Nyanmisaka

Can adjusting `-filter_threads` improve performance on you end? It works for me on some processors. To reflect the performance on an entry-level platform, I also ran it on RK3588(4xA76+4xA55). Somehow...

Make sense. It would be perfect if we could combine the advantages of both. With intrinsics optimitzed LUT, various tonemap methods and dovi reshaping can be accelerated. Even mid-range processors...

Can you push your patch to this branch, and let's see what is going on? The `signed -> unorm` normalization is usually done by hardware on GPGPU, and I guess...

vuyx/ayuv format fails after this change. But it does bring a perf uplift over other formats. ``` ./ffmpeg -hide_banner -init_hw_device vaapi=va:/dev/dri/renderD128 -init_hw_device qsv=qs@va -filter_hw_device qs \ -f lavfi -i nullsrc=s=1920x1080,format=vuyx...

8bit vs 10bit is unlikely to cause such a noticeable difference visually, it will only cause color banding. It is recommended that you add a comparison item, that is, a...

@podhorsky-ksj It should be a duplicate of https://github.com/jellyfin/jellyfin/issues/10698, which is fixed by https://github.com/jellyfin/jellyfin/pull/10710

Then I'm out of ideas. I haven't encountered a similar problem once. What version is vanilla ffmpeg? The changes in jellyfin-ffmpeg are basically for hardware acceleration.

> it is 6.1.1. . The changes are even for VAAPI? There are some, but not for older platforms. Maybe 6.1 contains some fixes that I'm not aware of. We're...

> Can you check if the "Throttle Transcodes" setting is enabled in the Playback settings? I've noticed that this option works poorly with hardware acceleration, and it can cause the...

Based on my years of experience fighting with AMDGPU on Linux, its hardware design and driver tolerance are very low, and the kernel driver is always changing and causing regression....