[MPlayer-dev-eng] mencoder problem
michaelni at gmx.at
Fri May 17 19:21:34 CEST 2002
On Friday 17 May 2002 18:42, D Richard Felker III wrote:
> > it does redistribute the bits, otherwise its a bug ...
> Well then why do I have a file encoded by lavc 2pass that's 100+ megs
> smaller than I requested, but not all the frames are qscale=3? It
> seems something is amis. Someone familiar with lavc 2pass control
> should probably check this, but I'm pretty sure there's some bug in
what did u set qmin to? there was a bug in there for qmin=1 ... (its fixed
and a tiny bit more info might help too (command line options,
> > > At least in the meantime until lavc's 2pass code is improved to deal
> > > with this situation, and perhaps in the longterm as well, it's
> > > probably desirable NOT to use qmin/max in the second pass of encoding.
> > > I propose the following possible solution, although I don't know if
> > > it's the best or most intuitive:
> > >
> > > In mencoder, have the default values for vqmin/vqmax be 0, which will
> > > mean initialize defaults based on pass number. For vpass 1, it would
> > > just initialize them with the old defaults that exist now. For vpass
> > > 2, it would initialize them to 1 and 31.
> > i doubt that would be good, why dont u just set vqmin=1 vqmax=31 for both
> > pass 1 and 2 ?
> I was under the impression that letting the qscale vary too widely in
> pass1 would be bad for estimating relative bitrate requirements,
> unless the bitrate tolerance is set really high. Right now, with
> vqmin=3 on the first pass, average bitrate often stays well below the
> requested bitrate on low motion scenes, leaving plenty of bits for the
> high motion scenes to get at least a decent qscale later. I would
> expect that if high motion scenes get *really* high qscales on the
> first pass, it would be harder to estimate what qscale should be used
> for them on the second pass. Of course I might just have no idea what
> I'm talking about, however.
sure u are right the more the qscale changes between the passes the more
problems there are but if the first pass is limited to qscale>=3 and the 2.
pass isnt than u have the same effect ...
More information about the MPlayer-dev-eng