[amelius]

> The media keeps missing the real lock in Nvidia has: CUDA. It's not the hardware. It's the ability for someone to use it painlessly.

Really? What if someone writes a new back-end to PyTorch, TensorFlow and perhaps a few other popular libraries? Then will CUDA still matter that much?

[p1esk]

if someone writes a new back-end to PyTorch

If that was easy to do surely AMD would have done it by now? After many years of trying?

[blitzar]

I am starting to wonder if AMD were even trying all this time.

[fritzo]

PyTorch has had an XLA backend for years. I don't know how performant it is though. https://pytorch.org/xla

[m00x]

It's pretty fast, just not as nice to use. You need statically defined tensors, and some functions are just not supported (last time I used it).

[pjmlp]

Can you do Unreal engine's Nanite, or Otoy Ray tracing in Pytorch?