• man page
• All Options

• ffmpeg with Nvidia GPU

• https://www.ramugedia.com/how-encode-decode-hevc-h-265-by-intel-quick-sync-video-hw--qsv-
• https://trac.ffmpeg.org/wiki/Hardware/QuickSync
• https://gist.github.com/andrearug/ee461152acfbbd3f52b3bb9ffb55b6b8
• https://geoffruddock.com/bulk-compress-videos-x265-with-ffmpeg/
• https://slhck.info/video/2017/02/24/crf-guide.html
• https://netflixtechblog.com/a-large-scale-comparison-of-x264-x265-and-libvpx-a-sneak-peek-2e81e88f8b0f
• https://www.youtube.com/watch?v=wi1BefrfTos
• https://video.stackexchange.com/questions/14907/how-to-downsample-4k-to-1080p-using-ffmpeg-while-maintaining-the-quality

• https://github.com/markus-perl/ffmpeg-build-script

• https://trac.ffmpeg.org/wiki/AudioChannelManipulation#a5.1stereo
• https://superuser.com/a/1616102
• https://datatracker.ietf.org/doc/html/rfc7845#section-5.1.1.5

FL, FC, FR, BL, BR, LFE → FL, FR

-ac 2 \
-af "pan=stereo|FL = 0.374107*FC + 0.529067*FL + 0.458186*BL + 0.264534*BR + 0.374107*LFE |FR = 0.374107*FC + 0.529067*FR + 0.458186*BR + 0.264534*BL + 0.374107*LFE" \

FL, FC, FR, SL, SR, BL, BR, LFE → FL, FR

-ac 2 \
-af "pan=stereo|FL = 0.274804*FC + 0.388631*FL + 0.336565*SL + 0.194316*SR + 0.336565*BL + 0.194316*BR + 0.274804*LFE | FR = 0.274804*FC + 0.388631*FR + 0.336565*SR + 0.194316*SL + 0.336565*BR + 0.194316*BL + 0.274804*LFE" \

FL, FC, FR, SL, SR, BC, LFE → FL, FR

-ac 2 \
-af "pan=stereo|FL = 0.321953*FC + 0.455310*FL + 0.394310*SL + 0.227655*SR + 278819*BC + 0.321953*LFE | FR = 0.321953*FC + 0.455310*FR + 0.394310*SR + 0.227655*SL + 278819*BC + 0.321953*LFE" \

FL, FC, FR, BL, BR → FL, FR

-ac 2 \
-af "pan=stereo|FL = 0.460186*FC + 0.650802*FL + 0.563611*BL + 0.325401*BR | FR = 0.460186*FC + 0.650802*FR + 0.563611*BR + 0.325401*BL" \

FL, FR, BL, BR → FL, FR

-ac 2 \
-af "pan=stereo|FL = 0.422650*FL + 0.366025*BL + 0.211325*BR | FR = 0.422650*FR + 0.366025*BR + 0.211325*BL" \

FL, FC, FR → FL, FR

-ac 2 \
-af "pan=stereo|FL = 0.414214*FC + 0.585786*FL | FR = 0.414214*FC + 0.585786*FR" \

Quality with QSV as not good at all, though this could have been my fault.

ffmpeg \
  -y \
  -hwaccel qsv \
  -qsv_device /dev/dri/card0 \
  -c:v hevc_qsv \
  -i input.mkv \
  -c:v hevc_qsv \
  -vf 'scale_qsv=w=1920' \
  -preset slow \
  -global_quality 25 \
  -c:a copy \
  -c:s copy \
  output.mkv

• libx264
• libx265

ffmpeg \
  -i MyMovie_4K.mkv \
  -vf scale=1920:-1 \
  -c:v libx265 \
  -preset veryslow \
  -c:a copy \
  MyMovie_1080p.mkv

Not sure if these are necessary.

-crf 28

• https://gist.github.com/kurlov/32cbe841ea9d2b299e15297e54ae8971

ffmpeg \
  -i in.mkv \
  -f srt \
  -i in.srt \
  -map 0:0 \
  -map 0:1 \
  -map 1:0 \
  -c:v copy \
  -c:a copy \
  -c:s srt \
  out.mkv

http://diantokam.blogspot.com/2016/12/adding-subtitle-and-scaling-film-using.html

ffmpeg \
  -i "input.mkv" \
  -vf subtitles=fullsubs.srt \
  "output.mkv"

  # export english subs
  ffmpeg \
    -y \
    -i "./${name}" \
    -map 0:s:0 \
    eng.srt

  # create a version with hardsubs
  # si=0
  ffmpeg \
    -y \
    -i "./${name}" \
    -vf subtitles="${name}":si=0 \
    -ac 2 \
    -af "pan=stereo|FL = 0.374107*FC + 0.529067*FL + 0.458186*BL + 0.264534*BR + 0.374107*LFE |FR = 0.374107*FC + 0.529067*FR + 0.458186*BR + 0.264534*BL + 0.374107*LFE" \
    -c:a aac \
    "${name}-hardsubs.mp4"

ffmpeg \
  -itsoffset -1 \
  -i full8.srt \
  -c copy \
  full8_delayed.srt

https://hub.docker.com/r/tiangolo/nginx-rtmp https://www.nginx.com/blog/video-streaming-for-remote-learning-with-nginx/

ffmpeg -h encoder=hevc_qsv

ffmpeg \
  -i  input.mkv \
  -ss 29:36 \
  -to 29:50 \
  -c:v copy \
  -c:a aac \
  output.mp4

1. Install docker
2. Install nvidia docker: https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html
3. Docker ffmpeg image: https://hub.docker.com/r/jrottenberg/ffmpeg

docker run \
  -v $(pwd):$(pwd) \
  -w $(pwd) \
  --runtime=nvidia jrottenberg/ffmpeg:4.1-nvidia \
  -hwaccel cuvid \
  -c:v h264_cuvid \
  -i input.mkv \
  -c:v hevc_nvenc \
  -c:a copy \
  -c:s copy \
  -preset slow \
  output_hevc.mkv

[h264 @ 0x56028de15ec0] Error creating a NVDEC decoder: 1
[h264 @ 0x56028de15ec0] Using more than 32 (34) decode surfaces might cause nvdec to fail.
[h264 @ 0x56028de15ec0] Try lowering the amount of threads. Using 16 right now.
[h264 @ 0x56028de15ec0] Failed setup for format cuda: hwaccel initialisation returned error.

• https://ffmpeg.org/pipermail/ffmpeg-user/2019-June/044610.html

Note that cuda hwaccel is an alias to the newer nvdec implementation, and to my knowledge, has no way of setting up decoder surface limits.

Here's a workaround you can try:

ffmpeg \
  -threads 4 \
  -hide_banner \
  -c:v h264_cuvid \
  -surfaces 8 \
  -nostats \
  -i rtmp://<ip>:<port>/myapp/mystream \
  -filter_complex [0]fps=fps=1[s0] \
  -map [s0] \
  -f rawvideo \
  -pix_fmt rgb24 pipe

Two changes:

1: Call up the h264_cuvid decoder directly, and apply a surface limit count of 8. This should be adequate.
2: Limit the thread count.

None of this worked.

• https://stackoverflow.com/a/48874015

[matroska @ 0x55a1a3738340] Non-monotonous DTS in output stream 0:1; previous: 490791, current: 490731; changing to 490791. This may result in incorrect timestamps in the output file.

Try -use_wallclock_as_timestamps 1 and -fflags +genpts

• https://stackoverflow.com/a/65501876

Add the following after the input file:

-max_muxing_queue_size 9999