[MEncoder-users] Adding harddup as an afterthought?

[Rich Felker]

On Sat, Sep 29, 2007 at 08:49:17PM +0200, Reimar Döffinger wrote:
> > Let me restate my question: if decoder gets
> > 
> >   P 0 0 0 0 0 0 0 0 0 B P
> > 
> > how should it predict the movements of the B frame?  By 0.5 of
> > movement of (the second) P, or by 0.9 of it?
> 
> There is no "0" for the decoder, the decoder gets P B P.
> All the "0"-frames do is tell the _player_ that it should display the P
> frames for 10 times the normal time. This is just a hack for containers
> that do not support arbitrary timestamps but only fixed fps.
> Again, the decoder is in no way involved in handling the 0-frames. In a
> framework like ffmpeg actually nobody besides the demuxer will even be
> able to see them.

So to answer his question: 0.5, of course.

Rich