Closed lastrosade closed 3 years ago
What would the use case for this feature be?
Some services like discord have size limits, many of my vp9 encodes end up at sizes over 100MB. It would be perfect if I could get two or three separate videos without having to cut the videos with ffmpeg as this often produces broken audio/video.
Say an encode would leave us with a 72MB file and we set a file size limit of 50MB and with -o outputfile.webm
Av1an should then produce two files outputfile_1.webm and outputfile_2.webm of about 36MB each.
Av1an should try to produce output files of roughly the same size, ie: avoid create one file of 50MB and one of 22MB.
This could have issues if we decide to copy the audio stream instead of re-encoding. Av1an may also not be able to produce chunks of sizes inferior to the specified file size limit if one was to use unreasonable settings.