nAo said:
Why would you 'waste' CELL cycles to do some vertex shading when you have a GPU designed for it?
IMHO it's better to use CELL for other, even complementary, tasks.
Yep.
Shifty Geezer said:
Seems to me Cell would ideally have been paired up with Xenos, to switch that whole GPU over to pixel shading and have Cell process the vertices. Unified shaders would offer more flexibility than fixed pipes.
I have had the exact same thought! Xenos could be used as a traditional GPU or every shader ALU could be dedicated to PS and have CELL do VS.
one said:
If unified shader is better, I guess Cell-based GPU might not be discarded in the first place... just my 2 cents.
What?
What does the success/failure of the CELL-based GPU's have to do with Unified Shader Architecture being good or bad?
It could very well be STI's (or Toshiba's)
design was flawed, or the fundamental aspects of the CELL implimentation are not as effecient as a totally dedicated PS hardware, or it could be lack of IPs, or they just stunk it up (like XGI, Matrox, S3, 3DLabs, and every other company that has dropped out of the 3D race).
NV, once very negative toward unified shading, has admitted they are going to it. ATI already spent a lot of research to get the R400 up and now has a working commercial product in R500 and will be going retail with the R600.
Sorry one, your 2 cents don't make much sense. Just because Sony did not go with a CELL based GPU does not mean unified shaders are inferior. Facts is both ATI and NV are going that direction. The only thing inferior was STI's attempt and the downplaying of Sony's competition.