Don't think so, none of their previous non GF100 gpu's had L2 cache, and they support CUDA as well. OpenCL doesn't require L2 either. I don't see the need for a mainstream GPU to have L2...Originally Posted by saaya
huh? but dont they need at least a tiny piece of l2 to NOT break cuda support and all?
Nice point, guess we have to ask him what was on his mind.
Reply With Quote
Bookmarks