Earlier quoted context omitted.
Move away from everything. Retvrn to monke.
No, further, return to sea. Exist as peaceful algae.
Why oh why didn't I take the blue pill?
Intel CEO: 'The entire industry is motivated to eliminate the CUDA market'
31–40 of 380 posts
Re: Intel CEO: 'The entire industry is motivated to eliminate the CUDA market'
#32What problem with cuda are they going to solve?
Re: Intel CEO: 'The entire industry is motivated to eliminate the CUDA market'
#33As another commenter said, it's CUDA. Intel and AMD and whoever can turn out chips reasonably fast, but nobody gets that it's the software and ecosystem. You have to out-compete the ecosystem. You can pick up a used Mi100 that performs almost like an A100 for 5x less money on eBay for example. Why is it 5x less? Because the software incompatibilities mean you'll spend a ton of time getting it to work compared to an Nvidia GPU.
Google is barely limping along with it's XLA interface to pytorch providing researchers a decent compatibility path. Same with Intel.
Any company in this space should basically setup a giant test suite of IDK, every model on hugging face and just start brute force fixing the issues. Then maybe they can sell some chips!
Intel is basically doing the same shit they always do here, announcing some open initiative and then doing literally the bare minimum to support it. 99% chance openvino goes nowhere. OpenAIs Triton already seems more popular, at least I've heard it referenced a lot more than openvino.
Re: Intel CEO: 'The entire industry is motivated to eliminate the CUDA market'
#34People don't seem particularly motivated to move away from CUDA to me. I've been poking around various models and tooling over the last couple months, and they pretty much all have something like
device = "cuda" if torch.cuda.is_available() else "cpu"
Re: Intel CEO: 'The entire industry is motivated to eliminate the CUDA market'
#35Intel and AMD have had years to provide similar capabilities on top of OpenCL. Maybe they should look into their own failures first.
Apple as well. Everyone's failure to commit in this situation enabled a highly-integrated competitor to clean up shop. It's funny how much clearer OpenCL's value prop is in hindsight...
Apple created OpenCL, and after disagreements with Khronos on how to move OpenCL forward, they stopped caring.
Apple also doesn't care about HPC, and has their own stuff on top of Metal Compute Shaders, just like Microsoft has DirectML.
Re: Intel CEO: 'The entire industry is motivated to eliminate the CUDA market'
#36Earlier quoted context omitted.
While we are at it, why not both? Also, can we move away from Windows and Office as well? And SharePoint, please? And add good search capabilities to Confluence.
Are you me? This list enumerates many specific pains I've had over the years. Adobe Acrobat might be next on this list.
You're just one of the borg that has existed and used computers and all felt the exact same thing.
Re: Intel CEO: 'The entire industry is motivated to eliminate the CUDA market'
#37Intel and AMD have had years to provide similar capabilities on top of OpenCL. Maybe they should look into their own failures first.
AMD definitely has and is doubling down on ROCm.
Re: Intel CEO: 'The entire industry is motivated to eliminate the CUDA market'
#38Re: Intel CEO: 'The entire industry is motivated to eliminate the CUDA market'
#39Sure, provide a 100% CUDA-compatible chip, then do the microsoft-style EEE.
Wouldn't a 100% Cuda-compatible toolchain suffice? Then you could write your software for Cuda and the software environment would run it on whatever is on your computer.
CUDA the language itself isn't necessarily the biggest obstacle for porting code, it's the usage of low level hardware features. You'd have to make a chip that's pretty dang close to Nvidia's before most hand-written CUDA code could run on it performantly.
Re: Intel CEO: 'The entire industry is motivated to eliminate the CUDA market'
#40Earlier quoted context omitted.
SYCL is a better analog to Cuda than OpenCL, and Intel have their own implementation of that. Don't really see anyone writing anything in SYCL though, and when I looked into trying it out it was a bit of a mess with different implementations, each supporting their own subset of OSs and hardware. https://www.intel.com/content/www/us/en/developer/tools/onea...
I am unfamiliar with the implementation, but why would it be difficult to implement a Cuda-compatible software layer on top of other platforms? This would be the first step. Then, if we want to move away from Cuda into hardware that's as ubiquitous and performant as Nvidia's (or better), someone would need to write an abstraction layer that's more convenient to use than Cuda. I did play a little bit with Cuda and Ope…
Because that is an herculean task, given the hardware semantics, the amount of languages that target PTX, the graphical debugging tools that expose every little detail of the cards like debugging on the CPU, and the libraries ecosystem.
Any Cuda-compatible software layer only has two options, be a second class CUDA implementation by being compatible with a subset like AMD ROCm and HIP effforts, or be compatible with everything always playing catchup.
The only way is to use middleware that just like in 3D APIs, abstract the actual compute API being used, as man language bindings are doing nowadays.