1# Changelog 2 3## [0.8.7] - 2021-05-08 4 5Encoder 6- Feature optimizations: creating new mode decision / encode-decode feature levels allowing better speed / quality trade-off granularity 7- Preset repositioning after adopting new levels 8- Preset 8 achieving similar speed levels as x265 medium tuned for VOD operating mode 9- New 1-pass and 2-pass VBR implementation ported from libaom and adapted to the SVT architecture - still a WIP 10- Cleaned up old VBR and CVBR RC code along with the lookahead mechanism associated with them 11- 2-pass encoding is only available for VBR mode and 1-pass with lookahead is used for CRF 12- Improvements for TPL algorithm to handle long clips and easy content 13- Memory optimizations, cleaning up data structures to reduce memory usage up to 2x memory reduction in multi-threaded VBR environment 14- Additional AVX2 and AVX512 optimizations 15- Cleaned up unused command line parameters and left the config params that are linked to ffmpeg / gst 16- Update documentation accordingly 17- Added HDR support and color primaries signalling (off by default until integrated with ffmpeg) 18 19Build and Testing 20- Bug fixes 21- Improve CI coverage 22- Improve Unit Test Coverage 23- Address C vs asm mismatches 24- Fix static analysis warnings / errors 25 26## [0.8.6] - 2020-11-28 27 28Encoder 29- Further quality-speed tradeoffs tuning for VOD use cases 30- Improved TPL support within 1-pass and 2-pass CRF moode 31- Continued non-optimized support for 2pass VBR and CRF 32- Align kernel nomenclature to prefix svt_aom for kernels brough from libaom to avoid symbol conflicts 33 34Build and Testing 35- Bug fixes 36- Improve CI 37- Added CI support for gitlab 38- Improve Unit Test Coverage 39- Address C vs asm mismatches 40- Fix static analysis warnings / errors 41- Add address sanitizer 42- Fix symbol conflicts with libaom and libvpx when staticly linked to ffmpeg 43 44## [0.8.5] - 2020-09-04 45Relicensing notice 46- Change the outbound license from BSD+Patent to the AOM license / patent 47 48Encoder 49- Added tpl support to adaptively change lambda and quantization parameters within the frame 50- Added multi staged hme support 51- Quality speed trade-offs tuned to VOD use cases 52- Added first level non-optimized support for 2pass VBR and CRF 53- Added combined cli two pass support with options for stats being written to a memory buffer and a specified file 54- Added non square partitioning optimizations 55- Improved lambda generation 56 57Build and Testing 58- Bug fixes 59- Improve CI 60- Improve Unit Test Coverage 61- Address C vs asm mismatches 62- Fix static analysis warnings / errors 63- Add address sanitizer 64- Fix symbol conflicts with libaom and libvpx when staticly linked to ffmpeg 65 66 67## [0.8.4] - 2020-06-26 68 69Build and Testing 70- Bug fixes 71- Improve CI 72- Improve Unit Test Coverage 73- Address C vs asm mismatches 74- Fix static analysis warnings / errors 75- Add address sanitizer 76- Various ffmpeg patch fixes 77 78## [0.8.3] - 2020-04-24 79 80Encoder 81- Presets optimization 82 83Build and Testing 84- Bug fixes 85- Xcode build support 86 87 88## [0.8.2] - 2020-04-18 89 90Encoder 91- Initial Super resolution support 92- Mode decision rate estimation enhancements 93- Altref improvements 94- Manual prediction structure support 95- Enhanced RDOQ support 96- Improved warp motion 97- Improved documentation and help menu 98- New command line parameters 99- Fix non-multiple of 8 resolution video corruption for 8bit 100- Improved multi-stage mode decision support 101- Added multi-stage motion estimation support 102- Chroma mode decision optimizations 103- Added 16bit pipeline support 104- Added Mode decision non-square partition weights 105- Enhanced qp-scaling assignment 106- Memory optimizations 107- Added AVX512 Optimizations 108- Added AVX2 Optimizations 109 110Decoder 111- Improved multi-threading support 112- Memory optimizations 113- Updated documentation 114- Loop filter flow optimization 115 116Encoder and Decoder 117- Encoder-Decoder-Common folder structure improvements 118 119Build and Testing 120- Bug fixes 121- Improve CI 122- Improve Unit Test Coverage 123- Address C vs asm mismatches 124- Support C only builds (for platforms other than x86) 125 126## [0.8.1] - 2020-01-28 127 128Encoder 129- Palette support for 10-bit 130- Added the first batch of detailed documentation to help developers 131 132Encoder and Decoder 133- Code cleanup and refactoring 134 135Build and Testing 136- Bug fixes 137- Improve CI 138- Improve Unit Test Coverage 139- Address C vs asm mismatches 140 141 142## [0.8.0] - 2019-12-20 143 144Encoder 145- Preset Optimizations 146- Single-core execution memory optimization [-lp 1 -lad 0] 147- Rate estimation update enhancements 148- Added on / off flags for feature run-time switching 149- Added auto-max partitioning algorithm 150- Multi-pass partitioning depth support 151- Remove deprecated RC mode 1 and shifter RC mode 2 and mode 3 to mode 1 and mode 2 respectively 152- Update Cost Calculation for CDEF Filtering 153- Intra-Inter Compound for 10-bit 154- Eigth-pel optimization 155- Added AVX512 Optimizations 156- Added AVX2 Optimizations 157 158Decoder 159- Initial multi-threading support 160- Decoder optimizations / cleanup 161 162Build and Testing 163- Bug fixes 164- Improve CI 165- Improve Unit Test Coverage 166- Address C vs asm mismatches 167 168## [0.7.5] - 2019-11-24 169 170Encoder 171- RDOQ for 10-bit 172- Inter Intra Class pruning at MD-Staging 173- Global Motion Vector support for 8-bit and 10-bit 174- Interpolation Filter Search support for 10-bit 175- Palette Prediction support 176- 2-pass encoding support 177- ATB 10-bit support at the encode pass 178- Simplified MD Staging [only 3 stages] 179- Inter-Inter and Inter-Intra Compound for 10-bit 180- Intra Paeth for 10-bit 181- Filter Intra Prediction 182- New-Near and Near-New support 183- OBMC Support for 8-bit and 10-bit 184- RDOQ Chroma 185- ATB Support for Inter Blocks 186- Temporal Filtering for 10-bit 187- Eight-pel support in predictive ME 188- MCTS Tiles support 189- Added AVX512 Optimizations 190- Added AVX2 Optimizations 191 192Decoder 193- SuperRes support 194- Reference Frame Scaling support 195- 12-bit support 196- Annex B support 197 198Build and Testing 199- Bug fixes 200- Improve CI 201- Improve Unit Test Coverage 202- Address C vs asm mismatches 203 204## [0.7.0] - 2019-09-26 205 206Encoder 207- Enhanced MRP Reference Frames 208- Intra Inter Compound 209- QP Modulation support 210- MFMV Support 211- MD Staging design [Up to 4 MD stages and 3 prediction classes: Intra / Inter / Compound] 212- Compound Motion prediction 213- 10-bit Mode Decision support for Intra 214- Thread safe resource allocation 215- Added AVX512 Optimizations 216- Added AVX2 Optimizations 217 218Decoder 219- Screen Content Tools 220- Temporal MV scan support 221- Inter support 222- Screen Content Tools support 223- Post Processing Filters support 224- Compound Mode (InterInter & InterIntra) Tool support 225- Decoder Film Grain support 226 227Build and Testing 228- Improve CI 229- Improve build scripts 230- Improve cmake lists 231- Improve Unit Test Coverage 232- API update 233- Bug fixes 234 235## [0.6.0] - 2019-06-28 236 237- Inital decoder implementation 238- Static library support 239- Alt-ref pictures - temporal filtering 240- Adaptive Transform Block for INTRA 241- Adaptive QP scaling 242- Decoder - Support for Tiles and 10bit 243- API - add option to calculate / report PSNR values 244- Support for segmentation 245- SIMD Optimizations 246- Downsampling 2x2 filtering 247- Handle incomplete SBs 248- MACROS / trailing / tabs-spaces cleanup 249 250## [0.5.0] - 2019-05-17 251 252- 8 bit / 10 bit 4:2:0 up to 4k60p resolutions 253- Presets 0-8 254- New API, FFmpeg, GStreamer plugins 255- Rate control support (VBR, CVBR) 256- Block sizes from 4x4 to 128x128 257- Non-square blocks 258- Tiles 259- Deblocking / CDEF / Restoration filters 260- Film Grain 261- Warped motion estimation 262- Intra block copy 263- Trellis quantized coefficient optimization 264- Support for 4 and 5 layers prediction structures 265- Chroma search in MD 266- Multi-reference picture support 267