[FFmpeg-trac] #4743(avformat:new): Opus ogg file is silent when looping playback.

[FFmpeg]

#4743: Opus ogg file is silent when looping playback.
--------------------------------------+----------------------------------
               Reporter:  dalecurtis  |                  Owner:
                   Type:  defect      |                 Status:  new
               Priority:  normal      |              Component:  avformat
                Version:  git-master  |               Keywords:  opus,ogg
             Blocked By:              |               Blocking:
Reproduced by developer:  0           |  Analyzed by developer:  0
--------------------------------------+----------------------------------
 ffmpeg -i sfx_f32le.wav -acodec opus sfx-opus.ogg
 ffmpeg -i sfx_f32le.wav -acodec libopus sfx-libopus.ogg

 Now play either file with ffplay and wait for playback to finish then seek
 to zero, observe silence.  That's as far as I could get with debugging
 with the CLI tools.

 Using Chrome's decoder implementation, I can see that seeking to -7000
 instead of the start time of -6500 (as indicated by first packet pts) will
 return all the correct packets. Otherwise the packets get consecutive pts
 values starting from 1, with all of their content marked for end trimming.

 [1:9:0728/153127:ERROR:ffmpeg_demuxer.cc(1258)] pts: 0
 [1:9:0728/153127:ERROR:ffmpeg_demuxer.cc(1258)] pts: 960
 [1:9:0728/153127:ERROR:ffmpeg_demuxer.cc(1258)] pts: 1920
 [1:9:0728/153127:ERROR:ffmpeg_demuxer.cc(1258)] pts: 2880
 [1:9:0728/153127:ERROR:ffmpeg_demuxer.cc(1258)] pts: 3840
 [1:9:0728/153127:ERROR:ffmpeg_demuxer.cc(1258)] pts: 4800
 [1:9:0728/153127:ERROR:ffmpeg_demuxer.cc(1258)] pts: 5760
 [1:9:0728/153127:ERROR:ffmpeg_demuxer.cc(1258)] pts: 6720
 [1:9:0728/153127:ERROR:ffmpeg_demuxer.cc(1258)] pts: 7680
 [1:9:0728/153127:ERROR:ffmpeg_demuxer.cc(1258)] pts: 8640
 [1:9:0728/153127:ERROR:ffmpeg_demuxer.cc(1258)] pts: 9600
 [1:9:0728/153127:ERROR:ffmpeg_demuxer.cc(1258)] pts: 10560
 [1:9:0728/153127:ERROR:ffmpeg_demuxer.cc(1258)] pts: 11520
 [1:9:0728/153127:ERROR:ffmpeg_demuxer.cc(1258)] pts: 12480
 [1:9:0728/153127:ERROR:ffmpeg_demuxer.cc(1258)] pts: 13440
 [1:9:0728/153127:ERROR:audio_discard_helper.cc(222)] End discard of 244
 out of 960
 [1:9:0728/153128:ERROR:ffmpeg_demuxer.cc(681)] Seeking to: -6500, -312 in
 stream 0
 [1:9:0728/153128:ERROR:ffmpeg_demuxer.cc(1258)] pts: 0
 [1:9:0728/153128:ERROR:audio_discard_helper.cc(222)] End discard of 960
 out of 960
 [1:9:0728/153128:ERROR:ffmpeg_demuxer.cc(1258)] pts: 1
 [1:9:0728/153128:ERROR:audio_discard_helper.cc(222)] End discard of 960
 out of 960
 [1:9:0728/153128:ERROR:ffmpeg_demuxer.cc(1258)] pts: 2
 [1:9:0728/153128:ERROR:audio_discard_helper.cc(222)] End discard of 960
 out of 960
 <snip>

 At first I thought out test clip was muxed incorrectly, but as you can see
 from the report creating an ogg opus file using ToT will result in the
 same issues. Since increasing the seek time seems to "fix" the issue, I'd
 guess that seeking isn't properly finding the beginning of the stream.

--
Ticket URL: <https://trac.ffmpeg.org/ticket/4743>
FFmpeg <https://ffmpeg.org>
FFmpeg issue tracker