AMD funded a drop-in CUDA implementation built on ROCm: It's now open-source • Hacker News Board

Re: AMD funded a drop-in CUDA implementation built on ROCm: It's now open-source

#41

Post by rubatuga » Mon, Feb 12, 2024, 3:06 PM UTC

blagie wrote: post #23

otoburb wrote: post #11

Earlier quoted context omitted.

They would always be at the mercy of NVIDIA's API. Without knowing the inner workings, perhaps a major concern with this approach is the need to implement on NVIDIA's schedule instead of AMD's which is a very reactive stance. This approach actually would make sense if AMD felt, like most of us perhaps, that the NVIDIA ecosystem is too entrenched, but perhaps they made the decision recently to discontinue funding beca…

They've been at mercy of Intel x86 APIs for a long time. Didn't kill them. What happens here is that the original vendor loses control of the API once there are multiple implementations. That's the best possible outcome for AMD. In either case, they have a limited window to be adopted, and that's more important. The abstraction layer here helps too. AMD code is !@#$%. If this were adopted, it makes it easier to fix t…

x86 is not the same, the courts forced the release of x86 architecture to AMD during an antitrust lawsuit

Re: AMD funded a drop-in CUDA implementation built on ROCm: It's now open-source

#42

Post by kkielhofner » Mon, Feb 12, 2024, 3:07 PM UTC

fariszr wrote: post #10

btown wrote: post #8

Why would this not be AMD’s top priority among priorities? Someone recently likened the situation to an Iron Age where NVIDIA owns all the iron. And this sounds like AMD knowing about a new source of ore and not even being willing to sink a single engineer’s salary into exploration. My only guess is they have a parallel skunkworks working on the same thing, but in a way that they can keep it closed-source - that this…

According to the article, AMD seems to have pulled the plug on this as they think it will hinder ROCMv6 adoption, which still btw only supports two consumer cards out of their entire line up[1] 1. https://www.phoronix.com/news/AMD-ROCm-6.0-Released

With the most recent card being their one year old flagship ($1k) consumer GPU...

Meanwhile CUDA supports anything with Nvidia stamped on it before it's even released. They'll even go as far as doing things like adding support for new GPUs/compute families to older CUDA versions (see Hopper/Ada and CUDA 11.8).

You can go out and buy any Nvidia GPU the day of release, take it home, plug it in, and everything just works. This is what people expect.

AMD seems to have no clue that this level of usability is what it will take to actually compete with Nvidia and it's a real shame - their hardware is great.

Re: AMD funded a drop-in CUDA implementation built on ROCm: It's now open-source

#43

Post by NekkoDroid » Mon, Feb 12, 2024, 3:07 PM UTC

marcus0x62 wrote: post #29

AndrewKemendo wrote: post #17

ROCm is not spelled out anywhere in their documentation and the best answers in search come from Github and not AMD official documents "Radeon Open Compute Platform" https://github.com/ROCm/ROCm/issues/1628 And they wonder why they are losing. Branding absolutely matters.

That, and it only runs on a handful of their GPUs.

If you are talking about the "supported" list of GPUs, those listed are only the ones they fully validate and QA test, other of same gen are likely to work, but most likely with some bumps along the way. In one of the a bit older phoronix posts about ROCm one of their engeneers did say they are trying to expand the list of validated & QA'd cards, as well as destinguishing between "validated", "supported" and "non-functional"

Re: AMD funded a drop-in CUDA implementation built on ROCm: It's now open-source

#44

Post by hnlmorg » Mon, Feb 12, 2024, 3:08 PM UTC

izacus wrote: post #19

btown wrote: post #8

Why would this not be AMD’s top priority among priorities? Someone recently likened the situation to an Iron Age where NVIDIA owns all the iron. And this sounds like AMD knowing about a new source of ore and not even being willing to sink a single engineer’s salary into exploration. My only guess is they have a parallel skunkworks working on the same thing, but in a way that they can keep it closed-source - that this…

Why do you think running after nVidia for this submarket is a good idea for them? The AMD GPU team isn't especially big and the development investment is massive. Moreover, they'll have the opportunity cost for projects they're now dominating in (all game consoles for example). Do you expect them to be able to capitalize on the AI fad so much (and quickly enough!) that it's worth dropping the ball on projects they're…

GPU for compute has been a thing since the 00s. Regardless of whether AI is a fad (it isn't, but we can agree to disagree on this one) not investing more in GPU compute is a weird decision.

Re: AMD funded a drop-in CUDA implementation built on ROCm: It's now open-source

#45

Post by FPGAhacker » Mon, Feb 12, 2024, 3:09 PM UTC

izacus wrote: post #19

btown wrote: post #8

Why would this not be AMD’s top priority among priorities? Someone recently likened the situation to an Iron Age where NVIDIA owns all the iron. And this sounds like AMD knowing about a new source of ore and not even being willing to sink a single engineer’s salary into exploration. My only guess is they have a parallel skunkworks working on the same thing, but in a way that they can keep it closed-source - that this…

Why do you think running after nVidia for this submarket is a good idea for them? The AMD GPU team isn't especially big and the development investment is massive. Moreover, they'll have the opportunity cost for projects they're now dominating in (all game consoles for example). Do you expect them to be able to capitalize on the AI fad so much (and quickly enough!) that it's worth dropping the ball on projects they're…

It was Microsoft’s strategy for several decades (outsiders called it embrace, extend, extinguish, only partially in jest). It can work for some companies.

Re: AMD funded a drop-in CUDA implementation built on ROCm: It's now open-source

#46

Post by SushiHippie » Mon, Feb 12, 2024, 3:10 PM UTC

sam_goody wrote: post #16

Aside from the latest commit, there has been no activity for almost 3 years (latest code change on Feb 22, 2021). People are criticizing AMD for dropping this, but it makes sense to stop paying for development when the dev has stopped doing the work, no? And if he means that AMD stopped paying 3 years ago - well, that was before dinosaurs and ChatGPT, and alot has changed since then. https://github.com/vosen/ZLUDA/co…

The code prior to this was all for the intel gpu zluda, and then the latest commit is all the amd zluda code, hence why the commit talks about the red team

Re: AMD funded a drop-in CUDA implementation built on ROCm: It's now open-source

#47

Post by currymj » Mon, Feb 12, 2024, 3:11 PM UTC

izacus wrote: post #19

btown wrote: post #8

Why would this not be AMD’s top priority among priorities? Someone recently likened the situation to an Iron Age where NVIDIA owns all the iron. And this sounds like AMD knowing about a new source of ore and not even being willing to sink a single engineer’s salary into exploration. My only guess is they have a parallel skunkworks working on the same thing, but in a way that they can keep it closed-source - that this…

Why do you think running after nVidia for this submarket is a good idea for them? The AMD GPU team isn't especially big and the development investment is massive. Moreover, they'll have the opportunity cost for projects they're now dominating in (all game consoles for example). Do you expect them to be able to capitalize on the AI fad so much (and quickly enough!) that it's worth dropping the ball on projects they're…

everyone buying GPUs for AI and scientific workloads wishes AMD was a viable option, and this has been true for almost a decade now.

the hardware is already good enough, people would be happy to use it and accept that's it's not quite as optimized for DL as Nvidia.

people would even accept that the software is not as optimized as CUDA, I think, as long as it is correct and reasonably fast.

the problem is just that every time i've tried it, it's been a pain in the ass to install and there are always weird bugs and crashes. I don't think it's hubris to say that they could fix these sorts of problems if they had the will.

Re: AMD funded a drop-in CUDA implementation built on ROCm: It's now open-source

#48

Post by mdre » Mon, Feb 12, 2024, 3:11 PM UTC

sam_goody wrote: post #5

I don't really follow this, but isn't it a bad sign for ROCm that, for example, ZLUDA + Blender 4's CUDA back-end delivers better performance than the native Radeon HIP back-end?

I'd say it's even worse, since for rendering Optix is like 30% faster than CUDA. But that requires the tensor cores. At this point AMD is waaay behind hardware wise.

Re: AMD funded a drop-in CUDA implementation built on ROCm: It's now open-source

#49

Post by bonton89 » Mon, Feb 12, 2024, 3:15 PM UTC

izacus wrote: post #19

btown wrote: post #8

Why would this not be AMD’s top priority among priorities? Someone recently likened the situation to an Iron Age where NVIDIA owns all the iron. And this sounds like AMD knowing about a new source of ore and not even being willing to sink a single engineer’s salary into exploration. My only guess is they have a parallel skunkworks working on the same thing, but in a way that they can keep it closed-source - that this…

Why do you think running after nVidia for this submarket is a good idea for them? The AMD GPU team isn't especially big and the development investment is massive. Moreover, they'll have the opportunity cost for projects they're now dominating in (all game consoles for example). Do you expect them to be able to capitalize on the AI fad so much (and quickly enough!) that it's worth dropping the ball on projects they're…

AMD also has the problem that they make much better margins on their CPUs than on their GPUs and there are only so many TSMC wafers. So in a way making more GPUs is like burning up free money.

Re: AMD funded a drop-in CUDA implementation built on ROCm: It's now open-source

#50

Post by anon291 » Mon, Feb 12, 2024, 3:16 PM UTC

rubatuga wrote: post #41

blagie wrote: post #23

Earlier quoted context omitted.

They've been at mercy of Intel x86 APIs for a long time. Didn't kill them. What happens here is that the original vendor loses control of the API once there are multiple implementations. That's the best possible outcome for AMD. In either case, they have a limited window to be adopted, and that's more important. The abstraction layer here helps too. AMD code is !@#$%. If this were adopted, it makes it easier to fix t…

x86 is not the same, the courts forced the release of x86 architecture to AMD during an antitrust lawsuit

You don't think the courts would force the opening of CUDA? Didn't a court already rule that API cannot be patented. I believe it was a Google case. As long as no implementation was stolen, the API itself is not able to be copyrighted.

Here it is: https://arstechnica.com/tech-policy/2021/04/how-the-supreme-...