|
|
|
|
|
|
by throwaway2a02
1948 days ago
|
|
|
My understanding is that the ETH and other algorithms designed to be run on GPUs have big memory and memory bandwidth requirements, which is what makes it more difficult (or less cost effective?) to implement on an ASIC. LE. I see you meant NVidia should be making the ASICs. In that case, that's exactly what they're doing, isn't it? This CMP line is exactly that. |
|
|