v3.0.2: Supernova

Good evening anyone who's reading. It's been a considerably amount of time since we were stated to release, but here we are.

This will be the last BIG release of svt-av1-psy for a long while.

I'll be starting to port over all of the relevant features to mainline svt-av1.

However, even after all relevant features will have been ported, I won't be working on "mainline" svt-av1-psy anymore; all future feature additions, bug fixes and optimizations will be done on an svt-av1-psy "fork".

Some members wish to distance themselves from the project for reasons I won't discuss here; they're not harmful to the svt-av1-psy project, those members just want to leave their work on svt-av1-psy as history.

Getting back on topic, we've decided to release our current coup-de-grâce efforts, svt-av1-psy 3.0.2: Supernova. We're going out with a bang!

PSY Updates

Features

• --sharp-tx has been added and enabled by default; it disables conventional transform optimizations to provide a sharper output overall. In 2.3.0-B, it was enabled by default for tune 0/3. It has the effect, although unintended at the time, of making psy-rd much stronger, which is why it has been kept default. However, for improved user control, we've decided to make it a user controllable setting.

• --hbd-mds has been added as a setting. It can be used to change the bit-depth of the mode decision pipeline. By default, it respects presets defaults. At 1/2/3, it forces 10-bit+ (HBD) mode decision, 8/10-bit hybrid mode decision and full 8-bit mode decision. It has been made available because on faster presets, especially with psy-rd, forcing 10-bit mode decision can make a huge difference to quality.

Quality & Performance

• --psy-rd can now be used with all tunes.

• --psy-rd has been optimized to slighly increase encoding speed.

• High quality svt-av1-psy psy-rd has been added, and is activated at --psy-rd>=0.6 while using presets 6 and slower. It can considerably increase the strength of psy-rd, greatly increasing visual quality at no rate cost. It's been implemented hrough some careful mode decision changes that occur when psy-rd influence is significant enough. The reason behind not implementing by default and limiting it to slower presets (<=P6) is that it has a non negligible speed penalty, which gets problematic for the fastest presets; it only provides it at higher psy-rd strengths, justifiying the limiting it to -psy-rd>=0.6.

• --spy-rd levels have been added for greater control over the visual tradeoff that spy-rd pathways provide. The available levels are 0/1/2: 0 is default, with spy-rd disabled. 1 is fully enabled spy-rd (intra mode decision sharpness and interpolation sharpness modifications) and 2 is partially enabled spy-rd (interpolation sharpness modifications only).

• --sharp-tx 1 --psy-rd 0.5 has been made default to increase high fidelity encoding performance.

• --noise-norm-strength 1 is now default. It provides a decent visual quality increase with minor tradeoffs. The recommended strength for tune 0/3 is still --noise-norm-strength 3.

Bug Fixes

• The 8-bit psy-rd path has been fully fixed by @juliobbv
• --tune 2 now behaves as expected.

Changes against mainline

• Full alt-ref temporal filtering quality has been restored to Preset 2 since it provided a significant increase in quality to alt-ref TF in darker scenes.
• The intra lambda RDO value has been reverted to 2.3.0-C values, from 50 to 65, to improve low light performance. While the change did improve energy retention in brighter scenes, it actually made dark scene performance worse.

Changes from mainline

• --frame-luma-bias has been changed to --luminance-qp-bias, following mainline's implementation.

• --lp threading behavior has changed and it has resulted in some svt-av1-psy features to not be bit-exact when --lp>1. This is expected and will not be adressed.

• Like mainline, --adaptive-film-grain is now enabled by default internally. In a future release, I will add a special feature to make encoder grain synthesis much stronger.

• A bunch of other optimizations, bug fixes, and lots of ARM64 NEON SIMD for you ARM64 peeps.

Support Us

As SVT-AV1-PSY's codebase has become more complex and the encoder's capabilities have increased dramatically, our efforts have scaled in kind. We have poured hours into coding, testing, distributing, and supporting this piece of open-source software entirely for free, and our work isn't stopping any time soon. This time, I've been all alone in rebasing, reworking and fixing bugs from some stuff that wasn't entirely our fault :)

If you appreciate the work that we do and you'd like to support us, we are always excited to see code contributions from outside of the core development team. Otherwise, you can support us monetarily via the links below.

Julio Barba: Coming Soon
BlueSwordM: Coming VERY Soon
Clybius: Coming Soon
Gianni Rosato: Donate

You can also visit our website at svt-av1-psy.com. Any support you can offer goes a long way, and we sincerely appreciate it.

Binaries (SvtAv1EncApp)

Linux builds:

x86-64-v3 (AVX2) ------------> MD5 = 82bd9fea8888498bcc35f5ae986cae7d
x86-64-v4 (AVX512) --------> MD5 = 650072893823ae7c28a779cdf32698e6

macOS builds:

Coming soon!

Windows builds:

Coming soon enough.

Unofficial Handbrake repository and builds can be found here: https://github.com/Nj0be/HandBrake-SVT-AV1-PSY
Do note that 3.0.2 hasn't been pushed to Handbrake svt-av1-psy yet.

Thanks for using SVT-AV1-PSY, and see you on the other side for SVT-AV1-PSYEX

v2.3.0-B: lighting up the scene!

Hello and good evening everyone! I hope you're all doing well, despite current events.

We've recently polished up the latest release of svt-av1-psy and have even added a few things for you to play with.

PSY Updates

Features

• --spy-rd has been added: it's a psychovisually-based metric focused on enhancing sharpness and detail retention, with a slight trade-off in blocking and banding artifacts, without deeply affecting RD/RDO decisions like psy-rd.
  It can be used independently or alongside psy-rd.

Quality & Performance

• --tune 0 can now be used alongside --psy-rd. Test it out and see how it performs for your usecases.
• --sharpness>0 now enables sharper transform optimizations
• Improved film grain synthesis parallelism: greatly improved speed on multi-threaded encodes using in-encoder grain synth!

Bug Fixes

• --psy-rd>0 has been fully fixed. This also improves general performance in many ways, performance and visual quality included.

Support Us

As SVT-AV1-PSY's codebase has become more complex and the encoder's capabilities have increased dramatically, our efforts have scaled in kind. We have poured hours into coding, testing, distributing, and supporting this piece of open-source software entirely for free, and our work isn't stopping any time soon.

If you appreciate the work that we do and you'd like to support us, we are always excited to see code contributions from outside of the core development team. Otherwise, you can support us monetarily via the links below.

Julio Barba: Coming Soon
BlueSwordM: Coming Soon
Clybius: Coming Soon
Gianni Rosato: Donate

You can also visit our website at svt-av1-psy.com. Any support you can offer goes a long way, and we sincerely appreciate it. ♥️

Binaries (SvtAv1EncApp)

Linux builds:

x86-64-v3 (AVX2) ------------> MD5 = 11746e2ee8cb9b198e7572910c1e9a18
x86-64-v4 (AVX512) --------> MD5 = 1e9cd2be5a207d8e6d0cb350b444734a

macOS builds:

Coming soon!

Windows builds:

Coming soon!

Unofficial Handbrake repository and builds can be found here: https://github.com/Nj0be/HandBrake-SVT-AV1-PSY

Thanks for using SVT-AV1-PSY, and see you on the other side for SVT-AV1-PSYEX

Full Changelog: v2.3.0-A...v2.3.0-B

v2.3.0-A

Hello everyone! Merry Christmas to those who celebrate – we have a special gift for you this holiday season in the form of SVT-AV1-PSY v2.3.0-A! 🎁

Also, this is going to be my last release working heavily on features for SVT-AV1-PSY, at least for a while. You can read more at my blog post, but it has been great fun and I'll miss working on features for this project very much. I'll still be acting as the project's maintainer for now, but very infrequently. Luckily, this project has always been a team effort, and I'm more than certain my talented team members will keep these gears turning without me, so keep tuning in for new #releases!

PSY Updates

Features

• Introduced --enable-tf 2 which adaptively scales temporal filtering strength based on 64x64 block error
• Introduced --psy-rd, a feature designed to help retain visual energy to improve perceptual fidelity

Quality & Performance

• New default --sharpness value of 1
• Allowed use of additional interpolation levels for sub pixel motion estimation to improve quality at preset -1
• Variance Boost now samples across three octiles for more consistent boosting behavior
• Increased visual sharpness on keyframes relative to user-configured --sharpness value when using Tune 3

Bug Fixes

• Fixed a bug preventing use of --fast-decode 2
• Fixed crashing with Tune 3 at high CRF values (#96)
• Fixed a bug where CRF and MBR values weren't printing correctly to the terminal

Support Us

As SVT-AV1-PSY's codebase has become more complex and the encoder's capabilities have increased dramatically, our efforts have scaled in kind. We have poured hours into coding, testing, distributing, and supporting this piece of open-source software entirely for free, and our work isn't stopping any time soon.

If you appreciate the work that we do and you'd like to support us, we are always excited to see code contributions from outside of the core development team. Otherwise, you can support us monetarily via the links below.

Julio Barba: Coming Soon
BlueSwordM: Coming Soon
Clybius: Coming Soon
Gianni Rosato: Donate

You can also visit our website at svt-av1-psy.com. Any support you can offer goes a long way, and we sincerely appreciate it. ♥️

Binaries

Linux builds: (MD5: df0ee3ec323a930c7db8796d50bf0e19)

• AVX2 (x86-64-v3)

macOS builds: (MD5: a40c52d5f99205bf47bbbe1feeee0b15)

• AArch64 (Apple Silicon)

Other Builds: Community Builds Thread

Thanks for using SVT-AV1-PSY! <3

Full Changelog: https://github.com/gianni-rosato/svt-av1-psy/commits/v2.3.0-A

v2.3.0

We made a lot of progress in v2.2.1-A & v2.2.1-B, but we still have more for you on the heels of mainline's release with SVT-AV1-PSY v2.3.0! 🌠

PSY Updates

Features

• New parameter --kf-tf-strength for controlling alternate reference frame temporal filtering on keyframes. This should enable users to more finely control temporal filtering, specifically in scenarios where the benefits of lower --tf-strength are desired without the accompanying bitrate increase. The default value for this new parameter is 1.

Quality & Performance

• Tune 4 now considers user-selected min & max QM values
• Tune 4's chroma boost curve has been modified to be more compression efficient
• The deblocking loop filter has been disabled at medium-high to high fidelity when using Tune 4

Documentation

• Removed deprecated no-dec build option from build.sh example command list

Bug Fixes

• Reverted optimized 32 and wider AVX2 block subtract due to a suspect memory issue causing crashing according to some users
• Corrected noise normalization strength not being printed correctly when set to 0 with Tune 3
• Fixed user-specified disablement of temporal filtering being ignored

Mainline Updates

API updates

• Preset shift: M12/M13 mapped to M11, M7-M11 shifted one position down. API does not change, all presets from MR-M13 will still be accepted
• svt_av1_enc_get_packet API is now a blocking call for low-delay enforcing a picture in, picture out model
• --fast-decode range changed from 0-1 to 0-2 available on all presets
• Introducing a new definition of --lp being levels of parallelism with a new field in the config structure level_of_parallelism
• logical_processors will be deprecated in the 3.0 release

Encoder

• NEW FAST DECODE MODE
• New fast-decode (2) to allow for an average AV1 software cycle reduction of 25-50% vs fast-decode 0 with a 1-3% BD-Rate loss across the presets
• Improved fast-decode (1) option to increase its AV1 software cycle reduction by ~10% while maintaining the same quality levels
• Improved --lp settings for high resolutions, with CRF gaining a ~4% improvement in speed and VBR gaining ~15%
• Further Arm-based optimizations improving the efficiency of previously written Arm-neon implementations by an average of 30%. See below for more information on specific presets
• Address speed regressions for high resolutions first pass encode by tuning the threading parameters, with 1080p showing the biggest gains
• Enabled AVX512 by default in cmake allowing for ~2-4% speedup
• Enabled LTO by default if using a new enough compiler
  • If LTO is a problem or causes one, it can be disabled by adding -DSVT_AV1_LTO=OFF to cmake to force it off.
  • Please report any issues that occur when using it.

Cleanup Build & Bug Fixes & Documentation

• third_party: update safestringlib with applicable upstream changes
• Improved the unit test coverage for Arm-neon code
• Updated documentation

Arm Improvements

• Speed comparison was done against v2.2 on AWS Graviton4 instances with Clang 19.1.1
• --lp 1 was used for all tests

Standard bitdepth (Bosphorus 1080p):

Preset	Uplift
0	1.15x
1	1.24x
2	1.29x
3	1.17x
4	1.22x
5	1.31x
6	1.40x
7 (against 2.2 preset 8)	1.50x
8 (against 2.2 preset 9)	1.61x
9 (against 2.2 preset 10	1.31x
10 (against 2.2 preset 11 )	1.29x
11 (against 2.2 preset 12/13)	1.24x

High bitdepth (Bosphorus 2160p):

Preset	Uplift
0	1.18x
1	1.19x
2	1.16x
3	1.27x
4	1.33x
5	1.27x
6	1.33x
7 (against 2.2 preset 8)	1.35x
8 (against 2.2 preset 9)	1.82x
9 (against 2.2 preset 10)	1.95x
10 (against 2.2 preset 11)	1.40x
11 (against 2.2 preset 12/13)	1.35x

Huge thanks to the mainline devs for these incredible improvements!

Support Us

As SVT-AV1-PSY's codebase has become more complex and the encoder's capabilities have increased dramatically, our efforts have scaled in kind. We have poured hours into coding, testing, distributing, and supporting this piece of open-source software entirely for free, and our work isn't stopping any time soon.

If you appreciate the work that we do and you'd like to support us, we are always excited to see code contributions from outside of the core development team. Otherwise, you can support us monetarily via the links below.

Gianni Rosato: Donate
Julio Barba: Coming Soon
BlueSwordM: Coming Soon
Clybius: Coming Soon

You can also visit our website at svt-av1-psy.com. Any support you can offer goes a long way, and we sincerely appreciate it. ♥️

Binaries

Linux builds: (MD5: fd3174b3fb83bf2b7c9212080d0e424c)

• AVX2 (x86-64-v3)

macOS builds: (MD5: 5a44e0e98bcd91fe7a0cf89fd2a5e5aa)

• AArch64 (Apple Silicon)

Other builds: Community Builds Thread (Unofficial)

Thanks for using SVT-AV1-PSY! <3

Full Changelog: https://github.com/gianni-rosato/svt-av1-psy/commits/v2.3.0

v2.2.1-B

Our first ever B release is here with SVT-AV1-PSY v2.2.1-B! 🕺

PSY Updates

Features

• New parameter --noise-norm-strength for controlling noise normalization. It is configurable with values 0 through 4, with a default value of 0. Noise normalization intelligently boosts higher frequency details in parts of the video where these details may otherwise be erased, helping preserve visual consistency in some scenarios.

Quality & Performance

• Tune 4 variance boost scaling is now a bit flatter
• Adjusted Tune 4 lambda weights
• Limited the maximum available transform size available to Tune 4 to 32x32
• Adjusted the default sharpness parameter used by Tune 4
• New custom Variance Boost curve for Tune 4
• Ported speedups for AVX2 hardware from aomenc

Documentation

• DLF 2 option is now properly reflected in the help menu

Bug Fixes

• Fixed a bug where on certain devices under specific circumstances, encoding extremely small images resulted in a floating point exception

Support Us

As SVT-AV1-PSY's codebase has become more complex and the encoder's capabilities have increased dramatically, our efforts have scaled in kind. We have poured hours into coding, testing, distributing, and supporting this piece of open-source software entirely for free, and our work isn't stopping any time soon.

If you appreciate the work that we do and you'd like to support us, we are always excited to see code contributions from outside of the core development team. Otherwise, you can support us monetarily via the links below.

Gianni Rosato: Donate
Julio Barba: Coming Soon
BlueSwordM: Coming Soon
Clybius: Coming Soon

You can also visit our website at svt-av1-psy.com. Any support you can offer goes a long way, and we sincerely appreciate it. ♥️

Binaries

Linux builds: (MD5: f619a782ec61606d99528a621f61b3fa)

• AVX2 (x86-64-v3)

macOS builds: (MD5: 758e673604be33ac1039f3b0ba5656a8)

• AArch64 (Apple Silicon)

Other builds: Community Builds Thread (Unofficial)

Thanks for using SVT-AV1-PSY! <3

Full Changelog: https://github.com/gianni-rosato/svt-av1-psy/commits/v2.2.1-B

v2.2.1-A

Our most significant PSY micro-release is here with SVT-AV1-PSY v2.2.1-A!!! 🏆

PSY Updates

Features

• Encoding with odd (non-mod2) dimensions is now possible
• Encoding at resolutions lower than 64x64 is now possible, down to as small as 4x4

Quality & Performance

• A new variance boost curve for Tune 4 has been introduced, optimized for still image encoding performance
• Improved color reproduction & overall picture quality in Tune 4 & Tune 2 through chroma qindex scaling (similar functionality is already present in Tune 3)
• Higher quality presets are available when encoding 16K
• Default --chroma-qm-min has been updated to 8 (from 0)

Documentation

• Updated Handbrake link to new, cross-platform Handbrake PSY builds

Bug Fixes

• The version number should now be correct when building from an archive instead of cloning with git
• AVIFs with dimensions exceeding 4K resolution should decode correctly in applications that previously refused to decode these images when they were produced by SVT-AV1(-PSY)

Other

• The enable-libdovi & enable-libhdr10plus flags in build.sh have been updated to enable-dovi & enable-hdr10plus. The old flags will still work, but are no longer documented & will be considered deprecated from this point forward.
• We have a website now! Come check us out at svt-av1-psy.com!

Support Us

As SVT-AV1-PSY's codebase has become more complex and the encoder's capabilities have increased dramatically, our efforts have scaled in kind. We have poured hours into coding, testing, distributing, and supporting this piece of open-source software entirely for free, and our work isn't stopping any time soon.

If you appreciate the work that we do and you'd like to support us, we are always excited to see code contributions from outside of the core development team. Otherwise, you can support us monetarily via the links below.

Gianni Rosato: Donate
Julio Barba: Coming Soon
BlueSwordM: Coming Soon
Clybius: Coming Soon

Any support you can offer goes a long way, and we sincerely appreciate it. ♥️

Binaries

Linux builds: (MD5: d3315ac26f07afa6b4d899eebd207562)

• AVX2 (x86-64-v3)

macOS builds: (MD5: e68976d9e899259188a7d96eaddb07bc)

• AArch64 (Apple Silicon)

Windows & Other builds: Community Builds Thread (Unofficial)

Thanks for using SVT-AV1-PSY! <3

Full Changelog: https://github.com/gianni-rosato/svt-av1-psy/commits/v2.2.1-A

v2.2.1

Just a quick update with a couple of bug fixes & documentation adjustments 🐞

PSY Updates

Documentation

• Fix --tf-strength parameter information in the README
• Add donation info

Bug Fixes

• Fix multiple-counting block bug in Tune 3 SSIM rdmult

Mainline Updates

Cleanup, bug fixes

• Fixed debug arm macOS builds by removing invalid assert

Support Us

As SVT-AV1-PSY's codebase has become more complex and the encoder's capabilities have increased dramatically, our efforts have scaled in kind. We have poured hours into coding, testing, distributing, and supporting this piece of open-source software entirely for free, and our work isn't stopping any time soon.

If you appreciate the work that we do and you'd like to support us, we are always excited to see code contributions from outside of the core development team. Otherwise, you can support us monetarily via the links below.

Gianni Rosato: Donate
Julio Barba: Coming Soon
BlueSwordM: Coming Soon
Clybius: Coming Soon

Any support you can offer goes a long way, and we sincerely appreciate it. ♥️

Binaries

Linux builds: (MD5: 300314556a376a85efb1dd8313c82171)

• AVX2 (x86-64-v3)
• AVX-512 (x86-64-v4)
• AVX-512 (Zen 3)

macOS builds: (MD5: ed10a223401dacc8b015b1813e9e4490)

• AArch64 (Apple Silicon)

Windows & Other builds: Community Builds Thread (Unofficial)

Thanks for using SVT-AV1-PSY! <3

Full Changelog: https://github.com/gianni-rosato/svt-av1-psy/commits/v2.2.1

v2.2.0

Long time no see! We're back with another exciting release: SVT-AV1-PSY v2.2.0! 🎉

PSY Updates

Features

• --tf-strength parameter for manually adjusting temporal filtering strength. By default, reduces temporal filtering strength by 4x
• --chroma-qm-min & --chroma-qm-max parameters for adjusting quantization matrix flatness for chroma planes
• --tune 4 - A new Tune called Still Picture has been introduced for AVIF encoding, with promising gains observed over aomenc, aomenc 4:4:4, and mainline SVT-AV1

Quality & Performance

• Enable 10-bit subpixel motion estimation processing for 10-bit inputs on Preset 3 & below
• Enable SUPERRES_AUTO_ALL for Tune 3 when superres-mode=4
  • Disables ALT-REF superres for speed and quality
• Add unipred bias for SSIM full mode decision cost when using Tune 3
• New perceptual QM curve when using Tune 3
  • Better chroma QM adjustment & better high fidelity performance
• Vastly improved screen content detection when --scm 2 is set. Currently only enabled in Tune 4

Documentation

• --help menu correctly prints maximum CRF value (70)
• Default Tune 2 is better reflected across documentation
• Updated build documentation
• Additional featured projects

Bug Fixes

• Scoped down QM disablement to only affect NEON-enabled systems
• Scoped down global sharpness features to --sharpness 1, as was the original intention

Mainline Updates

API updates

• No API changes on this release

Encoder

• Improve the tradeoffs for the random access mode across presets:
  • Speedup of ~15% across presets M0 - M8 while maintaining similar quality levels
• Improve the tradeoffs for the low-delay mode across presets
• Increased temporal resolution setting to 6L for 4k resolutions by default
• Added ARM optimizations for functions with c_only equivalent yielding an average speedup of ~13% for 4k 10-bit

Cleanup Build & bug fixes & documentation

• Profile-guided-optimized helper build overhaul
• Major cleanup and fixing of Neon unit test suite
• Address stylecheck dependence on public repositories

Support Us

As SVT-AV1-PSY's codebase has become more complex and the encoder's capabilities have increased dramatically, our efforts have scaled in kind. We have poured hours into coding, testing, distributing, and supporting this piece of open-source software entirely for free, and our work isn't stopping any time soon.

If you appreciate the work that we do and you'd like to support us, we are always excited to see code contributions from outside of the core development team. Otherwise, you can support us monetarily via the links below.

Gianni Rosato: Donate
Julio Barba: Coming Soon
BlueSwordM: Coming Soon
Clybius: Coming Soon

Any support you can offer goes a long way, and we sincerely appreciate it. ♥️

Binaries

Linux builds: (MD5: 3f30899255f474518d393e8f868db2d3)

• x86-64-v3 (AVX2)
• x86-64-v4 (compiler AVX-512)

macOS builds: (MD5: 26cdba09ef411f0a0ad6067308c2019b)

• AArch64 (Apple Silicon)

Windows builds: (MD5: 009bb6bd1e014f6660b9b07eb8399126)

• x86-64-v3 (compiler AVX-512)

Thanks for using SVT-AV1-PSY! <3

Full Changelog: https://github.com/gianni-rosato/svt-av1-psy/commits/v2.2.0

v2.2.0-rc1

Note: this is a release candidate, not an official release. Expect the lack of polish that usually accompanies pre-release software, as well as this potentially incomplete release changelog.

PSY Updates

Features

• --tf-strength parameter for manually adjusting temporal filtering strength. By default, reduces temporal filtering strength by 4x

Quality & Performance

• Enable 10-bit subpixel motion estimation processing for 10-bin inputs on Preset 3 & below
• Enable SUPERRES_AUTO_ALL for Tune 3 when superres-mode=4
  • Disables ALT-REF superres for speed and quality
• Add unipred bias for SSIM full mode decision cost when using Tune 3
• New perceptual QM curve when using Tune 3
  • Better chroma QM adjustment & better high fidelity performance

Documentation

• --help menu correctly prints maximum CRF value (70)
• Default Tune 2 is better reflected across documentation
• Updated build documentation

Bug Fixes

• Scoped down QM disablement to only affect NEON-enabled systems
• Scoped down global sharpness features to --sharpness 1, as was the original intention

Mainline Updates

API updates

• No API changes on this release

Encoder

• Improve the tradeoffs for the random access mode across presets:
  • Speedup of ~15% across presets M0 - M8 while maintaining similar quality levels
• Improve the tradeoffs for the low-delay mode across presets
• Increased temporal resolution setting to 6L for 4k resolutions by default
• Added ARM optimizations for functions with c_only equivalent yielding an average speedup of ~13% for 4k 10-bit

Cleanup Build & bug fixes & documentation

• Profile-guided-optimized helper build overhaul
• Major cleanup and fixing of Neon unit test suite
• Address stylecheck dependence on public repositories

Thanks for using SVT-AV1-PSY! ♥️

Full Changelog: https://github.com/gianni-rosato/svt-av1-psy/commits/v2.2.0-rc1

v2.1.0-A

Hi there! We're back with our micro-release format to announce some exciting changes in SVT-AV1-PSY v2.1.0-A! 🎉

PSY Updates

Features

• New parameter --max-32-tx-size, which restricts block transform sizes to a maximum of 32x32 pixels. This can be useful in very specific scenarios for improving overall efficiency
• Added support for HDR10+ JSON files via a new --hdr10plus-json parameter (thanks @quietvoid!). In order to build a binary with support for HDR10+, see our PSY Development page.
• New parameter --adaptive-film-grain, which helps remedy perceptually harmful grain patterns caused by extracting grain from blocks that are too large for a given video resolution. This parameter is enabled by default

Quality & Performance

• Disabled SSIM-cost transform decisions while keeping SSIM-cost mode decisions, generally improving Tune 3's efficiency & consistency
• Additional NEON optimizations for ARM platforms, providing a speed increase

Documentation

• All of the features present in this release have been documented, so associated documentation has been updated accordingly
• Build documentation updated to reflect the HDR10+ build option

Bug Fixes

• Disabled quantization matrices for presets 5 and higher due to a visual consistency bug (#56)

Thanks for using SVT-AV1-PSY! ♥️

Full Changelog: https://github.com/gianni-rosato/svt-av1-psy/commits/v2.1.0-A

Linux builds: (MD5: 8a801a369e20cd421badfc483b5c1e47)

• Generic
• x86-64-v3 (AVX2)
• x86-64-v4 (compiler AVX512, disabled native AVX512)
• RISC-V (RV64) (MD5: da95d67dafa0ae372bebff8e883547db)

macOS builds: (MD5: 242febb82425f26c42008d1162ac7513)

• AArch64 (Apple Silicon)

Windows builds: (MD5: 070a44d7187726c430f9fb3ca703e070)

• x86-64-v3 (AVX2)