I would like to point out that their research seems lacking.
The industry is not "moving to one big GPU".
The industry has already moved to 4, 6, or 8 "big" GPUs per station/node.
The industry is not "moving to one big GPU".
The industry has already moved to 4, 6, or 8 "big" GPUs per station/node.
I am absolutely curious as to what you're talking about because I have never seen this before, please correct me because I want to learn more.
Apple is talking about prosumers, not the rendering firms that need such massive GPU workloads. They'd be better off using server farms to render that much work. No one is going to work next to a noisy workstation with fans that has to disperse 2,000 watt worth of heat at full load.
Seemed more of a case that no one wants to actually code for OpenCL on AMD...