> On Aug 17, 2020, at 4:05 PM, Nicolas George <geo...@nsup.org> wrote: > > Steven Liu (12020-08-17): >> Output a warning message if the target duration of the segment is negative. >> Suggest user increase the hls_time value, >> and modify the target duration to one packet duration, >> because there maybe have bframe and then split not by keyframe, >> and the segment is very very small. > > Fiddling randomly with durations seem wrong. It will probably result in > invalid seeking. Have you found the exact cause of the issue? >
The split_by_time flag which enabled split segments at any position (non-key frame) is the root cause of the issue. The specification (RFC8216, 6.2.1) says explicitly: The server SHOULD attempt to divide the source media at points that support effective decode of individual Media Segments, e.g., on packet and key frame boundaries. > Preventing duration from being negative in the first place would be a > much better fix than adjusting it arbitrarily. > > Regards, > > -- > Nicolas George > > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe". _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
