Re:PyTorch 1.8, with AMD ROCm support
by NavinF on Friday March 05, 2021 @02:20AM
(#26353836)
Oh oof. Thanks for saving me time not having to look up ROCm benchmarks. I find it really surprising they they don't wanna compete on performance/$ at all by not supporting consumer cards
[ Parent
| on HN ]
No worries, for future reference you can check here (hopefully that page will report improved support in the future)
I think this is more of an issue that they have Compute optimised and Graphics optimised cards and Vega is their last compute optimised card.
It would be very nice for them to refresh their compute cards as well.
Going for market segmentation like that sounds like a pretty bad idea if you are already the underdog in the game.
“Making science accessible to everyone is important to us. That's one of the reasons why GeForce cards can run CUDA.” — Bryce Lelbach at Nvidia
AMD meanwhile considers GPU compute as a premium feature... which is a different approach.
Not having a shippable IR like PTX but explicitly targeting a given GPU ISA, making this unshippable outside of HPC and supporting Linux only also points in that direction.
Intel will end up being a much better option than AMD for GPU compute once they ship dGPUs... in their first gen.
Especially since more and more consumer workloads use compute as well. DLSS makes real time ray tracing at acceptable resolutions possible, your phone camera uses neural nets for every photo you take, Nvidias OptiX noise reduction is very impressive, and so on.
AMD doesn't seem to want to be part of future computer usages, it's a shame. "Armies prepare to fight their last war, rather than their next war".
They made new compute cards, but they aren’t available to customers. (Only businesses, under the Radeon Instinct brand)
With the price to match for those...
They aren’t available for business either, you can buy a Tesla card in a microcenter or on Newegg, and through a million partners ranging from small system builders to Dell and HP.
Good luck getting an instinct card.