[FFmpeg-trac] #8464(avformat:closed): `outpoint` directive of `concat` demuxer may be off by 1 frame

FFmpeg trac at avcodec.org
Fri Jan 10 01:33:44 EET 2020


#8464: `outpoint` directive of `concat` demuxer may be off by 1 frame
-------------------------------------+------------------------------------
             Reporter:  gdgsdg123    |                    Owner:
                 Type:  enhancement  |                   Status:  closed
             Priority:  normal       |                Component:  avformat
              Version:  git-master   |               Resolution:  invalid
             Keywords:  concat       |               Blocked By:
             Blocking:               |  Reproduced by developer:  0
Analyzed by developer:  0            |
-------------------------------------+------------------------------------
Changes (by gdgsdg123):

 * type:  defect => enhancement


Comment:

 Replying to [comment:3 cus]:
 > This is a simple timestamp precision issue. Input file is 60 fps,
 therefore the timestamp of the last frame is *exactly* 5287/60 which is
 88.116666666666 (infinite number of sixes), which is indeed smaller then
 the outpoint you specified 88.116667 therefore the last frame is included.
 Maybe invalid but it doesn't answer the real question...
 [[BR]]
 [[BR]]
 As those are values reported by FFprobe.

 [ticket:8460 gdgsdg123]:
 > {{{
 > ffprobe -hide_banner -show_entries "frame" -select_streams v:0 -of "xml"
 "temp.avi"
 > }}}
 > {{{
 > ffprobe -hide_banner -show_entries "frame=best_effort_timestamp_time"
 -select_streams v:0 -of "compact=nk=1:p=0" "temp.avi"
 > }}}
 [[BR]]
 [[BR]]
 [[BR]]
 Replying to [comment:3 cus]:
 > I suggest you round downwards when you specify the outpoint if you dont
 want to include the frame at outpoint.
 This approach seemed to produce a reliable output with various `outpoint`
 values tried. (not intensively tested)

--
Ticket URL: <https://trac.ffmpeg.org/ticket/8464#comment:4>
FFmpeg <https://ffmpeg.org>
FFmpeg issue tracker


More information about the FFmpeg-trac mailing list