Re: Towards Better Anti-aliasing

On Wed, Nov 2, 2011 at 10:11 AM, Mike Bostock <mbostock@cs.stanford.edu>wrote:

> > One reason could be that FSAA is (with the most common supersampling
> > schemes) not that precise, compared to some of the alternatives (it's a
> > rather crude, albeit quite effective, method).
>
> Point taken; any discrete anti-aliasing system is going to be less
> precise than a geometric solution (computing the exact sub-pixel area
> occluded by the front-most shape). But the hardware implementation for
> multisampling is quite good in my experience, and more than enough to
> compensate for the artifacts of not-FSAA.
>

On a lot of hardware/driver combinations, anything more than 2x2 MSAA is
excruciatingly slow. And 2x2 MSAA is very poor quality for a lot of
applications.

I wish MSAA worked better, because we could remove a lot of complexity from
our layout and rendering stack if we could rely on it, but we can't.

Rob
-- 
"If we claim to be without sin, we deceive ourselves and the truth is not
in us. If we confess our sins, he is faithful and just and will forgive us
our sins and purify us from all unrighteousness. If we claim we have not
sinned, we make him out to be a liar and his word is not in us." [1 John
1:8-10]