[FFmpeg-trac] #2667(avcodec:open): console spamming when decoding a h264 / mp4 stream
FFmpeg
trac at avcodec.org
Fri Jun 14 19:58:52 CEST 2013
#2667: console spamming when decoding a h264 / mp4 stream
------------------------------------+-----------------------------------
Reporter: gjdfgh | Owner:
Type: defect | Status: open
Priority: minor | Component: avcodec
Version: git-master | Resolution:
Keywords: h264 mov | Blocked By:
Blocking: | Reproduced by developer: 1
Analyzed by developer: 0 |
------------------------------------+-----------------------------------
Comment (by cehoyos):
Replying to [comment:13 ubitux]:
> Replying to [comment:11 cehoyos]:
> > > [...]
> > > The video is not supposed to be stalled like it is currently
> >
> > That brings me back to my original question:
> > Which application does not ''stall''?
>
> It seems totem/gstreamer is able to, according to the user.
totem plays the sample here exactly like ffplay and ffmpeg.
(Rereading the original post: Is it possible that this was just "wrong bug
tracker"?)
> > (Before you answer: Note that afaict most frames in the sample only
consist of "0" and "1" making decoding extra difficult imo if we do not
assume 1bit ppm or bmp.)
I just realize that this can be misunderstood: I meant h264 video packets
that consist only of 0x00 and 0x01 are most likely invalid;-)
> Are you sure you're not confused with some data in the (large) MP4
header?
Certainly possible, but is the large header really supposed to be in the
middle of the file?
> Also, the frames around this timestamp have very low differences, so
it's not surprising to have small data frames.
This is of course also possible but I would have expected those small data
frames to be >0 in size (many packets in the broken part of the file have
size 0) and not containing only 0x00 and 0x01.
--
Ticket URL: <https://ffmpeg.org/trac/ffmpeg/ticket/2667#comment:14>
FFmpeg <http://ffmpeg.org>
FFmpeg issue tracker
More information about the FFmpeg-trac
mailing list