libaom v3.2.0 Released

[Wan-Teh Chang]

Hello everyone,

libaom v3.2.0 has been released. The source code of the release can be
checked out from the git repository[1] using the release tag v3.2.0.
Alternatively, the release tarball[2], along with the signature file[3],
can be downloaded.

Thanks.

[1] https://aomedia.googlesource.com/aom
[2] https://storage.googleapis.com/aom-releases/libaom-3.2.0.tar.gz
[3] https://storage.googleapis.com/aom-releases/libaom-3.2.0.tar.gz.asc

This release includes compression efficiency and perceptual quality
improvements, speedup and memory optimizations, as well as some new
features.

- New Features
* Introduced speeds 7, 8, and 9 for all intra mode.
* Introduced speed 10 for real time mode.
* Introduced an API that allows external partition decisions.
NOTE: The external partition API should be considered experimental.
Until the external partition API is declared stable, breaking
changes may be made to this API in a future libaom release.
* SVC: added support for compound prediction.
* SVC: added support for fixed SVC modes.

- Compression Efficiency Improvements
* Intra-mode search improvement.
* Improved real time (RT) mode BDrate savings by ~5% (RT speed 5)
and ~12% (RT speed 6). The improvement was measured on the video
conference set.
* Improved real time mode for nonrd path (speed 7, 8, 9): BDrate
gains of ~3-5%.
* Rate control and RD adjustments based on ML research in VP9.
Gains of ~0.5-1.0% for HD.

- Perceptual Quality Improvements
* Added a new mode --deltaq-mode=3 to improve perceptual quality
based on a differential contrast model for still images.
* Added a new mode –deltaq-mode=4 to improve perceptual quality
based on user rated cq_level data set for still images.
* Weighting of some intra mode and partition size choices to better
manage and retain texture.

- Speedup and Memory Optimizations
* Further improved 2-pass good quality encoder speed:
o Speed 2 speedup: 18%
o Speed 3 speedup: 22%
o Speed 4 speedup: 37%
o Speed 5 speedup: 30%
o Speed 6 speedup: 20%
* Optimized the real time encoder (measured on the video conference
set):
o RT speed 5 speedup: 110%
o RT speed 6 speedup: 77%

- Bug Fixes
* Issue 3069: Fix one-pass mode keyframe placement off-by-one error.
* Issue 3156: Fix a bug in av1_quantize_lp AVX2 optimization.

[klaus zhang]

Hello 

I've noticed that speed 10 for real time mode is introduced  at relase v3.2.0,however, the value range is still [0,9] in `aomcx.h`,so how can I enbale speed 10 ?thanks.

/*!\brief Codec control function to set encoder internal speed settings,

* int parameter

*

* Changes in this value influences the complexity of algorithms used in

* encoding process, values greater than 0 will increase encoder speed at

* the expense of quality.

*

* Valid range: 0..9. 0 runs the slowest, and 9 runs the fastest;

* quality improves as speed decreases (since more compression

* possibilities are explored).

*/

AOME_SET_CPUUSED = 13,

[klaus zhang]

ret = aom_codec_control(&ctx_, AOME_SET_CPUUSED,10); runs fine,which means speed 10 for real time mode works fine.

 so problem is that comments for AOME_SET_CPUUSED is not updated in `aomcx.h`

[Wan-Teh Chang]

On Sun, Dec 26, 2021 at 9:49 PM klaus zhang <asplin...@gmail.com> wrote:
>
> ret = aom_codec_control(&ctx_, AOME_SET_CPUUSED,10); runs fine,which means speed 10 for real time mode works fine.
>
> so problem is that comments for AOME_SET_CPUUSED is not updated in `aomcx.h`

Thank you for the report. You are right. The comments are out of date.
I wrote a changelist to update the comments and the help message for
AOME_SET_CPUUSED:

https://aomedia-review.googlesource.com/c/aom/+/152241

In the good mode and all intra mode, 9 is the highest value of
AOME_SET_CPUUSED that can be set. However, the good mode treats speed
7..9 the same as 6, so speed 6 is essentially the highest speed in the
good mode.

Wan-Teh

[klaus zhang]

Hi,

Is there any performance benchmark of CPU speed 10 in comparison with CPU speed 9 with usage of AOM_USAGE_REALTIME ?

We use AV1 in real-time scenario(RTC products),so the encoding speed means much to us.And if v3.2.0 is proved that having a good performance improvements with usage of AOM_USAGE_REALTIME,I will update libaom library to v3.2.0 as fast as I can.

[Fyodor Kyslov]

Hello,

The performance improvement depends very much on the content, but in our tests we are seeing 10-35% encoding speed up on x86 arch on VGA resolution and on bitrates of 100-500k comparing to speed 9.

Average BDRate degradation is 9.1% comparing to speed 9.

Hope this helps.

Thanks,

-Fyodor

[klaus zhang]

Fyodor,thanks for your warm reply,I will try to update libaom to v3.2.0 ASAP.Anyway ,Are the test results available to us publicly ?If so,it will be wonderful.