Intel Releases SVT-AV1 2.0 For Even Faster AV1 Encoding

Intel has published SVT-AV1 2.0 as the newest major feature release to this leading open-source CPU-based AV1 video encoder. Along with various API changes, SVT-AV1 2.0 has yet more encode performance optimizations.

SVT-AV1 2.0 is succeeding SVT-AV1 1.8 due to various API breaks. A variety of adjustments to the API were made and have already been adapted for the likes of FFmpeg use. When it comes to the encoder performance in SVT-AV1 2.0, there are yet more optimizations. The release notes for SVT-AV1 2.0 sum up the changes as:

Improve the tradeoffs for the random access mode across presets:
- Speedup presets MR by ~100% and improved quality along with tradeoff improvements across the higher quality presets
- Improved the compression efficiency of presets M9-M13 by 1-4%
- Simplified VBR multi-pass to use 2 passes to allow integration with ffmpeg
- Continued adding ARM optimizations for functions with c_only equivalent
- Replaced the 3-pass VBR with a 2-pass VBR to ease the multi-pass integration with ffmpeg
- Memory savings of 20-35% for LP 8 mode in preset M6 and below and 1-5% in other modes / presets

Considering SVT-AV1 already is a remarkably fast AV1 video encoder, it's stunning to see the continued speed-ups and memory optimizations still being squeezed out of the codebase. It's also great to see ARM (non-x86) improvements continue to come to this Intel-spearheaded open-source effort.

Downloads and more information on the SVT-AV1 2.0 release via AOMediaCodec's GitLab. New SVT-AV1 benchmarks of the v2.0 release coming soon.