Associate
- Joined
- 14 Aug 2017
- Posts
- 1,196
I imagine if Intel are successful in getting software developers (not game developers) to implement more AVX-512 instructions, it won't matter how many additional cores AMD have, as they'll simply not be able to use the more efficient, faster AVX path.
Or those apps will accelerate with the GPU instead, which is far more capable and a better fit for stuff like tensorflow.