- cross-posted to:
- [email protected]
- [email protected]
- cross-posted to:
- [email protected]
- [email protected]
You must log in or register to comment.
That’s about time.
It’s only a matter of time. Just like physx and any other Nvidia gimmick, AMD will catch up and offer it for a fraction of the price.
Good. For my last GPU is was forced to go with Nvidia since I needed CUDA. ROCm was useless for consumer GPU’s back then. Support has gotten better since from what I’ve seen, so hopefully they continue and my next GPU will be team red again.
Can you run ML stuff on it? And why aren’t frameworks using Vulkan or OpenCL, if they’re both open standards?
Yes but barely, amd had a trash software group which has gotten much better.
But the nvidia stack is still sophisticated beyond belief, Cuda is still the foundation and the amount of effort nvidia put in is incredible.
They are modern day robber barons but they built a beautiful stack.
They should just let this mess quietly die and join Intel with their attempt at a cross-vendor standard.
Press X to doubt
Finally
Right! My thought was “fucking better be!” AMD lagging behind on general purpose compute is super disappointing.
Paywalled.
What is this about anyway ?Edit : Wow ! Great answers and nice project. Always good to see competition against Nvidia.
I was able to read without subscription or paywall. This is about AMD writing software that is open source and competes directly with Nvidia’s CUDA software framework.
Unable to read the paywalled article, but ROCm is like Nvidia’s CUDA for their GOUs. These used to be supported on and off for consumer GPUs, but get aupported mostly only for datacenter ones. Linux used to be supported, but they are bringing ROCm to windows as well.
Probably AMD is planning to improve support for consumer GPUs, since their GPUs are competitive to Nvidia’s at a lower price point, given that local LLMs, Image generation and other AIs are in a kind of a booming trend.
Currently, Nvidia has CUDA which is more or less the industry standard, so AMD with their ROCm and Intel with their OpenVino etc are trying to chip away at the monopoly.
weird, it isn’t for me. it is about AMDs equivalent of nVidias CUDA. AMD is trying to catch up in the area that they have unfortunately neglected for a very long time and in which nVidia has an (almost) unassailable lead. That means making things like machine learning / AI on their GPUs possible, or more easily and on more cards (especially more consumer cards) available.
I had to abandon a project recently because AMDs shitty software would refuse to install opencl for one of their integrated graphics chips.
Every time I’ve tried to use their subpar hardware it’s been met by exponentially more shitty subpar software. If I were in any business making decisions with real money I’d avoid AMD like a plague
I have a MacBook in my closet I can’t use at all because of an Nvidia integrated GPU. I’ve had nothing but problems with their shitty software too. Never had a single problem since I switched to AMD, and it was wayyyy cheaper. Your issue sounds like PEBCAK to me.
You know opencl AMD installations have problems if you have previously installed Nvidia software? So it sounds like your experience with AMD may have been soured by NVIDIA actually, and then you went out and gave them more money.
Guy. It was a prepackaged OEM device and the Nvidia install script bugged out on a fresh Ubuntu install. I don’t know what to tell you.
Guy.
