Steve Borho entropy: remove include of TComSampleAdaptiveOffset.h
    • 14 hours ago Steve Borho TComTrQuant: rename estBitsSbacStruct to EstBitsSbac
    • 14 hours ago Steve Borho entropy: nits
    • 14 hours ago Steve Borho entropy: hoist trivial functions for later optimizations
    • 14 hours ago Steve Borho merge CABAC into SBac class
    • 14 hours ago Steve Borho entropy: convert initSection into a constructor, since it was used as such
    • 15 hours ago Steve Borho TEncSearch: nits
    • 15 hours ago Steve Borho entropy: replace SBac and TEncSearch methods with TURecurse methods
    • 15 hours ago Steve Borho entropy: remove m_ prefix from struct TURecurse members
    • 34 hours ago Satoshi Nakagawa use std::swap() for readability
    • 18 hours ago Steve Borho search: remove braces around single line expressions, improve code flow
    • 18 hours ago Steve Borho Merge with stable
    • 18 hours ago Steve Borho Added tag 1.2 for changeset d6257335c537stable
    • 7 days ago Steve Borho csv: add slice-type stats to encode summary, avoid dup copies of summary string
    • 19 hours ago Steve Borho entropy: remove temporary helper function
    • 19 hours ago Steve Borho entropy: nits
    • 20 hours ago Steve Borho cu: remove get/set methods for its own member variable
    • 20 hours ago Steve Borho cu: remove more set methods
    • 20 hours ago Steve Borho cu: nit
    • 20 hours ago Steve Borho cu: assign m_rdGoOnSbacCoder without helper methods
    • 20 hours ago Steve Borho cu: pass bit counting flag as argument to encodeCU
    • 20 hours ago Steve Borho ctu: inline setThreadLocalData() for better clarity
    • 21 hours ago Steve Borho frameencoder: remove rarely used getRowCoder() and getBufferSBac()
    • 21 hours ago Steve Borho frameencoder: remove barely use helper function getRDGoOnSbacCoder
    • 21 hours ago Steve Borho ctu: rename m_sbacCoder to m_rowEntropyCoder
    • 21 hours ago Steve Borho entropy: allocate RDO coders as dual-dimensioned array
    • 22 hours ago Steve Borho framefilter: rename m_rdGoOnSbacCoder to m_sbacCoder - there is only one
    • 23 hours ago Steve Borho entropy: make m_cabac member of SBac class a non-pointer instance
    • 25 hours ago Steve Borho entropy: remove now unused Entropy class
    • 37 hours ago Steve Borho remove Entropy from TEncCu, TEncSearch, and CTURow
    • 37 hours ago Steve Borho frameencoder: remove most traces of the Encoder class
    • 26 hours ago Steve Borho frameencoder: do not pass member variable to its own function
    • 36 hours ago Steve Borho entropy: remove m_slice member from SBac class
    • 26 hours ago Steve Borho frameencoder: do not use Entropy object to code stream headers
    • 37 hours ago Steve Borho sao: remove redundant braces and other nits
    • 37 hours ago Steve Borho remove m_entropyCoder from FrameFilter and SAO code
    • 38 hours ago Steve Borho entropy: remove unnecessary API differences between Entropy and SBac
    • 39 hours ago Steve Borho entropy: reorder SBac methods for more clarity
    • 40 hours ago Steve Borho entropy: remove default argument to encodeIntraDirModeLuma()
    • 39 hours ago Steve Borho entropy: fix msvc warnings about signed/unsigned operations
    • 40 hours ago Steve Borho entropy: hoist all non-trivial Entropy methods into SBac
    • 40 hours ago Steve Borho entropy: remove all data members except m_entropyCoder
    • 41 hours ago Steve Borho port TEncEntropy and TEncSbac into internal Entropy and SBac classes
    • 35 hours ago Deepthi Nandakumar Merge with stable
    • 35 hours ago Deepthi Nandakumar TEncCU: bug fix introduced in commit 8a5b02464813stable 1.2
    • 36 hours ago Deepthi Nandakumar Merge with stable
    • 38 hours ago Min Chen TEncCU: save CABAC context missing after mergestable
    • 38 hours ago Min Chen TEncCU: save CABAC context missing after merge
    • 3 days ago Satoshi Nakagawa quant: returns numSig instead of absSum and lastPos
Version 1.1.258
  • Release Date: Jul 9, 2014
    • 23 hours ago Steve Borho Merge with default (prepare for 1.2)
    • 23 hours ago Steve Borho frameencoder: nitsdefault
    • 24 hours ago Steve Borho cu: do not copy a QP buffer into itself - prevents valgrind warning
    • 24 hours ago Steve Borho cu: cleanup and simplify initCU
    • 26 hours ago Steve Borho rc: allocate HRD SEI structures on demand, fix RCE vtable/memset bug
    • 4 days ago Min Chen avoid VS2008 and MinGW ambiguous pow() build warning
    • 5 days ago Deepthi Nandakumar frameencoder: remove warning about ambiguous pow function
    • 5 days ago Deepthi Nandakumar lambda-file: file close before return
Version 1.1.252
  • Release Date: Jul 4, 2014
    • 10 hours ago Steve Borho y4m: avoid implicit size_t to int conversion, avoid more MSVC warnings
    • 10 hours ago Steve Borho encoder: avoid int64_t to int conversions, avoid more MSVC warnings
    • 11 hours ago Steve Borho param: avoid spurious MSVC warning about buf being possibly uninitialized
    • 11 hours ago Steve Borho ratecontrol: fix cast of cutree offset, avoids MSVC warning
    • 11 hours ago Steve Borho frameencoder: match up integer argument types to avoid MSVC warning
    • 11 hours ago Steve Borho param: avoid spurious gcc warning about toksave possibly being uninitialized
    • 18 hours ago Steve Borho lamba-file: make improper number of constants a fatal error
    • 18 hours ago Steve Borho lambda-file: detect and report if lamda file containts too many values
    • 18 hours ago Min Chen simplify: getLumaIntraDir()[x] -> getLumaIntraDir(x)
    • 18 hours ago Steve Borho api: allow lambda tables to be user-specified via a text file
    • 21 hours ago Steve Borho rc: fix gcc warnings
    • 3 days ago Aarthi Thirumalai rc: write frame stats and cu stats to file in first pass
    • 7 days ago Aarthi Thirumalai rc: compute inter, intra and skip cus per frame for the first pass
    • 2 days ago Aarthi Thirumalai rc: move structure FrameStats to FrameEncoder class
    • 33 hours ago Satoshi Nakagawa add primitives.nquant for RDOQ
    • 30 hours ago Kavitha Sampath frame: initialize recon to avoid SAO read of uninitialized pixels beyond picture end
    • 34 hours ago Steve Borho encoder: fix --no-wpp behavior, keep TLD selection logic in one place
    • 42 hours ago Steve Borho input: streamline control logic of threaded file readers
    • 35 hours ago Steve Borho TEncEntropy: inline the only one caller of encodeInterDirPU
    • 36 hours ago Steve Borho TEncEntropy: hoist encodePredMode() after moving I-slice check to callers
    • 36 hours ago Steve Borho TEncEntropy: reorder nit
    • 36 hours ago Steve Borho TEncEntropy: hoist encodeSkipFlag() after moving I-slice check to callers
    • 36 hours ago Steve Borho TEncEntropy: hoist encodeSliceHeader() to header after simplifying
    • 36 hours ago Steve Borho TEncEntropy: nit
    • 36 hours ago Steve Borho TEncEntropy: hoist trivial functions to header for possible inlining
Version 1.1.218
  • Release Date: Jul 2, 2014
    • 3 hours ago Steve Borho TEncSearch: nit
    • 3 hours ago Steve Borho TComDataCU: remove redundant refIdx check from fillMvpCand
    • 3 hours ago Steve Borho Merge with stable
    • 3 hours ago Steve Borho TEncSearch: limit AMVP loop bounds in xCheckBestMVPstable
    • 13 hours ago Ashok Kumar Mishra remove redundant memset
    • 16 hours ago Satoshi Nakagawa split rate calculation functions to luma and chroma to simplify luma path
    • 4 days ago Satoshi Nakagawa fix emms: move selectLambda() into xRateDistOptQuant() and issue emms before it
    • 6 days ago Aarthi Thirumalai rc: rename texBits to coeffBits in RateControlEntry structure to maintain uniformity.
    • 29 hours ago Steve Borho encoder: RDOQ is only applicable at rd levels 4, 5, and 6
    • 5 days ago Aarthi Thirumalai rc: accumulate mv bits, coeff bits per frame
    • 39 hours ago Aarthi Thirumalai TComDataCU: remove warnings for shadowed declaration
    • 44 hours ago Albert Wang frameencoder: fix for error in VPS when AccessUnitDelimeter is turned on
    • 29 hours ago Steve Borho rest: document ipratio and pbratio parameters
    • 6 days ago Satoshi Nakagawa cli: add --ipratio and --pbratio
    • 10 days ago Steve Borho cli: use consistent bool naming convention
    • 10 days ago Steve Borho entropy: nit
    • 10 days ago Satoshi Nakagawa input: fix race condition
    • 4 days ago Min Chen reduce condition jmp in codeCoeffNxN
    • 4 days ago Min Chen improve count_nonzero by SSSE3
    • 4 days ago Sumalatha Polureddy psyrd: fix for inconsistent output
    • 4 days ago Deepthi Nandakumar framefilter: disable warning
    • 7 days ago Min Chen pass TLD into class FrameFilter
    • 7 days ago Satoshi Nakagawa refine intra tskip related.
    • 6 days ago Aarthi Thirumalai rc: initalize states to hold frame statistics used in two pass
    • 6 days ago Sumalatha Polureddy psy-rd: implement psy-rd in rdlevel=4,3 and 2
    • 6 days ago Ashok Kumar Mishra remove PCM mode support
    • 7 days ago Satoshi Nakagawa fix xGetIntraBitsQTChroma() for 4:2:2 [CHANGES OUTPUT 4:2:2 with tskip]
    • 7 days ago Deepthi Nandakumar psy-rd chroma: Adjust chroma lambda based on QP offsets.
    • 7 days ago Deepthi Nandakumar psy-rd chroma: add chroma offset tables for lambda calculations
    • 7 days ago Deepthi Nandakumar Chroma QP Offset: increase chroma QP when psy-rd is enabled.
    • 7 days ago Min Chen move calcSaoStatsRowCus_BeforeDblk into encode loop
    • 8 days ago Ashok Kumar Mishra TEncCU:fix for memory leak
    • 11 days ago Praveen Tiwari TEncCu: [CHANGES OUTPUT] xComputeCostIntraInInter, fix bug
Version 1.1.196
  • Release Date: Jun 25, 2014
    • 8 hours ago Ashok Kumar Mishra TEncCU:fix for large memory consumption
    • 21 hours ago Min Chen alloc concatenation memory for m_cuData
    • 18 hours ago Steve Borho entropy: remove TEncEntropyIf abstract class
    • 18 hours ago Steve Borho TEncEntropy: nit
    • 18 hours ago Steve Borho frameencoder: remove more redundant checks
    • 19 hours ago Steve Borho frameencoder: simplify cabac context save
    • 19 hours ago Steve Borho frameencoder: simplify SAO reset when not enabled on slice
    • 19 hours ago Steve Borho frameencoder: further simplify SAO coding
    • 19 hours ago Steve Borho frameencoder: simplify SAO coding logic
    • 21 hours ago Steve Borho frameencoder: simplify loop bounds
    • 21 hours ago Steve Borho frameencoder: further simplify the logic which sync's CABAC with upper-right LCU
    • 21 hours ago Steve Borho frameencoder: remove redundant widthInCU variable
    • 21 hours ago Steve Borho frameencoder: simplify logic slightly, numSubstreams can be > 1 only with WPP
    • 21 hours ago Steve Borho frameencoder: nits
    • 21 hours ago Steve Borho frameencoder: move trace code for clarity, remove dead bWaveFrontsynchro
    • 21 hours ago Steve Borho frameencoder: use m_param->bEnableWavefront directly
    • 21 hours ago Steve Borho frameencoder: simplify setup for encodeSlice()
    • 21 hours ago Steve Borho frameencoder: rename m_pic to m_frame
    • 21 hours ago Steve Borho frameencoder: remove unread bitsOriginallyInSubstreams
    • 33 hours ago Aarthi Thirumalai make cu level stats logging run time configurable
    • 22 hours ago Steve Borho api: comment nits
    • 22 hours ago Steve Borho replace TComPic with Frame class in common/frame.h
    • 18 hours ago Steve Borho frameencoder: fix --no-wpp, broken since 854f5e3072a0
    • 37 hours ago Sumalatha Polureddy encoder: move psy-rd logic together
    • 26 hours ago Steve Borho encoder: use free() to release strdup'd string
    • 44 hours ago Steve Borho TEncSearch: change pointer type of m_mref, to avoid repeated pointer setup
    • 44 hours ago Steve Borho cturow: use thread local storage for TEncCU and TEncSearch
    • 45 hours ago Steve Borho cu: remove m_loopFilterAcrossTilesEnabledFlag and getPU* default arguements
    • 27 hours ago Steve Borho frameencoder: comment nits
    • 37 hours ago Sumalatha Polureddy encoder:[CHANGES OUTPUT] disable rdoq when psyrd is enabled

Version 1.1.163
  • Release Date: Jun 21, 2014
    • 16 hours ago Steve Borho nal: include the start code bytes in the bounds check
    • 16 hours ago Steve Borho nal: nits
    • 16 hours ago Steve Borho nal: allow the concatenated substream buffer to be re-used from frame-to-frame
    • 16 hours ago Steve Borho fix checked build warning
    • 17 hours ago Steve Borho fix msvc build warnings
    • 19 hours ago Steve Borho move bitstream implementations into common
    • 20 hours ago Steve Borho bitstream: unify resetBits() and clear() methods
    • 20 hours ago Steve Borho encoder: remove redundant flush argument
    • 20 hours ago Steve Borho encoder: cleanup obsolete NAL data fields
    • 21 hours ago Satoshi Nakagawa x265_encoder_encode: don't return 0 while flushing.
    • 2 days ago Deepthi Nandakumar lambda: [CHANGES OUTPUT] Change the distortion weights applied to chroma.
    • 21 hours ago Steve Borho rc: disable MSVC warnings about using POSIX function names like unlink()
    • 4 days ago Ashok Kumar Mishra TComDataCU: refactor structure for better data locality
    • 22 hours ago Steve Borho cu: nits
    • 24 hours ago Min Chen move m_blockingStrength to local
    • 24 hours ago Min Chen static m_bLFCrossTileBoundary since all of LCU use same value
    • 24 hours ago Min Chen remove reduce code since trHeight always equal to trWidth
    • 3 days ago Aarthi Thirumalai rc: initialize the 2 pass states in rc
    • 3 days ago Aarthi Thirumalai param: log rc states of m_param in x265_param2string
    • 3 days ago Aarthi Thirumalai rc: add 2 pass states in RateControl
    • 3 days ago Aarthi Thirumalai rc: define default setting and validations for 2 pass states
    • 3 days ago Aarthi Thirumalai api: introduce param variables for two-pass encodes
    • 24 hours ago Steve Borho encoder: remove useless comment
    • 24 hours ago Steve Borho nal: refactor nal code, marshall packets directly into output buffer

Version 1.1.133
  • Release Date: Jun 19, 2014
    • 50 minutes ago Steve Borho pic: reorder initializations after moving m_reconPicYuv in e3418f7497e9
    • 12 hours ago Steve Borho pic: keep the recon TComPicYuv in the pool with the TComPicSym - another 10%
    • 13 hours ago Steve Borho frameencoder: make m_ssimCnt an unsigned int
    • 14 hours ago Steve Borho encoder: move EncStats into x265 namespace
    • 14 hours ago Steve Borho pic: keep a pool of TComPicSym, allocate only for encode process
    • 15 hours ago Steve Borho pic: nit
    • 15 hours ago Steve Borho pic: split picsym allocation from main create function, rename reinit()
    • 15 hours ago Steve Borho pic: destroy() no longer needs to be virtual, compress white-space
    • 15 hours ago Steve Borho pic: improve comments
    • 15 hours ago Steve Borho pic: remove unread m_bCheckLTMSB
    • 15 hours ago Steve Borho pic: remove unread m_bUsedByCurr
    • 16 hours ago Steve Borho TEncSearch: keep param pointer instead of top encoder pointer
    • 16 hours ago Steve Borho TEncSearch: nits
    • 16 hours ago Steve Borho framefilter: replace top pointer with param pointer
    • 16 hours ago Steve Borho frameencoder: replace m_cfg with m_param pointer, use m_top for leftovers
    • 17 hours ago Steve Borho frameencoder: remove unused bitcounter
    • 17 hours ago Steve Borho TComDataCU: remove hungarian prefixes from CU vars
    • 17 hours ago Steve Borho sym: do not return pointer reference from getCU()
    • 17 hours ago Steve Borho pic: move m_elapsedCompressTime and m_frameTime to frame encoder
    • 17 hours ago Steve Borho pic: move SSIM, PSNR and hash to frame encoder
    • 18 hours ago Steve Borho pic: move timing structures to RateControlEntry
    • 17 hours ago Steve Borho sei: remove unused RESERVED hash method, avoid compiler warnings
    • 18 hours ago Steve Borho pic: better name for picture timing SEI member
    • 18 hours ago Steve Borho writer: xConvertToUInt did not need to be a class member
    • 18 hours ago Steve Borho sei: allow SEI classes to implement write() directly to avoid two-pass encode
    • 18 hours ago Steve Borho sei: add m_ prefix to members of SEIDecodedPictureHash for consistency
    • 18 hours ago Steve Borho sei: repair decoded picture hash SEI
    • 20 hours ago Steve Borho psyrd: use zero stride in psyCost C reference, minor optimization
    • 25 hours ago Steve Borho TEncSearch: simplify square block dim logic
    • 2 days ago Satoshi Nakagawa fix: TOPSKIP refers outside of picture [OUTPUT CHANGE]
    • 33 hours ago Deepthi Nandakumar quant: Enforce QP range after adding bit-depth offsets
    • 30 hours ago Kavitha Sampath sei: disable HRD with warning when vbv is off
    • 34 hours ago Sumalatha Polureddy psyrd: use psyrdcost for PU/TU decision for inter and intra
    • 38 hours ago Steve Borho sei: first step towards simplifying SEI writing
    • 37 hours ago Steve Borho bitstream: give bit counter a useful constructor
    • 34 hours ago Deepthi Nandakumar ratecontrol: [CHANGES OUTPUT for 10-bit CRF] Remove QP_BD_OFFSET from Ratecontrol
    • 36 hours ago Deepthi Nandakumar ratecontrol: move validate of CRF params to x265_check_param

Version 1.1.99
  • Release Date: Jun 18, 2014
    • 10 hours ago Steve Borho writer: nits
    • 13 hours ago Min Chen cleanup unused TComLoopFilter::loopFilterPic
    • 13 hours ago Min Chen move m_bEdgeFilter to local
    • 13 hours ago Min Chen move lfcuParam to local
    • 13 hours ago Steve Borho nal: concatenate and escape row data while building row-start array for header
    • 13 hours ago Steve Borho pixel: prevent msvc warning
    • 2 days ago Satoshi Nakagawa refine tskip related
    • 3 days ago Satoshi Nakagawa TComTrQuant: lambda for each Cb and Cr
    • 27 hours ago Sumalatha Polureddy psyRd: Change psy-rd energy measurement
    • 4 days ago Steve Borho bitstream: simplify appendSubstream
    • 4 days ago Steve Borho sbac: simplify codeSliceHeader, x265 never codes dependent slices
    • 4 days ago Steve Borho slice: remove unused m_nextSlice and m_dependentSliceSegmentFlag
    • 4 days ago Steve Borho frameencoder: inline resetEntropy(), which was only called once
    • 4 days ago Steve Borho framefilter: remove empty end method
    • 4 days ago Steve Borho frameencoder: use numSubstreams directly
    • 4 days ago Steve Borho frameencoder: remove useless setBitstream call
    • 4 days ago Steve Borho frameencoder: remove redundant function calls and obsolete comments
    • 4 days ago Steve Borho frameencoder: explain why SAO is being called after compression
    • 4 days ago Steve Borho slice: remove unused m_tileOffstForMultES
    • 4 days ago Steve Borho frameencoder: remove need for intermediate bitstreamRedirect
    • 4 days ago Steve Borho frameencoder: remove sliceSegment, it was always false
    • 4 days ago Steve Borho frameencoder: remove unnecessary set of nextSlice()
    • 4 days ago Steve Borho encoder: remove unused m_bPCMInputBitDepthFlag
    • 4 days ago Steve Borho remove unused display orientation SEI
    • 4 days ago Steve Borho frameencoder: make per-row output streams member variables
    • 4 days ago Steve Borho frameencoder: nit, unify comment style in this function
    • 4 days ago Steve Borho frameencoder: remove redundant sets of SAO lambdas, reorder for clarity
    • 4 days ago Steve Borho frameencoder: nit
    • 4 days ago Steve Borho frameencoder: move QP/lambda initialization together
    • 4 days ago Steve Borho frameencoder: inline slice bounds determination, do not perform twice
    • 4 days ago Steve Borho slice: remove unused finalized flag
    • 4 days ago Steve Borho sao: remove unnecessary shim function
    • 4 days ago Steve Borho frameencoder: remove obsolete comment
    • 4 days ago Steve Borho dpb: no need to reset next slice in init function
    • 4 days ago Steve Borho frameencoder: declare bitstreamRedirect on stack, do not malloc. cleanups
    • 4 days ago Steve Borho remove unused gradual decoding refresh info SEI
    • 4 days ago Steve Borho frameencoder: move prefix SEI generation all together, before compressCTURows
    • 4 days ago Steve Borho frameencoder: fix header case
    • 5 days ago Aarthi Thirumalai rc: set isCbr flag to true only in vbv mode and maxrate equals bitrate
    • 4 days ago Steve Borho compress: add EMMS at return of xCompressInterCu
Version 1.1.88
  • Release Date: Jun 14, 2014
    • 7 days ago Steve Borho frameencoder: re-use a single bitstream object for all NAL and SEI
    • 36 hours ago Steve Borho compress: nit comment spacing
    • 36 hours ago Steve Borho weight: ensure weight table is initialized on failure, simplify weightAnalyse()
    • 36 hours ago Steve Borho compress: remove floating point math to avoid needing to use EMMS
    • 36 hours ago Steve Borho TComDataCU: nits
    • 36 hours ago Steve Borho frameencoder: remove obsolete check
    • 43 hours ago Steve Borho nal: msvc and its integer conversion warnings
    • 44 hours ago Steve Borho nal: greatly simplify NAL header generation, do not use TComOutputBitstream
    • 2 days ago Steve Borho encoder: set m_totalFrameThreads at the same time as param->frameNumThreads
    • 2 days ago Steve Borho param: clip bframe bias range as x264 does
    • 3 days ago Kavitha Sampath TComRom: fix fraction part variation in x265_lambda2_tab for higher QPs
    • 4 days ago Deepthi Nandakumar preset: improve the speed settings for fast preset.
    • 7 days ago Steve Borho Merge with stable
    • 7 days ago Steve Borho rc: fix potential build warningstable
Version 1.1.35
  • Release Date: Jun 6, 2014
    • 14 hours ago Kavitha Sampath weightb: reinitialize weight table when weightb is off
    • 15 hours ago Deepthi Nandakumar ratecontrol: change type of m_bframeBits to int64_t
    • 17 hours ago Steve Borho Merge with stable
    • 32 hours ago Aarthi Thirumalai rc: fixes to improve quality in vbvstable
    • 17 hours ago Aarthi Thirumalai rc: use actual bits to update vbv bufferstable
    • 32 hours ago Aarthi Thirumalai rc: set pbFactor to default with cutreestable
    • 19 hours ago Steve Borho combine NAL output semantics into one clean(er) interface
    • 19 hours ago Steve Borho nal: fix checked build
    • 20 hours ago Steve Borho encoder: no longer needs NALwrite.h
    • 20 hours ago Steve Borho TComBitstream: nits
    • 20 hours ago Steve Borho remove Encoder::bEnableRDOQTS, we did not allow separate config for it
    • 21 hours ago Steve Borho encoder: bEnableRDOQ->m_bEnableRDOQ for consistency
    • 21 hours ago Steve Borho encoder: param->m_param for consistency
    • 21 hours ago Steve Borho encoder: remove m_useLossless
    • 21 hours ago Steve Borho slicetype: convert lookahead "structs with methods" to classes
    • 21 hours ago Steve Borho rc: nit
    • 21 hours ago Steve Borho rc: convert RateControl to a class
    • 21 hours ago Steve Borho nal: check padded buffer size
    • 21 hours ago Steve Borho api: must match X265_ALLOC with X265_FREE for m_nalUnitData
    • 21 hours ago Steve Borho rc: fix type of actualBits
    • 24 hours ago Steve Borho nalwrite: simplify write method; single allocation with no reallocs
    • 28 hours ago Steve Borho rc: only pass param to rc constructor
    • 28 hours ago Steve Borho rc: further improvements to code clarity
    • 28 hours ago Steve Borho rc: nit
    • 28 hours ago Steve Borho rc: use member isCbr rather than making a shadow variable
    • 28 hours ago Steve Borho rc: move isCbr initialization to constructor, to avoid init() race
    • 28 hours ago Steve Borho Merge with stable
    • 32 hours ago Aarthi Thirumalai rc: fixes to improve quality in vbv
    • 32 hours ago Aarthi Thirumalai rc: use actual bits to update vbv buffer
    • 32 hours ago Aarthi Thirumalai rc: set pbFactor to default with cutree
    • 33 hours ago Kavitha Sampath SEI: fix bus error, cleanup m_decodingUnitInfoSEIEnabled
    • 29 hours ago Steve Borho sei: white-space nits
    • 29 hours ago Steve Borho sei: remove SEIPictureTiming fields we do not signal, cleanup
    • 39 hours ago Satoshi Nakagawa fix: uninitialized read m_totalFrameThreadsstable
    • 30 hours ago Steve Borho rest: reorganize analysis options into their own section
    • 35 hours ago Kavitha Sampath fix crash in hrd when init function is called with null pointer
    • 34 hours ago Deepthi Nandakumar Merge with stable
    • 35 hours ago Deepthi Nandakumar TEncCU: fix bug in no-intra, intra-analysis should still happen for P-framesstable
    • 40 hours ago Deepthi Nandakumar Merge with stable
    • 40 hours ago Deepthi Nandakumar preset: correct ultrafast settings (in line with doc) by disabling loop-filter.stable
Version 1.0.156
  • Release Date: Jun 4, 2014
    • 11 hours ago Steve Borho TEncSearch: rename variable to avoid shadowing an earlier 'part'
    • 11 hours ago Steve Borho primitives: move more aliasing to Setup_Alias_Primitives
    • 36 hours ago Satoshi Nakagawa refine cbf==0 path: remove clearing coeff and resi
    • 13 hours ago Steve Borho TEncSearch: remove unnecessary pragma
    • 19 hours ago Steve Borho Merge with default (prepare for 1.1)stable
    • 19 hours ago Steve Borho psy: use m_lambdaSAD multiplier for psy cost, m_lambdaSSE for bits
    • 21 hours ago Steve Borho change license email globally
    • 4 days ago Ashok Kumar Mishra add support for lossless encode
    • 22 hours ago Steve Borho TEncCu: nits
    • 4 days ago Ashok Kumar Mishra fix to support multiple color space format in TComSampleAdaptiveOffset and TEncCu structure
    • 23 hours ago Aarthi Thirumalai rc: improve visual quality in high bit depth encodes.
    • 22 hours ago Steve Borho cmake: nit
    • 22 hours ago Steve Borho cmake: encoder.cpp no longer needs -Wno-unused-parameter
    • 22 hours ago Steve Borho cmake: move all "GCC-like" logic together, use consistent var names
    • 4 days ago Den Scherbin Disable exceptions for all compilers which support fno-exceptions
    • 4 days ago Aarthi Thirumalai TEncSearch: use actual frame thread count to decide motion search limits
    • 37 hours ago Deepthi Nandakumar preset: set superfast, veryfast, faster and fast presets to rd 2.
    • 37 hours ago Deepthi Nandakumar preset: For ultrafast preset, turn off lft, change rdLevel from 3 to 2.
    • 4 days ago Den Scherbin Disable exceptions. Fixes Clang linker errors.
Version 1.0.131
  • Release Date: May 30, 2014
    • 13 hours ago Steve Borho nits
    • 16 hours ago Steve Borho nits
    • 16 hours ago Steve Borho pool: allow thread private data structures
    • 18 hours ago Steve Borho rest: add missing --no-repeat-headers option
    • 24 hours ago Steve Borho TEncCu: nit
    • 30 hours ago Deepthi Nandakumar psy-rd: bug fix in merge mode, use psyCosts for all decisions whenever psy-rd is enabled
    • 30 hours ago Deepthi Nandakumar TEncCU: inserting runtime checking for m_totalPsyCost
Version 1.0.126
  • Release Date: May 29, 2014
    • 20 hours ago Kavitha Sampath SEI: fix bug in picture timing SEI when interlace encoding is enabled
    • 27 hours ago Deepthi Nandakumar psyRD: psyRD costs are saved into m_totalPsyCost.
    • 27 hours ago Deepthi Nandakumar TComDataCU: introduce new variable m_psyEnergy
    • 28 hours ago Deepthi Nandakumar TComDataCU: introduce new variable m_totalPsyCost
    • 28 hours ago Deepthi Nandakumar TComDataCU: whitespace nits
    • 28 hours ago Deepthi Nandakumar TComDataCU: renaming m_totalCost to m_totalRDCost
    • 3 days ago Satoshi Nakagawa add -D__STDC_LIMIT_MACROS=1
    • 2 days ago Steve Borho nits
Version 1.0.111
  • Release Date: May 24, 2014
    • 16 hours ago Steve Borho rdcost: use floating point math to calculate psy rd cost
    • 16 hours ago Steve Borho rdcost: fix dc portion of psy energy calculation
    • 8 days ago Steve Borho api: allow minimum decoder level to be specified
    • 23 hours ago Steve Borho param: introduce warnings about measurements with psy-opts enabled
    • 29 hours ago Steve Borho TEncSearch: readability nits
    • 4 days ago Satoshi Nakagawa cleanup bReusePred, unify absTUPartIdx to absPartIdx
Version 1.0.103
  • Release Date: May 23, 2014
    • 84 minutes ago Steve Borho rc: simplify updateVbvPlan()
    • 3 hours ago Deepthi Nandakumar rest: refine rdLevels description.
    • 3 hours ago Deepthi Nandakumar psyrd: is enabled only at rdLevels 5 and 6.
    • 5 hours ago Santhoshini Sekar rc: always use frameSizeEstimted for bits in updateVbvPlan
    • 8 hours ago Santhoshini Sekar rc: rename variable vbvMinRate to isCbr
    • 12 hours ago Satoshi Nakagawa fix: segmentation fault --tune zerolatency
    • 22 hours ago Steve Borho cmake: re-enable wd4800 for TLibEncoder
    • 22 hours ago Steve Borho cmake: further cleanups
    • 22 hours ago Steve Borho cmake: re-enable wd4244 and wd4127 in common/vec
    • 22 hours ago Steve Borho cmake: re-enable wd4800 for TLibCommon, fix remaining warnings
    • 22 hours ago Steve Borho frameencoder: fixup comment following variable rename
    • 22 hours ago Steve Borho lowres: use a better name for cuTree offset buffer
    • 23 hours ago Steve Borho cmake: merge TLibCommon source groups
    • 23 hours ago Steve Borho move TComRdCost.h to encoder/rdcost.h, cleanup refacto
Version 1.0.78
  • Release Date: May 21, 2014
    • 2 days ago Satoshi Nakagawa refine merge related
    • 3 days ago Satoshi Nakagawa encodeCoeffNxN(): call only if non-zero coeff exists
    • 3 days ago Satoshi Nakagawa cleanup unused arg
    • 15 hours ago Steve Borho psyrd: use EMMS before checking calcPsyRdCost
    • 3 days ago Steve Borho psyrd: use actual recon pixels for intra predictions (bug spotted by valgrind)
    • 3 days ago Steve Borho rest: improve --rd docs
    • 3 days ago Steve Borho ratecontrol: initialize singleFrameVbv (bug spotted by valgrind)
    • 3 days ago Steve Borho slicetype: initialize bReady (bug spotted by valgrind)
Version 1.0.54
  • Release Date: May 16, 2014
    • 7 hours ago Satoshi Nakagawa cleanup clear()
    • 4 hours ago Steve Borho TEncSearch: remove unused (and shadowed) variable
    • 4 hours ago Steve Borho checked: convert a pile of asserts() into optional runtime checks
Version 1.0.50
  • Release Date: May 14, 2014
    • 2 days ago Satoshi Nakagawa fix 4:2:2 coeff rate calc
    • 3 days ago Satoshi Nakagawa refine intra reference samples
    • 24 hours ago Steve Borho ratecontrol: do not low-clip first I frame with ABR
    • 4 days ago Satoshi Nakagawa reduce xModeBitsIntra() call
    • 24 hours ago Steve Borho rest: fix typo
    • 33 hours ago Steve Borho rest: add API docs with links to appropriate sections
    • 2 days ago Steve Borho rest: add a note to the tune documentation about using --tune ssim/psnr
    • 2 days ago Steve Borho api: even moar
    • 2 days ago Steve Borho api: moar tpyo
    • 2 days ago Steve Borho api: tpyo
    • 2 days ago Steve Borho api: remove mention of x265_encoder_reconfig(), which doesn't yet exist
    • 2 days ago Steve Borho api: remove _t suffix from x265_param in comments
    • 5 days ago Satoshi Nakagawa reduce malloc call to avoid fragment and overhead
    • 4 days ago Steve Borho rdcost: fix comments for lambda fix8 variables, refs b77ca886ef3b
Version 1.0.38
  • Release Date: May 10, 2014
    • 18 hours ago Min Chen reduce code size on intra_pred_ang8_*
    • 19 hours ago Steve Borho TComPic: nits
    • 21 hours ago Steve Borho rest: fix indent of min-crf
    • 22 hours ago Steve Borho cli: expose param.bRepeatHeaders to the CLI, reorder help for more clarity
Version 1.0.26
  • Release Date: May 9, 2014
    • 13 hours ago Aarthi Thirumalai slicetype: update the lowresCosts and intraCosts when aq/cutree is enabled
    • 13 hours ago Kavitha Sampath SEI: enable generation of recovery point SEI message for keyframes
    • 13 hours ago Steve Borho cmake: bump X265_BUILD for qpfile and crf-min
    • 14 hours ago Gopu Govindaswamy x265: implemented crf-min that specifies a minimum rate factor value for encode
    • 23 hours ago Steve Borho rest: document --qpfile
    • 37 hours ago Gopu Govindaswamy cli: added qpfile feature to force frame qp and slicetype using file

Version 1.0.4
  • Release Date: May 4, 2014
    • 19 hours ago Steve Borho Merge with stable
    • 19 hours ago Steve Borho Added tag 1.0 for changeset cea97c4d7945stable
    • 28 hours ago Deepthi Nandakumar interlace: set sourceScanType to 0 to indicate interlaced videostable 1.0
    • 27 hours ago Aarthi Thirumalai fix g_chromaScale for 420 videos accordingly to the HEVC spec.stable
    • 20 hours ago Steve Borho rdcost: use less fractional bits for lambda scale, and larger ints for chromastable
    • 28 hours ago Deepthi Nandakumar interlace: set sourceScanType to 0 to indicate interlaced video
    • 2 days ago Steve Borho Merge with stable
Version 0.9.132
  • Release Date: May 2, 2014
    • 25 hours ago Ashok Kumar Mishra fix WP issue for 422 and 444 formats
    • 35 hours ago Steve Borho rest: initial documentation for threading details and considerationsstable
    • 35 hours ago Steve Borho csv: always write summary info if file handle is openstable
    • 44 hours ago Steve Borho Merge with default (code freeze for 1.0)stable
    • 44 hours ago Steve Borho uncrustify - enforce coding style mechanicallydefault
    • 46 hours ago Steve Borho slicetype: nit
Version 0.9.126
  • Release Date: Apr 30, 2014
    • 24 hours ago Steve Borho TComRom: hard-code the lambda tables
    • 29 hours ago Aarthi Thirumalai remove list data from lowres costs prior to use
    • 26 hours ago Aarthi Thirumalai fix g_chromaScale to be full length
    • 4 days ago Satoshi Nakagawa fix: 4:2:2 rdLevel <= 1
    • 30 hours ago Ashok Kumar Mishra fix hash mismatch for 422 format with HM 14.0_RExt decoder
    • 2 days ago Steve Borho slicetype: remove unused auto var
    • 2 days ago Steve Borho slicetype: rename rowsCompleted to bFrameCompleted
    • 2 days ago Aarthi Thirumalai slicetype: fix incorrect initialization of fenc->rowSatds when wpp is enabled.
    • 2 days ago Steve Borho asm: adjust x264_ prefixes
Version 0.9.117
  • Release Date: Apr 26, 2014

Rate this software:

Your Rating:
You have not voted yet!

Average: 4.43
Total Ratings: 7

 




Software Submissions