[Sandy] tri-only branch?

[Makc]

then... test and report!

On 6/17/08, Thomas Pfeiffer <kiroukou at gmail.com> wrote:
> >
> > well, we may change color material to use single beginFill - endFill
> > per shape3d, if that is *really* the only place where quads outperform
> > tris. I still think it was because of doubled number of array access.
> >
>
> If that is true, QUAD mode and bitmapmaterial with matrix cache disabled
> would be faster than TRI.
> (I mean we disable the mapping matrix cache in all cases).
> If QUAD is faster here too, that is array access which slow down everything
> in Sandy pipeline.
> _______________________________________________
> Sandy mailing list
> Sandy at osflash.org
> http://osflash.org/mailman/listinfo/sandy_osflash.org
>