[MPlayer-users] lavc vs. xvid (and improving lavc quality)

[Loren Merritt]

On Tue, 8 Jun 2004, Wayde Milas wrote:
> On Tue, 2004-06-08 at 04:07, Alexei Khlebnikov wrote:
> > On Mon, 07 Jun 2004 17:45:56 -0400
> > Jason Tackaberry <tack at sault.org> wrote:
> >
> > > I played around a bit with the other cmp functions, and cmp=3 is a
> > > pretty good improvement over cmp=0.  Right now, at least, I prefer cmp=3
> > > with post-processing (spp,noise) over cmp=10 without post-processing.
> > > But for those times when I want to circulate video with the expectation
> > > that it won't have pp, NSSE is a significant improvement.  I get the
> > > best of both worlds now.  I have very little to complain about. :)
> >
> > Have you tried cmp=6? It's called "rate distortion optimal, slow", seems to be
> > the description of the best cmp method. I am especially interested in not-very-high
> > bitrates (1CD rips).
> >
> > Anyone else's comments are welcome too, off course. What is the best cmp method
> > for average movie, regadrless of speed?
>
> I have. 3 almost is always optimal for me, except for
> cartoons/animation, when I think 7? (cant remember) is.

On all the anime I've encoded, 2 (hadamard) and 3 (dct) have produced
essentially identical results (no psnr difference nor noticable
artifacts). Both are better than 6. So I use 2, because it's faster.

--Loren Merritt