[XviD-devel] TOOSMALL-LIMIT + Trellis

Michael Militzer michael at xvid.org
Wed Sep 10 00:55:23 CEST 2003


Quoting Edouard Gomez <ed.gomez at free.fr>:

> Michael Militzer (michael at xvid.org) wrote:
> > > [1] The patch is still available for review:
> > > http://ed.gomez.free.fr/vrac/xvid-patches/mbtransquant-cleanup.diff
> > 
> > I'm not competent enough. (yeah, I'll say that always from now on when
> > it saves me  from further work ;-)) - Christoph or  Skal should have a
> > look...
> 
> Ok michael is not competent ;-) but what christoph?
> 
> Is at least the new code logical, not comparing things that can't be
> compared ? Just fill in the following form:
> [ ] Yes, it's logical, please apply.
> [ ] No, you got it wrong, don't commit.

If you just ask if it's logical to return the coefficient sum also after
trellis quantization, I'd say yes. It simply makes things more consistent
and easier to understand - however, it's generally not useful to perform
coefficient thresholding after trellis quantization (as said, results
after trellis quant _should_ be R-D optimal, so coeff thresholding can
only make results worse). But since TOO_SMALL_LIMIT is set to 1 by default,
that's no problem. And as I already explained, I want to have coeff thresh-
holding for cartoon mode and I want to perform it even with trellis
enabled. So I want your patch, it's logical, please apply ;-)

bye,
Michael


More information about the XviD-devel mailing list