5 ms·

If not CUDA, then what can you use with those?

by 29athrowaway 3y ago

DeathArrow 3y ago
Games? Rendering? Transcoding video?
- sylware 3y ago
  Video is getting a lot of direct ASIC blocks (look at the latest VPE block in AMD GPUs). I guess those chips are for the movies/video industry, online or not. Because for the "consumer", the CPU is already very efficient, and I don't think we would save interesting about of battery usage in a "real usage" perspective. I may be very wrong, but I don't watch hours and hours in a row of ultra high quality videos on a small screen, that off the AC plug, the battery is unusable in a matter of a few years anyway... it not less.
  - imtringued 3y ago
    Encoding video in real time is expensive. You can run your blog on a 5€/month machine. You'll most likely need ten times that money for a single real time stream using nothing but software decoding, rendering and encoding. If you think, "hey I'm going to use a GPU for this" then congratulations, you've increased your cloud bill by a factor of 20.
    - sylware 3y ago
      Lastest AMD GPU does realtime encoding with AV1. It has directly an ASIC block for that. Just mmap the video engine command circular buffer, get some dma buffers, get a interrupt circular buffer, and you are good to go (I think the dma-buffer sync framwork between drivers and user space is still a WIP, and I dunno how the interrupt circular buffer is handled).
anotherhue 3y ago
CUDA is at the API level (ish, I know). There's plenty of room for new silicon that can expose different APIs. Compare to switching from x86 to ARM.
- ssijak 3y ago
  why no CUDA rosetta?
  - JonChesterfield 3y ago
    Nvidia could build that but doesn't want it to exist. Outside of nvidia you'd have to reverse engineer their machine code which would be a massive undertaking. The ISA ISA is published by AMD and Intel, you could build tooling from the docs alone if you wish.
  - c0n5pir4cy 3y ago
    AMD has released HIP and a tool called HIPIFY which kind of behaves like this but at the source level¹. Rather than try and just translate CUDA to work on AMD compute they are more focused on higher level tooling. Currently they seem to have a particular focus on AI frameworks and tools like PyTorch/Tensorflow/ONNX. They have sponsored and helped with a lot of PyTorch development for example, so PyTorch support for AMD is much better than it was this time last year². ¹(https://github.com/ROCm/HIP https://github.com/ROCm/HIP) ²(https://pytorch.org/blog/experience-power-pytorch-2.0/ https://pytorch.org/blog/experience-power-pytorch-2.0/)
doikor 3y ago
Simulating nuclear explosions, effects of decay on nuclear weapon stockpiles, etc (this is literally what the El Capitan supercomputer mentioned in the article is for)
- neverrroot 3y ago
  Wrong reply, please ignore.
  - hutzlibu 3y ago
    You can also delete your wrong comments .. (within 2 h I think)
    - neverrroot 3y ago
      I couldn’t see anything about deleting, only edit. Will check out the web version on a desktop next time, maybe it’s available there.
      - hutzlibu 3y ago
        Not sure, if you have to have more karma for it (would not make sense to me), but for me the delete button is right next to the edit button.
    - mkl 3y ago
      Within 2h, and if no one has replied.
    - jacquesm 3y ago
      Until you commented...
neverrroot 3y ago
Everything is possible, especially what they have in mind, where they will do a custom implementation. CUDA is important for the existing ecosystem, but that doesn’t make it the only show in town.
bayindirh 3y ago
HPC admin here. ROCm, which is AMD's equivalent of CUDA. The thing is you don't have to directly interface with CUDA or ROCm. Once the framework you want to use supports these, you're done. AMD is consistently getting used on the TOP500 machines, and this gives them insane amounts of money to improve ROCm. CUDA has an ecosystem and hardware moat, but not because they're vastly superior, but because AMD both prioritized processors first, and NVIDIA played dirty (OpenCL support and performance shenanigans, anyone?). This moat is bound to be eaten away by both Intel and AMD, and compute will be commoditized. NVIDIA foresaw this and bought Mellanox and wanted ARM to be a complete black box, but it didn't work out. Ethernet consortium woke up and got agitated by the fact that only ultra-low latency fabric provider is not independent anymore, so they're started to build alternatives to Infiniband ecosystem. Interesting times are ahead. Also there's OneAPI, but I'm not very knowledgeable about it. It's a superAPI which can target many compute platforms, like OpenCL, but takes a different approach. It compiles to platform native artifacts (CUDA/ROCm/Intel/x86/custom, etc.) IIRC.
- justinclift 3y ago
  > they're started to build alternatives to Infiniband ecosystem. Cool, that sounds interesting. Anything you can point at? :)
  - timschmidt 3y ago
    UltraEthernet
    - justinclift 3y ago
      Thanks. :)
  - viewtransform 3y ago
    Discussion on ultrafast ethernet and partners at AMD's December presentation. https://youtu.be/tfSZqjxsr0M?t=5198 https://youtu.be/tfSZqjxsr0M?t=5198
- formerly_proven 3y ago

If not CUDA, then what can you use with those?

DeathArrow 3y ago

Games? Rendering? Transcoding video?

sylware 3y ago
Video is getting a lot of direct ASIC blocks (look at the latest VPE block in AMD GPUs). I guess those chips are for the movies/video industry, online or not. Because for the "consumer", the CPU is already very efficient, and I don't think we would save interesting about of battery usage in a "real usage" perspective. I may be very wrong, but I don't watch hours and hours in a row of ultra high quality videos on a small screen, that off the AC plug, the battery is unusable in a matter of a few years anyway... it not less.
- imtringued 3y ago
  Encoding video in real time is expensive. You can run your blog on a 5€/month machine. You'll most likely need ten times that money for a single real time stream using nothing but software decoding, rendering and encoding. If you think, "hey I'm going to use a GPU for this" then congratulations, you've increased your cloud bill by a factor of 20.
  - sylware 3y ago
    Lastest AMD GPU does realtime encoding with AV1. It has directly an ASIC block for that. Just mmap the video engine command circular buffer, get some dma buffers, get a interrupt circular buffer, and you are good to go (I think the dma-buffer sync framwork between drivers and user space is still a WIP, and I dunno how the interrupt circular buffer is handled).

anotherhue 3y ago

CUDA is at the API level (ish, I know). There's plenty of room for new silicon that can expose different APIs. Compare to switching from x86 to ARM.

ssijak 3y ago
why no CUDA rosetta?
- JonChesterfield 3y ago
  Nvidia could build that but doesn't want it to exist. Outside of nvidia you'd have to reverse engineer their machine code which would be a massive undertaking. The ISA ISA is published by AMD and Intel, you could build tooling from the docs alone if you wish.
- c0n5pir4cy 3y ago
  AMD has released HIP and a tool called HIPIFY which kind of behaves like this but at the source level¹. Rather than try and just translate CUDA to work on AMD compute they are more focused on higher level tooling. Currently they seem to have a particular focus on AI frameworks and tools like PyTorch/Tensorflow/ONNX. They have sponsored and helped with a lot of PyTorch development for example, so PyTorch support for AMD is much better than it was this time last year². ¹(https://github.com/ROCm/HIP https://github.com/ROCm/HIP) ²(https://pytorch.org/blog/experience-power-pytorch-2.0/ https://pytorch.org/blog/experience-power-pytorch-2.0/)

doikor 3y ago

Simulating nuclear explosions, effects of decay on nuclear weapon stockpiles, etc (this is literally what the El Capitan supercomputer mentioned in the article is for)

neverrroot 3y ago
Wrong reply, please ignore.
- hutzlibu 3y ago
  You can also delete your wrong comments .. (within 2 h I think)
  - neverrroot 3y ago
    I couldn’t see anything about deleting, only edit. Will check out the web version on a desktop next time, maybe it’s available there.
    - hutzlibu 3y ago
      Not sure, if you have to have more karma for it (would not make sense to me), but for me the delete button is right next to the edit button.
  - mkl 3y ago
    Within 2h, and if no one has replied.
  - jacquesm 3y ago
    Until you commented...

neverrroot 3y ago

Everything is possible, especially what they have in mind, where they will do a custom implementation. CUDA is important for the existing ecosystem, but that doesn’t make it the only show in town.

bayindirh 3y ago

HPC admin here. ROCm, which is AMD's equivalent of CUDA. The thing is you don't have to directly interface with CUDA or ROCm. Once the framework you want to use supports these, you're done. AMD is consistently getting used on the TOP500 machines, and this gives them insane amounts of money to improve ROCm. CUDA has an ecosystem and hardware moat, but not because they're vastly superior, but because AMD both prioritized processors first, and NVIDIA played dirty (OpenCL support and performance shenanigans, anyone?). This moat is bound to be eaten away by both Intel and AMD, and compute will be commoditized. NVIDIA foresaw this and bought Mellanox and wanted ARM to be a complete black box, but it didn't work out. Ethernet consortium woke up and got agitated by the fact that only ultra-low latency fabric provider is not independent anymore, so they're started to build alternatives to Infiniband ecosystem. Interesting times are ahead. Also there's OneAPI, but I'm not very knowledgeable about it. It's a superAPI which can target many compute platforms, like OpenCL, but takes a different approach. It compiles to platform native artifacts (CUDA/ROCm/Intel/x86/custom, etc.) IIRC.

justinclift 3y ago
> they're started to build alternatives to Infiniband ecosystem. Cool, that sounds interesting. Anything you can point at? :)
- timschmidt 3y ago
  UltraEthernet
  - justinclift 3y ago
    Thanks. :)
- viewtransform 3y ago
  Discussion on ultrafast ethernet and partners at AMD's December presentation. https://youtu.be/tfSZqjxsr0M?t=5198 https://youtu.be/tfSZqjxsr0M?t=5198
formerly_proven 3y ago