On Saturday 05 March 2005 20:21, Daniel Phillips wrote:
> Hi all,
>
> I've taken some time to research FSAA methods in general, and
> supersampling in particular.  Spoiler: my conclusion is that we should
> implement 2x2 supersampling in the initial release of the FPGA
> prototype, and follow up later with rotated grid supersampling, and
> still later with a multisampling implementation.
<snip>

Excellent work Daniel!

Just a short question though, knowing next to nothing about this. How does the 
blending at the edges work theoretically? Alpha blending gives problems with 
continuity I'd think. Let's say we render two adjacent white polygons on a 
black background, and we render them one at a time. If we blend the edge 
pixels with the background then a pixel that is cut right through the middle 
will be 75% white instead of 100%. And if you render the same polygon twice 
then it will be 75% white instead of 50%. How do you prevent that?

The simple 2x2 supersampling approach does not have this problem, but how does 
this work for multisampling?

The reason I'm asking is that I'm wondering whether we can't cheat, and do 
something that gives somewhat better results than 2x2 antialiasing. I'm 
afraid that 2x2 won't be fast enough, and it won't be pretty enough either, 
making the entire thing a bit useless. Which is an opinion based on nothing, 
so I'd be happy to hear counterarguments :).

Lourens
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)

Reply via email to