It now supports AMD GPUs since 3 weeks ago, check the latest commit at the repo:

https://github.com/vosen/ZLUDA

The article also mentions exactly this fact.

That's exactly what the post from 3 weeks ago discussed.

AMD dropped ZLUDA funding, the remains were opensourced and it won't ever be a viable option.

> it won't ever be a viable option

For production workloads, I generally agree. It's an unsupported hack with a questionable future, I wouldn't do anything money-making with it.

However, for tinkering and consumer workloads, it already works pretty well. Enough of cuDNN and cuBLAS work to run PyTorch and in turn, Stable Diffusion with https://github.com/lshqqytiger/ZLUDA - there's even a fairly user-friendly setup process already in https://github.com/vladmandic/automatic .

I was able to get a personal non-ML related project working on my AMD card in just a few minutes, which saved me a lot of development time before I then deployed the production workload on NV hardware (this is probably why AMD pulled the plug on the project - it's almost more of a boost to NV than anything else, AMD really need people to be writing code on ROCm to deploy on AMD datacenter hardware).

Given that PyTorch supports AMD cards with ROCm support, I'm curious why would you be coding in CUDA (very low level!) as opposed to PyTorch ?

> a personal non-ML related project

Not everything is DNNs and tensors...

As for the Stable Diffusion thing, a silly edge case - because MIOpen (and therefore PyTorch-on-ROCm) doesn't work on Windows yet (they're slated to ship it next month, I think).

> Not everything is DNNs and tensors...

Sure, but you could use PyTorch for cuBLAS/cuRAND etc type functionality too.

I think it is a relevant zeitgeist reaction to NVIDIA banning translation of CUDA to non-NVIDIA hardware post from earlier.

by earlier do you mean 2021?

I mean with regard to the recency of posts on HN. The community sees a post, and then related posts begin to trickle in.

We upvote it because we found it interesting. What's the point of arguing against other people's interest? What do you think will come from your outrage at the Internet?

Yes, we did. Any other questions?

Your best option is not to upvote or comment on content you don't find interesting! That's how this all works.

Not everyone sits on hacker news 24/7 - I missed the previous post 22 days ago and I appreciate it being here again.

Let the community decide with upvotes, that’s the point.

Other people do not engage HN in the same pattern you do.

Do we really need such self centered complaints posted over and over?

Go touch grass. Seems like you spend too much time browsing this website.

Are you suggesting that it's wrong to presume to police other people's speech?