Also, try: ffplay -max_delay 50000 -sync video -loglevel debug rtsp:// 128.197.178.104/mpeg4/media.amp
(replace the rtsp://xxx with the url of your source video) if rtp packets are being dropped, there will be bright yellow warnings in the error output that look something like: [mpeg4 @ 027A5B40] RTP: missed 1 packetsq= 363KB sq= 0B f=0/0 . --luke On Tue, Nov 22, 2011 at 5:48 PM, Luke Clemens <lclem...@gmail.com> wrote: > i'm the guy who originally wrote that "proposed fix" for the bug. the fix > definitely resolves the issue with many ip cameras, but i can say with > certainty that if you're having the problem with TCP as well, then you have > a different problem. also while some of the errors are the same (like the > MB ones), there are some that i've never seen. i agree with camera man - if > you still have the problem even with TCP, then it can't be the network MTU. > my guess would be there is something in your own sending software that's > corrupting the stream. maybe a codec setting in lib555 or something? > > try enabling debug mode and then a super simple app like: > > while (1) { av_read_frame(ctx, pkt); } > > if you have any errors in that, then you can probably blame the RTP layer. > if not, then it is most likely an h.264 thing. sorry i can't help much, but > at least it's a start... > > --luke > > On Fri, Nov 11, 2011 at 7:21 AM, Johannes Ebersold < > johannes.ebers...@dfki.de> wrote: > >> ** >> On 11/11/2011 02:54 PM, Camera Man wrote: >> >> On 11/11/2011 08:27 AM, Andrey Utkin wrote: >> >> What is the sender app? >> What is transport protocol - TCP or UDP? >> Seems possible to me that UDP packets get trunkated to network MTU. >> That explains why both your decoder and ffplay playback it badly - you >> receive stream that's already corrupted. >> >> Note that if it's UDP, there's apparently a bug that appears with many >> IP cameras, see this thread: >> <http://www.mail-archive.com/libav-user@ffmpeg.org/msg00442.html><http://www.mail-archive.com/libav-user@ffmpeg.org/msg00442.html>for >> a description of the bug (and some sort of workaround). >> >> If you're using TCP, this bug is not to blame. >> >> >> _______________________________________________ >> Libav-user mailing >> listLibav-user@ffmpeg.orghttp://ffmpeg.org/mailman/listinfo/libav-user >> >> >> Hi, >> >> thanks for the hint. I saw this thread and the proposed fixed and tried >> to fix the bug. But, as mentioned, this did not work, even disabling >> pthreads completely did not fix the problem. It seems, like this issue is >> not related to my issue, although the problems described in this thread >> (and ticket 285) are similar to mine. >> Furthermore does my issue appear with TCP as wells as UDP. >> >> Regards, Johannes >> >> _______________________________________________ >> Libav-user mailing list >> Libav-user@ffmpeg.org >> http://ffmpeg.org/mailman/listinfo/libav-user >> >> > > > -- > - > - > - > - > Luke Clemens > http://clemens.bytehammer.com > -- - - - - Luke Clemens http://clemens.bytehammer.com
_______________________________________________ Libav-user mailing list Libav-user@ffmpeg.org http://ffmpeg.org/mailman/listinfo/libav-user