[x264-devel] Re: non monotone timestamps...

[bond]

> Hi
>
> On Wed, Aug 30, 2006 at 08:36:16PM +0200, bond wrote:
> > > bond wrote:
> > > >>> [...]
> > > >>>
> > > >> 'ctts' atom is actually created correctly. If a problem resides, it
is
> > > >> most likely cause pts and/or dts of the packets are incorrect.
> > > >>
> > > >> Problem with B pyramid is just that lavf will not let packets pass
> > since
> > > >> they have non monotone timestamps as mentioned in the first post of
the
> > > >> thread.
> > > >
> > > > the ctts atom is NOT created correctly with b-pyramid in ffmpeg!
> > > >
> > > > i have just tested it again with revision 6129. i compared the ctts
to
> > what
> > > > mp4box and mp4creator create for the stream and let mp4box analyse
the
> > ctts
> > > > atom resulting into the logfile attached showing the errors
> > > >
> > > > my commandline used:
> > > > ffmpeg -s 352x144 -i "c:\...\matrix.yuv" -y -vcodec h264 -bf
3 -bpyramid
> > 1
> > > > ffmpeg.mp4
> > > >
> > >
> > > Can't you see "error, non monotone timestamps 2002 >= 1001"....
> >
> > yes, i see such errors
>
> this and the bpyramid in mov issue should be fixed in latest ffmpeg svn

works great, thanks a lot!

a minor error seems to be still there with the last two entries in the ctts
atom:
ffmpeg writes:
Sample 40 - DTS 39 - CTS 39
Sample 41 - DTS 40 - CTS 41
Sample 42 - DTS 41 - CTS 46
Sample 43 - DTS 42 - CTS 44
Sample 44 - DTS 43 - CTS 43
Sample 45 - DTS 44 - CTS 45
Sample 46 - DTS 45 - CTS 49
Sample 47 - DTS 46 - CTS 48

as you can see CTS time 47 is missing, instead it jumps directly to 48

mp4creator writes it correctly this way:
Sample 40 - DTS 39 - CTS 39
Sample 41 - DTS 40 - CTS 41
Sample 42 - DTS 41 - CTS 46
Sample 43 - DTS 42 - CTS 44
Sample 44 - DTS 43 - CTS 43
Sample 45 - DTS 44 - CTS 45
Sample 46 - DTS 45 - CTS 48
Sample 47 - DTS 46 - CTS 47

-- 
This is the x264-devel mailing-list
To unsubscribe, go to: http://developers.videolan.org/lists.html