71–80 of 218 posts
Earlier quoted context omitted.
ROCm supports Vega, Vega 7nm, and CDNA just fine. The issue is that AMD has split their compute into two categories: * RDNA -- consumer cards. A new ISA with new compilers / everything. I don't think its reasonable to expect AMD's compilers to work on RDNA, when such large changes have been made to the architecture. (32-wide instead of 64-wide. 1024 registers. Etc. etc.) * CDNA -- based off of Vega's ISA. Despite bei…
That makes a lot more sense, thanks. They could do with making that a lot clearer on the project. Still handicaps them compared to Nvidia where you can just buy anything recent and expect it to work. Suspect it also means they get virtually no open source contributions from the community because nobody can run or test it on personal hardware.
This makes absolutely no sense to me, and I have a Ph.D:
"* RDNA -- consumer cards. A new ISA with new compilers / everything. I don't think its reasonable to expect AMD's compilers to work on RDNA, when such large changes have been made to the architecture. (32-wide instead of 64-wide. 1024 registers. Etc. etc.) * CDNA -- based off of Vega's ISA. Despite being "legacy ISA", its pretty modern in terms of capabilities. MI100 is competitive against the A100. CDNA is likely going to run Frontier and El Capitan supercomputers. ROCm focused on CDNA. They've had compilers emit RDNA code, but its not "official" and still buggy. But if you went for CDNA, that HIP / ROCm stuff works enough for the Oak Ridge National Labs. Yeah, CDNA is expensive ($5k for MI50 / Radeon VII, and $9k for MI100). But that's the price of full-speed scientific-oriented double-precision floating point GPUs these days.
I neither know nor care what RDNA, CDNA, A100, MI50, Radeon VII, MI100, or all the other AMD acronyms are. Yes, I could figure it out, but I want plug-and-play, stability, and backwards-compatibility. I ran into a whole different minefield with AMD. I'd need to run old ROCm, downgrade my kernel, and use a different card to drive monitors than for ROCm. It was a mess.
NVidia gave me plug-and-play. I bought a random NVidia card with the highest "compute level," and was confident everything would work. It does. I'm happy.
Intel has historically had great open source drivers, and if it give better plug-and-play and open source, I'll buy Intel next time. I'm skeptical, though. The past few year, Intel has a hard time tying their own shoelaces. I can't imagine this will be different.
As TFA rightly points out, unless something drastically changes in the next ~6mo, Intel is going to launch into the most favorable market situation we've seen in our lifetimes. Previously, the expectation is that they needed to introduce something that was competitive with the top end cards from nVidia and AMD. With basically all GPU's out of stock currently they really just need to introduce something competitive wi…
Earlier quoted context omitted.
Unfortunately that support is via ROCm, which doesn't support the last three generations (!) of AMD hardware: https://github.com/ROCm/ROCm.github.io/blob/master/hardware....
ROCm supports Vega, Vega 7nm, and CDNA just fine. The issue is that AMD has split their compute into two categories: * RDNA -- consumer cards. A new ISA with new compilers / everything. I don't think its reasonable to expect AMD's compilers to work on RDNA, when such large changes have been made to the architecture. (32-wide instead of 64-wide. 1024 registers. Etc. etc.) * CDNA -- based off of Vega's ISA. Despite bei…
yeah, but that's exactly what OP said - Vega is three generations old at this point, and that is the last consumer GPU (apart from VII which is a rebranded compute card) that ROCm supports.
On the NVIDIA side, you can run at least basic tensorflow/pytorch/etc on a consumer GPU, and that option is not available on the AMD side, you have to spend $5k to get a GPU that their software actually supports.
Not only that but on the AMD side it's a completely standalone compute card - none of the supported compute cards do graphics anymore. Whereas if you buy a 3090 at least you can game on it too.
As TFA rightly points out, unless something drastically changes in the next ~6mo, Intel is going to launch into the most favorable market situation we've seen in our lifetimes. Previously, the expectation is that they needed to introduce something that was competitive with the top end cards from nVidia and AMD. With basically all GPU's out of stock currently they really just need to introduce something competitive wi…
Intel has the manufacturing capability to really beat up Nvidia. Even if the cards don’t perform like top-tier cards they could still win bigly here. Very exciting!
They're a shared resource; however, if you're willing to pay the money, you could monopolize their resources and outproduce anybody.
1 - https://epsnews.com/2021/02/10/5-fabs-own-54-of-global-semic...
I find myself needing, for the first time ever, a high-end video card for some heavy video encoding, and when I look, they're all gone, apparently in a tug of war between gamers and crypto miners. At the exact same time, I am throwing out a box of old video cards from the mid-nineties (Trident, Diamond Stealth) and from the looks of it you can list them on eBay but they don't even sell. Now Intel is about to leap int…
That, and the oligopoly of AMD and NVidia. Their grip is so tight they dictate terms to card makers. For example; you can't build an NVidia GPU card unless you source the GDDR from NVidia. Between them the world supply of high end GDDR is monopolized.
Intel is going to deliver some badly needed competition. They don't even have to approach the top of the GPU high end; just deliver something that will play current games at 1080p at modest settings and they'll have an instant hit. Continuing the tradition of open source support Intel has had with (most) of their GPU technology is something else we can hope for.
Earlier quoted context omitted.
That makes a lot more sense, thanks. They could do with making that a lot clearer on the project. Still handicaps them compared to Nvidia where you can just buy anything recent and expect it to work. Suspect it also means they get virtually no open source contributions from the community because nobody can run or test it on personal hardware.
I bought an ATI card for deep learning. I'm a big fan of open source. Less than 12 months later, ROCm dropped support. I bought an NVidia, and I'm not looking back. This makes absolutely no sense to me, and I have a Ph.D: "* RDNA -- consumer cards. A new ISA with new compilers / everything. I don't think its reasonable to expect AMD's compilers to work on RDNA, when such large changes have been made to the architectu…
Its right there in the ROCm introduction.
https://github.com/RadeonOpenCompute/ROCm#Hardware-and-Softw...
> ROCm officially supports AMD GPUs that use following chips:
> GFX9 GPUs
> "Vega 10" chips, such as on the AMD Radeon RX Vega 64 and Radeon Instinct MI25
> "Vega 7nm" chips, such as on the Radeon Instinct MI50, Radeon Instinct MI60 or AMD Radeon VII, Radeon Pro VII
> CDNA GPUs
> MI100 chips such as on the AMD Instinct™ MI100
--------
The documentation of ROCm is pretty clear that it works on a limited range of hardware, with "unofficial" support at best on other sets of hardware.
Earlier quoted context omitted.
This is a problem for AMD especially Probably not. AMD has had their N7/N6 orders in for years. They're just budging in line with their superior firepower. Intel even bought out first dibs on TSMC 3nm out from under Apple. There's no evidence this is happening and people with TSMC experience say it's not happening. Nvidia and especially AMD took a risk not being in the fabrication business Yes, and it paid off dramat…
Do you have sources for any of your claims? Other than going fabless being a fantastic way to cut costs and management challenges, but increase longterm supply line risk, none of that is anything that I've heard. Here are sources for my claims. AMD on TSMC 3nm for Zen5. Will be squeezed by Intel and Apple- https://videocardz.com/newz/amd-3nm-zen5-apus-codenamed-stri... Intel consuming a good portion of TSMC 3nm- http…
As TFA rightly points out, unless something drastically changes in the next ~6mo, Intel is going to launch into the most favorable market situation we've seen in our lifetimes. Previously, the expectation is that they needed to introduce something that was competitive with the top end cards from nVidia and AMD. With basically all GPU's out of stock currently they really just need to introduce something competitive wi…
GPU stock is rising and prices falling. It’s too late now.