July 28, 20169 yr the main problem with the performance of AMD cards and CPUs is that they are locked into a contract with Global Foundries for their silicon. if they could get it from TSMC like nVidia does, they could clock the cores to the same speeds and combined with all the innovations they have crammed in there, they could be way ahead But in the case with Blender, Blender supports CUDA GPU parallel rendering. Blender does have OpenCL support for a limited selection of AMD cards, but it is still early support, and not optimized.
Create an account or sign in to comment