I expect this would be the case with alpha compositing at least... Afaict subpixel rendering is supersampling width with multiples of 3, then modulo via the ho

I expect this would be the case with alpha compositing at least... Afaict subpixel rendering is supersampling width with multiples of 3, then modulo via the horizontal coordinate into rgb triplets and divide by supersampling factor. I guess if kerning and position are also subpixel it might become more tricky to cache results efficiently.