[Ffmpeg-devel] Partial frame decode in avcodec?

Matt Schuckmann mschuck
Thu May 11 00:21:50 CEST 2006


Everything I've got says it's an implementation of MPEG4 that has been
modified (not very well) to support very large hi-res frames. In it's
current state B frames are typically larger than I frames if there is any
motion at all.

Did my statement about no P frames throw you? Maybe it supports them but we
don't use them, only I frames and B frames, I'm told because it's easier to
do reverse playback without P frames. 



-----Original Message-----
From: ffmpeg-devel-bounces at mplayerhq.hu
[mailto:ffmpeg-devel-bounces at mplayerhq.hu] On Behalf Of M?ns Rullg?rd
Sent: Wednesday, May 10, 2006 2:44 PM
To: FFMpeg development discussions and patches
Subject: Re: [Ffmpeg-devel] Partial frame decode in avcodec?

"Matt Schuckmann" <mschuck at imoveinc.com> writes:

>> What happens if a predicted block depends on pixels outside the
>> region?  Are you sure your old decoder didn't simply decode the full
>> frame and crop it for you?
>
> I don't know about the predictive block question I haven't looked into it
to
> closely. I'm kind of new to video compression but maybe the fact that the
> old codec doesn't use P frames has something to do with it or maybe it
goes
> back and decodes just those blocks in needs. 

What is this old codec really?  I thought you said it was an mpeg4
implementation, but maybe I misunderstood you.

> I'm certain the old codec only decodes the specified region (or something
> close to it) because testing shows that if I have it decode only 1/4 of
the
> frame the processing time drops by about 75%. 

That sounds rather convincing.

-- 
M?ns Rullg?rd
mru at inprovide.com

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel at mplayerhq.hu
http://mplayerhq.hu/mailman/listinfo/ffmpeg-devel





More information about the ffmpeg-devel mailing list