[FFmpeg-devel] Re: [PR] avcodec/h264: account for reference B-frames in reorder depth (PR #24614)

[Nil Fons Miret via ffmpeg-devel]

If I understand correctly, this is what ffmpeg already does in strict
mode: `sps->num_reorder_frames` takes the maximum derived from the
level and coded picture size. This fixes my sample and is more general
than the PR's heuristic, but it has bigger tradeoffs: the worst-case
value adds latency and memory usage to hold the N frames, and changes
`has_b_frames` from the adaptively-found value to the level max, which
breaks the expectations of some FATE tests. I set up the PR this way
to keep the adaptive approach in normal mode while fixing the frame
drop, but if there is consensus to just port the strict mode behavior
to normal mode, I would be okay with that too.
_______________________________________________
ffmpeg-devel mailing list -- ffmpeg-devel@ffmpeg.org
To unsubscribe send an email to ffmpeg-devel-le...@ffmpeg.org