[FFmpeg-devel] [bug/patch] MPEG-TS muxer: PCR not in sync with PTS/DTS

[Michael Niedermayer]

On Mon, Feb 08, 2010 at 12:52:32PM -0500, Mike Scheutzow wrote:
> > I'm having a problem with this patch. I'm writing a segmenter that
> > needs to produce TS stream with packets that have pts and dts delayed
> > by an offset. The patch however cause lot of empty packets at the
> > beginning of each segment (to compensate for delayed pts/dts).
> >
> > Is there any way to avoid this?
> >
> > Thanks.
> > -Matej
>
> Please don't top-post on this list.
>
> It is my understanding that the mpeg-ts muxer is supposed to generate 
> Constant Bitrate (CBR) transport streams.  In the current code, the PCR 
> *always* starts at 0, so the behavior you call "a problem" is actually the 
> correct behavior.
>
> One solution would be to add a new feature to the mpeg-ts mux so that the 
> initial PCR can be specified by the caller (using metadata?) or maybe 
> calculated from the first DTS received.
>
> I'll work on a patch if Baptiste indicates he is OK with the concept.

iam not baptiste but i think
1. we need VBR support
2. its the muxers job to figure all timestamps out from the DTS/PTS
   do proper interleaving and all that in a fully system/video/audio
   compliant way (as long as the codec streams are compliant)
3. only add minimal overhead

our PS muxer can do all that,what is so different on TS that the existing
code from our PS muxer cannot just be used?

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Complexity theory is the science of finding the exact solution to an
approximation. Benchmarking OTOH is finding an approximation of the exact
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20100208/ced1bf42/attachment.pgp>