[FFmpeg-devel] About guess_correct_pts / AVFrame.best_effort_timestamp

[Nicolas George]

Le nonidi 29 pluvi?se, an CCXIX, Luca Barbato a ?crit?:
> because h264 in nut or h264 in mkv shouldn't be interpreted as h264 in
> avi just to name one...

To what extent should it be interpreted differently?

I thought that, specifically for H.264, concatenating the payload of the
packets would just give an elementary stream that would decode just the
same.

> If you want to extract information from the codec and use it to help
> getting valid information to sync streams you should do that in a place
> in which you are aware of ALL the information available, this place
> usually it's outside the decoder...

I try to follow your reasoning: the place where all the information is
available is after the decoder, isn't it? Before the decoder, all the
information embedded in the payload is not decoded yet.

Regards,

-- 
  Nicolas George
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20110218/b299562c/attachment.pgp>