Lmst

macOS Tahoe se prépare pour l'encodage AV1 http://dlvr.it/TLJwnx #macOS #AV1

#Handbrake has been doing a pretty great job for me when encoding my Blurays to #AV1. I have noticed a few times though, especially in movies with lots of film grain in dark scenes, it can be pretty destructive on the grain in those dark scenes. I've discovered two changes to fix it.

1) Encode the movie to 10 bit color, even if the source file is 8 bit.
2) Add the following to "Additional Options" under the video tab of Handbrake.

film-grain-denoise=0:film-grain=20

🌕 讓 rav1d 視訊解碼器提速 1%
➤ 透過分析和優化，提升 AV1 解碼器的效率
✤ https://ohadravid.github.io/posts/2025-05-rav1d-faster/
這篇文章記錄了作者嘗試優化 Rust 版本的 AV1 視訊解碼器 rav1d 性能的過程。作者參與了 memorysafety.org 舉辦的性能提升競賽，透過分析 profiler 的資料，發現了兩個小的性能瓶頸，並針對其中一個瓶頸進行了優化，成功將解碼速度提升了略高於 1%，且未使用任何不安全的程式碼。優化重點集中在 cdef 濾鏡的實現上，發現 rav1d 使用了較大的暫存緩衝區來歸零，而 C 版本的 dav1d 則使用更精簡的方式。
+ 這篇文章很棒！詳細解釋瞭如何透過分析性能瓶頸，並針對性地進行優化，即使是很小的改進也能帶來實際的效能提升。
+ 令人印象深刻的是作者在未使用不安全程式碼的前提下完成了優化，這對於 Rust 語言來說尤其重要，也證明瞭 Rust 在性能上的潛力。
#視訊解碼 #性能優化 #Rust #AV1

@wyatt I've seen others interpret it as rav1d offering a $20,000 bug bounty for dav1d, the C decoder. If you find exploitable undefined behavior that the dav1d devs have to spend 5% of CPU time to prevent, you win.

#rav1d #dav1d #av1 #BugBounty #UndefinedBehavior #riir #rustlang

End-to-end, fully #opensource HDR10 pipeline. Hantro #AV1 10bit decoding accelerator, #Wayland color management protocol, #GStreamer video player. All running on the Radxa Rock5 RK3588. Drop by our table at Embedded Recipes to check it out! #Embedded #OpenSource

$20K Bounty Offered for Optimizing Rust Code in Rav1d AV1 Decoder

https://www.memorysafety.org/blog/rav1d-perf-bounty/

#HackerNews #Rust #Bounty #Rav1d #AV1 #Decoder #CodeOptimization

Boosteroid Ultra subscribers are getting a visual boost! They have begun rolling out support for the advanced AV1 codec, promising an even smoother, higher-fidelity cloud gaming experience.
#CloudGaming #Boosteroid #AV1

https://clouddosage.com/boosteroids-next-gen-cloud-gaming-with-av1-codec-support/

I've been watching a couple of lectures on a phone during a train ride, and I feel like the following is my favorite minimum-bandwidth format: ~1MB/min - 480p 100kbit/s #AV1 video + 32kbit/s #Opus audio.

For non-action content (lectures, "talking-head" videos, ...) it's more than adequate for a phone screen over constantly-changing low-bandwidth connection. Text is nicely readable (over a couple of frames), especially if scene changes are far apart. And I'm more listening than watching anyway.

It's wild how effective hardware encoding is.

I needed to convert a 1080p 2 hour lecture down to a reasonable size (so that I can view it via slow mobile). I encoded it to #AV1 using VAAPI on my #Arc750. I'm getting 20x realtime encode speed 🤯

I'm happy to get 0.2x using software encoding with libsvtav1.

Adveritas ( #AV1 ) has released " Quarterly Activities Report & Appendix 4C " on Mon 28 Apr at 08:34 AEST #tax #ArtificialIntelligence #government #UnitedStates #Australia
https://grafa.com/asset/adveritas-ltd-673-av1.asx?utm_source=asxmktsensitive&utm_medium=mastodon&utm_campaign=av1.asx

@ezra you do realize that #AV1 is a superset of #VP9?

https://wetdry.world/@ezra/114407858544159170

You could add #VP3.2 (#Theora), & #VP8 instead...

@ezra personally I think #Apple should be forced.to implement #AV1 & #WebM and that @EUCommission could mandate #OpenStandards.

As for quality, were it not for #patents and espechally #SoftwarePantents, we'd have way better #Codecs, but this way people engineering a #FLOSS #Codec are constantly required to specifically avoid infringement of Patents and other IP by not using anything remotely related to it...

Also if #H265 wasn't #patented and facing #PatentTrolls and #rentseekers we'd see way better quality there as well.

Cuz #CCSS & #FLOSS can't share code nor expertise at all.

And yes, Apple (#MPEGLA-Member) is at the root of the problem!

That's why:

Duration: 00:00:10.23, start: 0.000000, bitrate: 30013 kb/s
...
Stream #0:1[0x2](eng): Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, stereo, fltp, 192 kb/s (default)
Stream #0:2[0x3](eng): Video: hevc (Main) (hvc1 / 0x31637668), yuvj420p(pc, bt709), 1920x1080, 29725 kb/s, SAR 1:1 DAR

Let's see how far I can compress it down with #av1 until "I see some pixels"

dav1d, le décodeur AV1 le plus optimisé qu'Apple refuse d'utiliser http://dlvr.it/TKMlSN #dav1d #AV1

Done this with the #AMD 5600G's iGPU, now testing #Jellyfin hardware transcoding with my #Intel #ArcA380 GPU - one question though, is it normal for CPU usage to be rather high at least in the early parts of streaming (while transcoding), even with hardware transcoding?

I'm just trying to figure out if it is actually hardware transcoding - I'm assuming it is, bcos on the admin dashboard I'm seeing that it's transcoding as AV1 and I'm sure my Ryzen 7 1700 would not be able to do/handle that, esp considering that I'm testing with 4 streams playing concurrently? but the CPU usage is rather high the first minute or two or more from when the stream starts, it does lower down afterwards - the video playback is perfectly fine, no stuttering or anything like that.

My method of passthrough is the same as I did with the 5600G, that is a simple passthrough to the #LXC container, then to the #Docker container running Jellyfin. I don't think I noticed this high CPU usage when testing the 5600G, and the only minor configuration difference between the two was I'd used #VAAPI with the 5600G and disabled #AV1 encoding (since idt it supports it), while on the Arc A380 GPU I'm using Intel's #QSV and have enabled AV1 encoding.

Am I correct to assume that hardware transcoding is indeed working? Cos again, I'm quite certain my Ryzen 7 1700 would definitely NOT be able to handle this lol esp since I only give the LXC container 2 cores.

Screenshot of a terminal window showing htop on the top half, and docker logs at the bottom. Htop shows high CPU usage running multiple ffmpeg processes.

Playback info message on Jellyfin admin dashboard of one of the 4 streaming sessions, saying that it's 'Transcoding', and that 'The media is being converted into a format that is compatible with the device that is playing the media.'

The format its transcoding seems to be '420 Kbps MP4 AV1 AAC'.

The reason it gave for transcoding says, 'The number of audio channels is not supported. The video's bitrate exceeds the limit'.

🎬 AV1 is supposed to make streaming better, so why isn’t everyone using it?
—@verge

｢ Netflix, Microsoft, Google, Amazon, and Meta banded together to develop video compression’s latest evolution: AV1. The companies say the video codec is around 30 percent more efficient compared to other standards like HEVC and the Google-developed VP9, allowing it to deliver higher-quality video at a lower bandwidth ｣

https://www.theverge.com/tech/635020/av1-streaming-netflix-youtube-google-adoption

#av1 #codecs #opensource

Updated Numbers for Eve-VP9 & Eve-AV1

https://lemmy.world/post/27915584

Updated numbers for Eve-AV1 & Eve-VP9 are public! The last time they were updated was around 2019, so this should give a more relevant look at how they perform. It takes a lot to improve a mature video encoder, and I'm very impressed with the Eve lineup. I may be biased, though.

#av1 #video #transcoding #encoding #codec #vp9