[x264-devel] pic_struct / hrd patch

[Trahald]

the cpb_removal_delay should be the same as if the frame were not an IDR. ie
should not reset. The frame after that IDR however should act as tho the IDR
DID reset.
ie
CPB REMOVAL DELAY
idr p  b  b  b  p  p   IDR  p  b  b
0   2  4  6  8 10 12 14    2  4  6

On Sun, Sep 20, 2009 at 2:55 AM, Alex Giladi <alex.giladi at gmail.com> wrote:

> Trahald,
> In case we use cpb=0 only after the first buffering SEI,  what should
> be happening in case of a random access point, i.e. starting a
> playback from an arbitrary IDR (assuming that we repeat headers in
> front of each IDR)?
> Alex.
>
> On Sat, Sep 19, 2009 at 10:23 AM, Trahald <wewk584 at gmail.com> wrote:
> > Elecard buffer analyser gives the message
> >
> > "Can't continue the analysis.
> >
> > The value of cpb_removal_delay equals 0 in the fist access unit of a
> > buffering period that does not initialise the HRD. It will ause invalid
> > timings of coded picture removal"
> >
> > On Sat, Sep 12, 2009 at 10:47 PM, Jason Garrett-Glaser
> > <darkshikari at gmail.com> wrote:
> >>
> >> On Sat, Sep 12, 2009 at 7:36 PM, Alex Giladi <alex.giladi at gmail.com>
> >> wrote:
> >> > All,
> >> > The attached patch adds support for buffering and picture timing
> >> > SEI's, allowing use of pulldown or/and writing NAL HRD.
> >> > The patch borrows a lot from two earlier patches, by Alex Izvorski and
> >> > by an unknown author (patch found on x264.nl).
> >> > Hope you'll find it useful.
> >> > Alex.
> >>
> >> This patch is probably slightly wrong due to the fact that, like
> >> Trahald's, it ignores NAL escape bytes, which do count as part of NAL
> >> HRD.
> >>
> >> I am working on a fix for this.
> >>
> >> Dark Shikari
> >> _______________________________________________
> >> x264-devel mailing list
> >> x264-devel at videolan.org
> >> http://mailman.videolan.org/listinfo/x264-devel
> >
> >
> > _______________________________________________
> > x264-devel mailing list
> > x264-devel at videolan.org
> > http://mailman.videolan.org/listinfo/x264-devel
> >
> >
> _______________________________________________
> x264-devel mailing list
> x264-devel at videolan.org
> http://mailman.videolan.org/listinfo/x264-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x264-devel/attachments/20090920/a58b58b8/attachment.htm>