Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

What Is AMD ROCm?

threedots.ovh

Post reply on HN
Print view View original

31–40 of 57 posts

Re: What Is AMD ROCm?

#31

Post by edgyquant »

dagmx wrote: post #12
nicolaslem wrote: post #7

Earlier quoted context omitted.

On the other hand AMD has a fraction of the engineering resources Intel and Nvidia have. They need to make some choice and looking back at the last few years it seems that their choice to focus their efforts on hardware and gaming paid off.

On the flip side, I think that they haven't focused on a compelling compute story means that anyone doing anything other than pure gaming is better served by an Nvidia card.

Yeah I bought an AMD card a few years when they released the new architecture to support them. I ended up grabbing an nvidia card because I don’t play many games but I want to be able to run tensorflow etc but after a year AMD still had little support for any machine learning.

Re: What Is AMD ROCm?

#32

Post by shmerl »

There should be some Rust based generic GPU programming solution that's not tied to any specific GPU. That should be able to replace CUDA and Co. in the long run.

Re: What Is AMD ROCm?

#33

Post by my123 »

slavik81 wrote: post #30
my123 wrote: post #24

Earlier quoted context omitted.

> Isn't 4.5 also the version, that kicked Vega64 to the curb? ROCm 4.5 is the _last_ version to support the Vega10 ASIC (MI25, Vega56, Vega64). https://github.com/RadeonOpenCompute/ROCm/#amd-instinct-mi25... The next ROCm release after 4.5 is sometime in Q1 next year. So it's on planned death really soon. It is transitioning to _that_ comical AMD "enabled in the codebase but not tested and not supported" state, rotti…

The MI25 line predates when I joined AMD, so I don't know much about it. Were they for sale to the general public? I was concerned about that as well, but I don't personally know anyone who owns a gfx900 card. I'm a little unclear on what impact it will have on the community.

> Were they for sale to the general public?

The MI25 wasn't targeted at the general public, but it wasn't hard to buy one. And the customer products using that same die, Vega 56 and 64, were sold quite a bit.

What affected the community severely might be the combination of both GFX8 and gfx900 going away, leaving only MI50 (Vega20, also used in Radeon VII consumer variant) and no support (only unofficial enablement) for the later customer products. Because those went GFX10/Navi.

Re: What Is AMD ROCm?

#34

Post by jjoonathan »

blip54321 wrote: post #20

I tried ROCm. I bought a supported card (RX570/RX580 series). Within 12 months, AMD dropped support. Newer versions of ROCm didn't work with the card. Older versions didn't actually work either, since all other tooling assumed newer versions. Dependency hell. When things kinda started working in one context, where I could use old tooling (not the one I wanted to use ROCm in), it CUPy was slower than CPU, and then har…

Yep. This is the same lesson that I learned in 2013 and again in 2015 with OpenCL, both for professional apps and for programming. On the professional app side, there was a lot of "support" for OpenCL that had caveats severe enough to make it unusable. Sure, blender supports OpenCL! You just have to make a custom build and then the GPU target is slower than the CPU target. Sure, Adobe apps support OpenCL -- but any GPU render contexts are solid black boxes and forum posts indicate it has been that way for a year and nobody cared to fix it. Same thing with programming and debugging: there were slides suggesting feature parity with CUDA on a bunch of fronts that just weren't implemented or locked up the computer if you tried to actually use them. It got so bad that I walked away from my sunk costs, sold my AMD cards, ate the ebay tax, ate the nvidia tax, and bought cards that actually worked.

Now I'm in "twice bitten, once shy" mode with AMD. I hate paying the green tax as much as the next guy and I desperately want to have a second source of professional GPUs, but I'm not going to be the guinea pig. Not again. Not for the 3rd time. I want to see someone else successfully using AMD cards for common ML workflows and for blender before I even consider risking it again.

Re: What Is AMD ROCm?

#35

Post by zwaps »

VHRanger wrote: post #2

It's something that has very little uptake because it's not supported on mainline GPUs? I want to use it for compute on something like a rx 6800 and to my knowledge can't

AMD's support - not just for new or old GPUs, but in terms of all sorts of compatibility changes, usability parity and regressions is staggeringly bad.

Given that compute is important, and many people use their GPUs for compute of some sort, I simply can not understand how that part is so poorly executed on the part of AMD that, in terms of actual application, you might call it entirely absent.

I mean, this is a company that produces compute cards, which supposedly someone in the world must buy and use... but who? Why? I have never seen anyone, and for good reason. And it seems like AMD just... doesn't care?

Like, it's just not a part of their organizational strategy... Compute is on the powerpoint slides, but... no one (can) use it?

It's been going on for years now and I don't get it.

Re: What Is AMD ROCm?

#36

Post by synergy20 »

sorenjan wrote: post #28
blip54321 wrote: post #20

I tried ROCm. I bought a supported card (RX570/RX580 series). Within 12 months, AMD dropped support. Newer versions of ROCm didn't work with the card. Older versions didn't actually work either, since all other tooling assumed newer versions. Dependency hell. When things kinda started working in one context, where I could use old tooling (not the one I wanted to use ROCm in), it CUPy was slower than CPU, and then har…

Khrono's SYCL is the open alternative to CUDA. It's what Intel is using for their oneAPI, there's a CUDA backend and even a HIP backend.

Nice try, still no way to compare with CUDA, Nvidia is so much ahead and it's hopeless to catch up.

Re: What Is AMD ROCm?

#37

Post by enos_feedler »

bubblethink wrote: post #8
VHRanger wrote: post #2

It's something that has very little uptake because it's not supported on mainline GPUs? I want to use it for compute on something like a rx 6800 and to my knowledge can't

AMD is much smaller in comparison, and their main focus with ROCM is to get pytorch and tensorflow to work with enterprise GPUs. Everything else is long tail in terms of scale.

If this is the goal they are better off working with Google on MLIR since they are also focusing on hw acceleration of pytorch and tensorflow code

Re: What Is AMD ROCm?

#38

Post by ahartmetz »

to11mtm wrote: post #9
nicolaslem wrote: post #7

Earlier quoted context omitted.

On the other hand AMD has a fraction of the engineering resources Intel and Nvidia have. They need to make some choice and looking back at the last few years it seems that their choice to focus their efforts on hardware and gaming paid off.

ATI/AMD has always had finicky drivers and engineering decisions IMO. I guess on the plus side they at least have a more open driver than NVidia (AFAIK nouveau doesn't get any support from them, at least AMD tries to maintain their open source driver on some level.) And yet, every time I've tried an ATI/AMD Card, the driver experience even in windows has been pretty off-putting, and while I suppose we are finally at…

> AMD tries to maintain their open source driver on some level.

I think you somehow mistyped "AMD has open source drivers of absofuckinglutely excellent quality supporting hardware of the last ten years or so". No really, they are great. For graphics, that is.

Re: What Is AMD ROCm?

#39

Post by a_bonobo »

VHRanger wrote: post #2

It's something that has very little uptake because it's not supported on mainline GPUs? I want to use it for compute on something like a rx 6800 and to my knowledge can't

I tink the main workhorse supercomputer GPUs are not out yet.

They announced the MI200 GPU with 128GB of memory, two supercomputers (Setonix, Frontier) are suppose to include them but both will only be launched next year https://www.tomshardware.com/news/setonix-supercomputer-mi20...

Re: What Is AMD ROCm?

#40

Post by supernovae »

zmachinaz wrote: post #26

I have been using ROCm for 2y+. The investment in this infrastructure was a big mistake. The biggest burner was the need to do a clean install on each new ROCm release. Clean here means manually finding and deleting all traces from the previous ROCm version, and recompilation of apps like pytorch. Good upgrades took hours, bad ones days ... . Finally I settled to freeze the system and not touch it anymore until retir…

this is a problem with nvidia too.. i just made all my infra easy to reprovision and start clean and workloads ran as containers..
Post reply on HN