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

LibreCUDA – Launch CUDA code on Nvidia GPUs without the proprietary runtime

github.com

Post reply on HN
Print view View original

101–110 of 162 posts

Re: LibreCUDA – Launch CUDA code on Nvidia GPUs without the proprietary runtime

#101

Post by smokel »

segmondy wrote: post #72
daft_pink wrote: post #39

I think the point of open cuda is to run it on non NVIDIA gpus. Once you have to buy NVIDIA gpus what’s the point. If we had true you competition I think it would be far easier to buy devices with more vram and thus we might be able to run llama 405b someday locally. Once you already bought the NVIDIA cards what’s the point

Some of us are running llama 405B locally already. All my GPUs are ancient Nvidai GPUs. IMO, the point of an open cuda is to force Nvidia to stop squeezing us. You get more performance for the buck for AMD. If I could run cuda on AMD, I would have bought new AMD gpus instead. Have enough people do that and Nvidia might take note and stop squeezing us for cash.

> the point of an open cuda is to force Nvidia to stop squeezing us

Nobody is forcing you to buy GPUs.

Your logic is flawed in the sense that enough people could also simply write alternatives to Torch, which, by the way, is already open source.

Re: LibreCUDA – Launch CUDA code on Nvidia GPUs without the proprietary runtime

#102

Post by funcDropShadow »

galkk wrote: post #93
IgorPartola wrote: post #82

Earlier quoted context omitted.

Hear hear. Yes practically if you need to run a workload on a closed source system or if that’s your only option to get the performance then you have to do what you have to do. But in the long run open source wins because once an open source alternative exists it is just the better option. As a bonus, with open source platforms you are much less subject to whims of company licensing. If tomorrow Nvidia decided to cha…

> But in the long run open source wins because once an open source alternative exists it is just the better option. This is just wishful thinking. Anything close to real professional use, not related to IT, and closed source is king: office work, CAD, video editing, music production, and those domains immediately came to mind. Nowhere there open source can seriously challenge commercial, closed sourced competitors. Y…

Counter example: blender. It may not be winning in video editing, but it has serious market share in 3d rendering. Different players are investing money in it and extend it with their own stuff.

Re: LibreCUDA – Launch CUDA code on Nvidia GPUs without the proprietary runtime

#103

Post by funcDropShadow »

smokel wrote: post #101
segmondy wrote: post #72

Earlier quoted context omitted.

Some of us are running llama 405B locally already. All my GPUs are ancient Nvidai GPUs. IMO, the point of an open cuda is to force Nvidia to stop squeezing us. You get more performance for the buck for AMD. If I could run cuda on AMD, I would have bought new AMD gpus instead. Have enough people do that and Nvidia might take note and stop squeezing us for cash.

> the point of an open cuda is to force Nvidia to stop squeezing us Nobody is forcing you to buy GPUs. Your logic is flawed in the sense that enough people could also simply write alternatives to Torch, which, by the way, is already open source.

Nobody is forcing you to buy a computer.

Nobody is forcing you to live under a roof.

Nobody is forcing you to eat.

Re: LibreCUDA – Launch CUDA code on Nvidia GPUs without the proprietary runtime

#104

Post by samstave »

sharpshadow wrote: post #98
samstave wrote: post #44

Earlier quoted context omitted.

If you use all the GPTs enough - you'll see them clear as day... And by saying "Tell us how you really feel" reveals, you may not have thought of The Implications of the current state of AI. (I can give you a concrete example of the WEF guardrails: I have a LBB of some high profile names that are all related around a specific person, then I wanted to see how they were related to one another from a publicly available…

I know from this book[1] how deep and far reaching even just publicly available data can get you. Furthermore with financials even so far that one can predict upcoming events from analysing the money flow. Very powerful people are very good at hiding. It’s no surprise that they want themselves excluded from various searches and are successfully able to do so. Would be interesting to know if the data is excluded alrea…

Im fully convinced now that both openai and anthropic have agi. mayhaps not in whatever a 'conventional definition is' -- in fact, I think its far more insidious: I think that its a computing logic/reasoning system which, when fully connected to the internet and whatever other networks they can give it access to - it has Omniscient Capability.

We've known of echelon being a fully capable phone surveillance system since the 70s.

We knew of a lot of capability the agencies etc have had over the decades.

I wonder when Sam Altman may visit Antarctica?

Re: LibreCUDA – Launch CUDA code on Nvidia GPUs without the proprietary runtime

#105

Post by d4mi3n »

galkk wrote: post #93
IgorPartola wrote: post #82

Earlier quoted context omitted.

Hear hear. Yes practically if you need to run a workload on a closed source system or if that’s your only option to get the performance then you have to do what you have to do. But in the long run open source wins because once an open source alternative exists it is just the better option. As a bonus, with open source platforms you are much less subject to whims of company licensing. If tomorrow Nvidia decided to cha…

> But in the long run open source wins because once an open source alternative exists it is just the better option. This is just wishful thinking. Anything close to real professional use, not related to IT, and closed source is king: office work, CAD, video editing, music production, and those domains immediately came to mind. Nowhere there open source can seriously challenge commercial, closed sourced competitors. Y…

What if Linux itself? The plethora of open programming languages? Tools like OpenSSH?

Commercial, closed source products generally benefit from a monopoly within a specific problem domain or some kind of regulatory capture. I don’t think that means an open source alternative isn’t desirable or viable, just that competing in those contexts is much more difficult without some serious investment—be it political, monetary, or through many volunteered hours of work.

Another comment mentioned Blender which is a great example of a viable competitor in a specific problem domain. There are others if you look at things like PCB circuit design, audio production/editing, and a surprising amount of fantastic computer emulators.

Re: LibreCUDA – Launch CUDA code on Nvidia GPUs without the proprietary runtime

#106

Post by georgehotz »

ZoomerCretin wrote: post #100
georgehotz wrote: post #94

Cool to see one of these in C, particularly if it can be binary compatible. Why not s/libreCuInit/cuInit? If you are interested in open source runtimes, tinygrad has them in Python for both AMD and NVIDIA, speaking directly to the kernel through ioctls and poking the command queues. https://github.com/tinygrad/tinygrad/blob/master/tinygrad/ru... https://github.com/tinygrad/tinygrad/blob/master/tinygrad/ru...

Incredible! Any plans to support SASS instructions for Nvidia GPUs, or only PTX?

We'll get there as we push deeper into assemblies. RDNA3 probably first, since it's documented and a bit simpler.

Re: LibreCUDA – Launch CUDA code on Nvidia GPUs without the proprietary runtime

#108

Post by smokel »

funcDropShadow wrote: post #103
smokel wrote: post #101

Earlier quoted context omitted.

> the point of an open cuda is to force Nvidia to stop squeezing us Nobody is forcing you to buy GPUs. Your logic is flawed in the sense that enough people could also simply write alternatives to Torch, which, by the way, is already open source.

Nobody is forcing you to buy a computer. Nobody is forcing you to live under a roof. Nobody is forcing you to eat.

Sorry for the harsh comment.

I just found it highly unlikely that Nvidia would change its ways due to this, and I don't really see how we're being "squeezed". Nvidia are delivering amazing products (as are AMD), and it is not going to be any cheaper this way.

Building this kind of hardware is not something a hacker can do over the weekend.

Re: LibreCUDA – Launch CUDA code on Nvidia GPUs without the proprietary runtime

#109

Post by raxxorraxor »

galkk wrote: post #93
IgorPartola wrote: post #82

Earlier quoted context omitted.

Hear hear. Yes practically if you need to run a workload on a closed source system or if that’s your only option to get the performance then you have to do what you have to do. But in the long run open source wins because once an open source alternative exists it is just the better option. As a bonus, with open source platforms you are much less subject to whims of company licensing. If tomorrow Nvidia decided to cha…

> But in the long run open source wins because once an open source alternative exists it is just the better option. This is just wishful thinking. Anything close to real professional use, not related to IT, and closed source is king: office work, CAD, video editing, music production, and those domains immediately came to mind. Nowhere there open source can seriously challenge commercial, closed sourced competitors. Y…

It is completely ok to use commercial software in a commercial environment. It isn't and shouldn't be the goal of open source to provide the best consumer product.

In the grand scheme of things I believe open source at least provides serious competition and that commercial software has its own work to do.

Also, a lot of not all professional work uses open source components. Research is a field where it shines and there it matters a lot.

Adobe has to work for its money as well as its competitors get more powerful by the day. And everyone hates their creative cloud.

Re: LibreCUDA – Launch CUDA code on Nvidia GPUs without the proprietary runtime

#110

Post by lmpdev »

daft_pink wrote: post #39

I think the point of open cuda is to run it on non NVIDIA gpus. Once you have to buy NVIDIA gpus what’s the point. If we had true you competition I think it would be far easier to buy devices with more vram and thus we might be able to run llama 405b someday locally. Once you already bought the NVIDIA cards what’s the point

The point might not necessarily be for consumers

Linus wasn’t writing Linux for consumers (arguably the Linux kernel team still isn’t), he needed a Unix-like kernel on a platform which didn’t support it

Nvidia is placed with CUDA in a similar way to how Bell was with Unix in the late 1980s. I’m not sure if a legal “CUDA Wars” is possible in the way the Unix Wars was, but something needs to give

Nvidia has a monopoly and many organisations and projects will come about to rectify it, I think this is one example

The most interesting thing to see moving forward is where the most just place is to draw the line for Nvidia they deserve remuneration for CUDA, but the question is how much? The axe of the Leviathan (US government) is slowly swinging towards them, and I expect Nvidia to pre-emptively open up CUDA just enough to keep them (and most of us) happy

After a certain point for a technology so low in the “stack” of the global economy, more powerful actors than Nvidia will have to step in and clear the IP bottleneck

Tech giants are powerful and influence people more than the government, but I think people forget how powerful the government can be when push comes to shove over such an important piece of technology

—————

PS my comparison of CUDA to Unix isn’t perfect, mostly as Nvidia has a hardware monopoly as it stands, but as they don’t fab it themselves it’s just a design/information at the end of the day. There’s nothing physically preventing other companies producing CUDA hardware, just obvious legal and business obstacles

Perhaps a better comparison would be Texas Instruments trying to monopolise integrated circuits (they never tried). But if Fairchild Semiconductors hadn’t’ve independently discovered ICs, we might have seen a much slower logistic curve than we have had with Moore’s law (assuming competition is proportional to innovation)

Post reply on HN