RE: [AD] improvements on alpha blending...

["Vincent Penquerc'h" <vincent@xxxxxxxxxx>]

> and then with that table, say we wanna calculate the blending of
> pixels 255, 30, 10 and 20, 30, 150
> we sould only do
> r = average[255][20];
> g= average[30][30];
> b= average[10][150];

On second thought, this can't be done for alpha blending, as you'd
need 256 of these tables (you have three parameters). This can be
done only for transparent mode drawing (where the "alpha" stays
constant for the whole image.
It would still be a win there for large images, but the applicable
domain is smaller.

-- 
Lyrian