I've seen un explainable '-ss' (i.e., seek) problems.
I've seen trivial changes to FFmpeg commands produce a flood of un explainable "non-monotonous DTS"
errors when the trivial change had nothing to do with anything that would change DTS. (And I find it
tragic-funny that FFmpeg 'fixes' the bogus DTS problem by taking the previous DTS and adding '1' to
it instead of adding a whole frame.)
We've all probably experienced these types of faults and then just blew them off and tried something
different in order to get our jobs done. For example, Gyan showed me '-bsf noise' as a better, more
reliable way to cut than using '-ss'.
Ticket_11055.m2ts is the first sample I've found that makes a clear cut case that something's wrong
with FFmpeg's timestamp service -- a 'service' that's probably deeply buried. There may be several
such 'services' created as developers encounter case issues that prompt them to write their own
timestamp service 'fixes'.
May I remind you that
ffmpeg -f framecrc
ffmpeg -vf showinfo
ffprobe -show_frames
all produce differing results.
ffmpeg -f framecrc
is the only one that agrees with the actual packet DTSes & PTSes.
_______________________________________________
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user
To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
