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

It’s Time to Open Up the GPU

gpuopen.com

Post reply on HN
Print view View original

21–30 of 95 posts

Re: It’s Time to Open Up the GPU

#21

Post by cs702 »

This is an AMD project motivated by the fact that NVIDIA has become the de facto standard for GPU computation in deep learning and machine learning applications.

TensorFlow, Torch, Theano, CNTK, CAFFE, and every other deep learning framework out there works out of the box with NVIDIA hardware via the CUDA stack, but not with AMD hardware, e.g., via the OpenCL stack.

Which is a shame, because the entire NVIDIA hardware + software stack is closed.

It would be great to see this new project succeed, and maybe even force NVIDIA to open source key or all components its stack down the road.

To the team at AMD running this project: if you are reading this, step one is to get built-in support into all major deep learning and machine learning frameworks.

Re: It’s Time to Open Up the GPU

#22

Post by robbies »

wtallis wrote: post #20
tracker1 wrote: post #19

Wasn't OpenCL supposed to deal with this? Obligatory XKCD[1] [1] https://m.xkcd.com/927/

OpenCL is basically a framework to make it more convenient to run shaders for non-graphics purposes. It's still a high-level cross-vendor hardware abstraction layer. This sounds like it's more about officially documenting and supporting the kinds of things people have been reverse-engineering: https://news.ycombinator.com/item?id=10605156

Yes to your OpenCL stuff, no to your reverse-engineering stuff. The post you linked is about generating your own native shaders on GCN. GPUOpen could possibly expose that, but that's a difficult problem to expose to developers considering the variety of architectures exposed at any time by the IHVs (including just inside one IHV at any time). If they did go down that path, they would open up their internal bytecode-to-native-ISA shader compiler...which they haven't done yet, and have shown no indication of doing.

Re: It’s Time to Open Up the GPU

#23

Post by robbies »

cs702 wrote: post #21

This is an AMD project motivated by the fact that NVIDIA has become the de facto standard for GPU computation in deep learning and machine learning applications. TensorFlow, Torch, Theano, CNTK, CAFFE, and every other deep learning framework out there works out of the box with NVIDIA hardware via the CUDA stack, but not with AMD hardware, e.g., via the OpenCL stack. Which is a shame, because the entire NVIDIA hardwar…

I can only speculate on this aspect (from inside the industry), but it's been plain to see that AMD has been hemorrhaging engineers (though they still have plenty of talent). This relates to your request about getting built-in support: I'm not sure AMD has the manpower to do that work, and they might be hoping the OSS community helps them with that...

Re: It’s Time to Open Up the GPU

#24

Post by gravypod »

cs702 wrote: post #21

This is an AMD project motivated by the fact that NVIDIA has become the de facto standard for GPU computation in deep learning and machine learning applications. TensorFlow, Torch, Theano, CNTK, CAFFE, and every other deep learning framework out there works out of the box with NVIDIA hardware via the CUDA stack, but not with AMD hardware, e.g., via the OpenCL stack. Which is a shame, because the entire NVIDIA hardwar…

That is and will continue to be one of the biggest concerns for the GPU market for some time now.

I assume that after a few years, factories will be buying computers, cameras, and GPUs instead of people for inspection.

This will be a large market, and that's what NVIDIA understood when they started making ties with different companies. AMD has a lot of catch up to do, but in my opinion they can do it.

Re: It’s Time to Open Up the GPU

#25

Post by appleflaxen »

cs702 wrote: post #21

This is an AMD project motivated by the fact that NVIDIA has become the de facto standard for GPU computation in deep learning and machine learning applications. TensorFlow, Torch, Theano, CNTK, CAFFE, and every other deep learning framework out there works out of the box with NVIDIA hardware via the CUDA stack, but not with AMD hardware, e.g., via the OpenCL stack. Which is a shame, because the entire NVIDIA hardwar…

I'm sad to say, but it's true. AMD has been claiming "this is the year we really open up" since at least 2007[1]. Well, I'm happy to hear them recommit, but I don't believe them anymore.

Now that they are losing ground in the marketplace on the GPU side with deep learning, and their CPUs can't compete with intel on the performance side, their claims ring hollow.

Imagine where we would be if they open sourced their drivers in 2007 like they said they would do.

1. http://arstechnica.com/gadgets/2007/05/amd-launches-the-hd-2...

Re: It’s Time to Open Up the GPU

#26

Post by poopchute »

Sleaker wrote: post #12

Doesn't give off the impression of being super open when you have samples say they are only for Radeon cards. This seems almost like a marketing strategy/guise from AMD, and if you go look at other websites for this article it literally says AMD launches GPUopen.com. AMD is only mentioned once in the article.. shady in my opinion and not very 'open' either. When one company continually calls out for 'openness' in a m…

Did you even read the post? The whole thing talks about how this is AMD's new version of mantle which is specifically for gcn cards. It even says who its written by at the bottom, the senior manager of engineering at AMD.

Re: It’s Time to Open Up the GPU

#27

Post by robbies »

ChuckMcM wrote: post #11

First, nullify all the patents ... Seriously, back around the turn of the century I was pretty heavily into graphics and rendering and was frustrated by all the constraints on getting access to documentation. I finally figured out a way to pin down a vendor with all the necessary NDAs and all the necessary agreements so that I could actually get 100% access to the inner workings of the GPU. As long as I read the docu…

I'm actually amazed that you were able to get an IHV to give you that access.

Interestingly, most console developers currently have this sort of access. If they are just targeting consoles (or better, a specific platform), they're able to get a lot done. The tricky part is if they have to target PC and consoles. They have the access they want on consoles, but can't replicate on PC.

Actually, on top of that, replicating that task on PC would mean giving developers a nightmarish mix of varying, quirky architectures to target. Realistically, what most _engine_ devs want on PC are more current and lower level access APIs (DX12 and Vulkan) and maybe even an OSS shader compiler. The compiler wouldn't be even for them to poke at. I think they'd hope that the OSS community might do a better job than the IHVs at maintaining the compiler.

Re: It’s Time to Open Up the GPU

#28

Post by forrestthewoods »

cs702 wrote: post #21

This is an AMD project motivated by the fact that NVIDIA has become the de facto standard for GPU computation in deep learning and machine learning applications. TensorFlow, Torch, Theano, CNTK, CAFFE, and every other deep learning framework out there works out of the box with NVIDIA hardware via the CUDA stack, but not with AMD hardware, e.g., via the OpenCL stack. Which is a shame, because the entire NVIDIA hardwar…

Step one is to produce competitive hardware. If AMD's hardware sucks then it doesn't matter if it's supported by frameworks or not.

Re: It’s Time to Open Up the GPU

#29

Post by pjmlp »

cs702 wrote: post #21

This is an AMD project motivated by the fact that NVIDIA has become the de facto standard for GPU computation in deep learning and machine learning applications. TensorFlow, Torch, Theano, CNTK, CAFFE, and every other deep learning framework out there works out of the box with NVIDIA hardware via the CUDA stack, but not with AMD hardware, e.g., via the OpenCL stack. Which is a shame, because the entire NVIDIA hardwar…

Maybe if Apple and Khronos cared to define OpenCL native support for C++ and Fortran since day 1, this would have turned out to be different.

Re: It’s Time to Open Up the GPU

#30

Post by pcwalton »

FussyZeus wrote: post #5
badlucklottery wrote: post #4

Earlier quoted context omitted.

Smooth for me on Win7/FF44b+IE11. Chrome 48 doesn't appear to like it much though.

Safari up-to-date on a MacBook Pro, so much chug. Must be a webkit thing.

Works fine for me on Firefox too. It likely is a WebKit issue.
Post reply on HN