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

Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

blogs.nvidia.com

Post reply on HN
Print view View original

291–300 of 347 posts

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#291

Post by fluffything »

adev_ wrote: post #178
fluffything wrote: post #165

Earlier quoted context omitted.

> Correction: Google developped an internal CUDA implementation for their own need based on LLVM that Nvidia barely supported it for their own need afterwards. This is widely inaccurate. While Google did developed a PTX backend for LLVM, the student that worked on that as part of a GSOC got later hired by NVIDIA, and ended up contributing the current NVPTX backend that clang uses today. The PTX backend that Google co…

> While Google did developed a PTX backend for LLVM, the student that worked on that as part of a GSOC got later hired by NVIDIA, and ended up contributing the current NVPTX backend that clang uses today. You more or less reformalized what I said. It might become used one day behind a proprietary blob, rebranded blob of NVidia, but fact is that today, close to nobody use it for production in the wild and it is not ev…

> CUDA fortran, yes you said it, CUDA fortran. The rest is OpenACC.

You can also mix C, OpenACC, and CUDA C, and compile to x86-64. So I'm really not sure about what point you are trying to make here.

You were claiming that OpenACC and CUDA only runs on nvidia's hardware, yet I suppose you now agree that this isn't true I guess.

I do agree that PGI is still nvidia owned, but there are other compilers that do what PGI does.

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#292

Post by fluffything »

mambru wrote: post #228
fluffything wrote: post #146

Earlier quoted context omitted.

> Correction: Nobody will be able to use the AMD hardware (outside of computer graphics) because everybody has been locked-in with CUDA on Nvidia. NVIDA open-sourced their CUDA implementation to the LLVM project 5 years ago, which is why clang can compile CUDA today, and why Intel and PGI have clang forks compiling CUDA to multi-threaded and vectorized x86-64 using OpenMP. That you can't compile CUDA to AMD GPUs isn'…

>Intel and PGI have clang forks compiling CUDA to multi-threaded and vectorized x86-64 using OpenMP. Where are these forks?

For PGI, all pgi compilers can do this, just pick x86-64 as the target. There are also other forks online (just search for LLVM, CUDA, x86 as keywords), some university groups have their forks on github, where they compile CUDA to x.

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#293

Post by ksec »

bhouston wrote: post #187
dijit wrote: post #18

Earlier quoted context omitted.

I don't like that people downvoted you for asking a question. If someone thinks the question is stupid or not doesn't mean that a downvote is warranted. (nor an upvote, answer the question and move on.) To answer though; it's just a coincidence, as you might already know Nvidia uses famous scientists (especially in the field of electricity) as the names of their microarchitectures. * Volta (Alessandro Volta, inventor…

And even before that there was: * GeForce (for Andrea Geforce, the first to use the color electric green) * Riva (for Jose Riva, the discoverer that you can use TNT to generate electricity.)

Wow, I didn't know that. Wasn't it done in contest by Gamers?

So what does "Radeon" mean?

Edit: Yes according to Wiki. So we may never know whether Before was actually dedicated to the person. I remember vaguely reading PC Gamer at the time that was not the case at all.

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#294

Post by pixelpoet »

GauntletWizard wrote: post #255
m463 wrote: post #248

Earlier quoted context omitted.

Remember the long view - patents cause people to hurry to publish and share their ideas publically. Why shouldn't they be granted liberally. In a few years, the temporary monopoly falls away and the benefit passes to everyone. I think they should work to make them even cheaper and easier to file.

20 years is a long time. For some fields, it is perfectly reasonable, but 20 year patents on many recent CS inventions would have significantly bottlenecked development of the industry - Look at how much mess was created by the JPEG patents, for example, and similar problems have existed for every other not-explicitly-libre A/V codec.

While we're on the subject of patents and Nvidia, their patent on using quasi Monte Carlo in rendering is allowing them to hold basically the whole path tracing world hostage, e.g. possibly forcing people to use CUDA who might otherwise have used OpenCL.

They didn't even invent the numerical methods themselves (pure mathematics from other countries from long ago), they were just first to file for a particular application.

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#295

Post by slavik81 »

adev_ wrote: post #190
amelius wrote: post #164

Earlier quoted context omitted.

> Nobody will be able to use the AMD hardware (outside of computer graphics) because everybody has been locked-in with CUDA on Nvidia. But numpy can be ported. So can pytorch. I don't think the lock-in is that big of an issue. GPUs do only simple things, but do them fast.

> But numpy can be ported. So can pytorch. Letting AMD or Intel port themselves everything that has been developed in CUDA like it was done for Pytorch is not substainable and will always lag behind. It can only help to create a monopoly on the long term.

As Hip continues to implement more of CUDA, I think we'll see more developers doing it themselves when the barrier to porting is smaller. AMD has a lot of work to do, and I don't know whether they'll succeed or not, but IMO they have the right strategy.

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#296

Post by m4rtink »

saiya-jin wrote: post #287
m4rtink wrote: post #272

Earlier quoted context omitted.

Tesla was also the name of a Czechoslovak elektronics company, known among other thing for their electron microscopes. Even though this Tesla is long gone, it has a lasting legacy here in Brno due to many electron microscope manufacturers (Delong, FEI, Thermo Fisher, etc.) being present and often libking their origin or many employees to the old Tesla company.

Ha, my cousin works in one of those electron microscope companies. Thought it was some one-off company saving some costs maybe due to good graduates coming out of Brno technical university, good to know there is more to it!

One of the old Tesla company buildings is now the Brno Museum of Technology and the have (among many other) an exposition about the history of electron microcopy in Brno. One of the mashines that is part of the exposition (a huge experimental electron beam litography machine) is pretty much bolted in place in the same spot the old Tesla company built it all those years ago. There are even some "legends" about how they built it so well, that it is still holding vacuum inside today.

Another legend says that they once could not get a new model of electron microscope working before showing it on an international exposition. The image was always blurry. They ran out of time, so they just took it to the exposition to fix it there. But it worked flawlessly there! Turns out this was due to the tram troley lines running next to the company building interferring with the electron optics. :)

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#297

Post by ksec »

Polylactic_acid wrote: post #236
adev_ wrote: post #121

Earlier quoted context omitted.

> AMD could release a 2x powerful GPGPU tomorrow for half the price and most current NVIDIA users wouldn't care because what good is that if you can't program it?. Correction: Nobody will be able to use the AMD hardware (outside of computer graphics) because everybody has been locked-in with CUDA on Nvidia. They can not even change even if they want to: it is pure madness to reprogram an entire GPGPU software stack e…

Amd doesn't even have software for GPGPU on some of their cards. I have an rx5700xt and I cant use it for anything but gaming because ROCm doesn't support navi cards, a whole year after its release.

I think at this point AMD wants anything Compute to concentrate on CDNA, and graphics remain on RDNA.

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#298

Post by ksec »

amelius wrote: post #164
adev_ wrote: post #121

Earlier quoted context omitted.

> AMD could release a 2x powerful GPGPU tomorrow for half the price and most current NVIDIA users wouldn't care because what good is that if you can't program it?. Correction: Nobody will be able to use the AMD hardware (outside of computer graphics) because everybody has been locked-in with CUDA on Nvidia. They can not even change even if they want to: it is pure madness to reprogram an entire GPGPU software stack e…

> Nobody will be able to use the AMD hardware (outside of computer graphics) because everybody has been locked-in with CUDA on Nvidia. But numpy can be ported. So can pytorch. I don't think the lock-in is that big of an issue. GPUs do only simple things, but do them fast.

I have been stating similar thing on GPU for a very long time.

The GPU hardware is ( comparatively ) simple.

It is the software that sets GPU vendors apart. For Gaming, that is Drivers. For Compute that is CUDA.

On a relative scale, getting a decent GPU design may have a difficulty of 1, getting a decent Drivers to work well on all existing software is 10, getting the whole ecosystem system around your Drivers / CUDA + Hardware is likely in the range of 50 to 100.

As far as I can tell, under Jensen's leadership, the chance of AMD or even Intel to shake up Nvidia's grasp in this domain is partially zero in the foreseeable future.

That is speaking as an AMD shareholder and really wants AMD to compete.

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#299

Post by w0utert »

fluffything wrote: post #94
wegs wrote: post #23

Earlier quoted context omitted.

Yeah. That's been my general problem with adopting NVidia for anything. They make good hardware, but there's a lot of lock-in, and not a lot of transparency. That introduces business risk. I'm not in a position where I need GPGPU, but if there wasn't that risk, and generally there were mature, open standards, I'd definitely use it. The major breakpoint would be when libraries like Numpy do it natively, and better yet…

> They make good hardware, but there's a lot of lock-in, and not a lot of transparency. This sounds like you'd like NVIDIA to open-source all their software. I see this type of request a lot, but I don't see it happening. NVIDIA's main competitive advantage over AMD and Intel is its software stack. AMD could release a 2x powerful GPGPU tomorrow for half the price and most current NVIDIA users wouldn't care because wh…

>> NVIDIA's main competitive advantage over AMD and Intel is its software stack. AMD could release a 2x powerful GPGPU tomorrow for half the price and most current NVIDIA users wouldn't care because what good is that if you can't program it?

I always wonder why it is so hard for AMD to develop a true competitor to CUDA, but for AMD hardware? Not try to solve GPGPU programming through open standards like OpenCL, just copy the concept of CUDA wholesale. They could still build it on top of LLVM etc and release the whole thing as open-source, but have the freedom to not have to deal with design-by-committee frameworks like OpenCL, so they can ensure focus on GPU programming and nothing else, and only on those platforms where the majority of the demand is. There is not much wrong with OpenCL, it's just not nearly as good/capable/easy-to-use as CUDA if all you are interested in is GPGPU programming.

AMD is a big company with a lot of revenue, especially recently, so why would it be so hard to have a team working full-time on creating a direct CUDA knock-off ASAP?

Re: Nvidia CEO Introduces Nvidia Ampere Architecture, Nvidia A100 GPU

#300

Post by magicalhippo »

zelon88 wrote: post #168
wmf wrote: post #159

Earlier quoted context omitted.

That's been around since MMX and AltiVec. It took a while for GPUs to adopt subword SIMD though.

SIMD works great for doing the same thing to multiple pieces of data, but it doesn't do the scaling up that I described. I'm no chip engineer, so maybe what I'm envisioning isn't possible. In essence, instead of making 4x 64-bit cores you make 128x 2-bit cores and then some architecture on the die to select groups of cores to build a processor of the required size, execute some instructions with that processor, and t…

What you're describing is basically a modern FPGA[1]. You can wire it up as you want at runtime, and they can contain specialized hardware like hardware multipliers and fast local memory to accelerate certain workloads.

[1]: https://en.wikipedia.org/wiki/Field-programmable_gate_array

Post reply on HN