[FFmpeg-devel] Supporting container formats with PTS gaps (Ogg)

Michael Niedermayer michaelni
Mon Feb 2 21:19:43 CET 2009


On Mon, Feb 02, 2009 at 11:56:01AM -0800, Baptiste Coudurier wrote:
> Michael Niedermayer wrote:
> > On Mon, Feb 02, 2009 at 06:17:30PM +0200, Tomer Barletz wrote:
> >> On Mon, Feb 2, 2009 at 5:53 PM, Michael Niedermayer <michaelni at gmx.at> wrote:
> >>> mpeg does not store all timestamps, that is true,
> >> What do you mean by 'mpeg'? do you mean mpeg-2 video?
> > 
> > mpeg 1 and 2, PS and TS
> > 
> > 
> >>> but the mpeg muxer,
> >>> at least ours needs all timestamps (or at least more than the minimum
> >>> guranteed from mpeg input). Kinda tricky to conform to the
> >>> buffer and interleaving requirements of mpeg without knowing the
> >>> timestamps at all.
> >> As far as I've seen, the PES encoder does not generate a PES header
> >> for each frame, so maybe I don't understand something - but the muxer
> >> definitely does not create all timestamps.
> > 
> > there are limits on the buffer sizes, if you do not know the dts you do
> > not know when a frame is removed thus you do not know if your buffer will
> > overflow, underflow or neither.
> > 
> > and please keep in mind we dont implement a suboptimal CBR-only muxer but
> > try to use the buffers to their full extend permitted by the spec.
> 
> Indeed.
> 

> > and there is no such thing called a PES encoder
> 
> There could be, I've seen such files in the wild.

of course, ive seen them too, i just meant we dont have a seperate PES
encoder/muxer.

[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Dictatorship naturally arises out of democracy, and the most aggravated
form of tyranny and slavery out of the most extreme liberty. -- Plato
-------------- 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/20090202/e77b63c3/attachment.pgp>



More information about the ffmpeg-devel mailing list