A technical question re: Xenos and RSX

I think I've read something about developers liking the RSX because of it's efficiency in rejecting pixels (early Z???) to avoid wasting bandwidth from overdraw...

If only I could remember where I read it...

It's probably here in the forum somewhere... Anyone want to lend a hand and link it?

Now that is an interesting question. I remember reading somewhere about a backface cull being supported in the RSX. Could this be a major benefit?
 
wouldn't most bandwidth comparisons be meaningless anyway because the ROP functions of xenos have been moved to the daughter die? i mean, you can't just say it takes nGB/s to render this frame on both systems and have it be a meaningful number, because parts of xenos have considerably more bandwidth than other parts (256GB/s from ROPs to eDRam, for example).
 
Now that is an interesting question. I remember reading somewhere about a backface cull being supported in the RSX. Could this be a major benefit?
Backface culling is not a feature unique to RSX. It is common in GPUs.
 
Back
Top