[FFmpeg-devel] GoC 2008 and H264 SVC decoder

Nico Sabbi Nicola.Sabbi
Mon Mar 31 12:20:55 CEST 2008


On Monday 31 March 2008 11:59:01 Nicolas George wrote:
> Le duodi 12 germinal, an CCXVI, 

;)))

> > because both formats are very bad, but for different reasons
> > explained countless times in this ML
>
> Could you, without too much trouble, find one of these messages? I
> would like to try and dump the arguments in <URL:
> http://wiki.multimedia.cx/index.php?title=Matroska
>

search Steve Lhomme's posts: he even went as far as
denying that all the mess needed to disentangle different
codecs in mkv is a serious problem

>  >, in a "shortcomings" section (or rather "pro and cons", as I am
>  > always
>
> suspicious of references that only give one point of view).
>
> > IIRC formats that have both pts and dts are few: mpeg*,
> > nut, mkv, mov, maybe asf;  surely not avi and ogg. All other
> > formats aren't worth considering for carrying H264
>
> By the way, that is a question that was bothering me for a long
> time: what is the point of DTS (decoding timestamps, right?)? And
> why is it necessary for H.264 and not other codecs?

for all codecs, not only for H264, but H264 also has B-pyramid
that require a massive frame reordering.
There's a nice and long thread in nut-devel (search broadcast)
that discusses the importance of DTS, PTS and PCR in a multiplex

>
> My naive idea is that a frame needs to be decoded before it is
> presented, and after all frames it depends on have been decoded,
> and that is enough. Do you have any references where this is
> explained?
>

ISO-13818-1.







More information about the ffmpeg-devel mailing list