51–60 of 147 posts
There is an unofficial Bruhnspace project...but it is really sad that AMD has made a management decision that prevents use of a perfectly viable Ryzen laptop to make use of these libraries.
Unlike...say a M1
Earlier quoted context omitted.
It’s worse.... It’s Linux only so no Mac, Windows or WSL. No support what so ever for APUs which means if you have a laptop without a dedicated GPU you’re out of luck (tho discrete mobile GPUs aren’t officially supported either and often do not work). They’ve not only haven’t been supporting any of their consumer based R“we promise it’s not GCN this time”DNA GPUs, but since December last year (2020) they’ve dropped s…
> or WSL Does WSL not support PCIe passthrough?
And even then, almost all AMD customer GPUs are not supported. (Everything but Vega)
APUs? Nope too.
Earlier quoted context omitted.
Oh oof. Thanks for saving me time not having to look up ROCm benchmarks. I find it really surprising they they don't wanna compete on performance/$ at all by not supporting consumer cards
I think this is more of an issue that they have Compute optimised and Graphics optimised cards and Vega is their last compute optimised card. It would be very nice for them to refresh their compute cards as well.
Now that major frameworks finally started supporting ROCm, AMD has half-abandoned it (IIRC the last consumer cards supported were the Vega ones, cards from 2 generations ago). I hope this will change.
I work for AMD, but this comment contains exclusively my personal opinions and information that is publicly available. ROCm has not been abandoned. PyTorch is built on top of rocBLAS, rocFFT, and Tensile (among other libraries) which are all under active development. You can watch the commits roll in day-by-day on their public GitHub repositories. I can't speak about hardware support beyond what's written in the docs…
CUDA works with basically any card made in the last 5 years, consumer or compute. ROCm seems to work with a limited set of compute cards only.
There's a ROCm team in Debian [1] trying to push ROCm forward (ROCm being open), but just getting supported hardware alone is already a major roadblock, which stalls the effort, and hence any contributions Debian could give back.
Ryzen APU - the embedded GPU inside laptops is not supported by AMD for running rocm There is an unofficial Bruhnspace project...but it is really sad that AMD has made a management decision that prevents use of a perfectly viable Ryzen laptop to make use of these libraries. Unlike...say a M1
Another reason is certainly that they simply don't need to - just like Intel's iGPU, people working with deep learning opt for discrete GPUs (either built-in or external), both just isn't an option (yet?) for M1-based systems.
The audience would be a niche within a niche and the cost-benefit-ratio doesn't seem to justify the effort for them.
Ryzen APU - the embedded GPU inside laptops is not supported by AMD for running rocm There is an unofficial Bruhnspace project...but it is really sad that AMD has made a management decision that prevents use of a perfectly viable Ryzen laptop to make use of these libraries. Unlike...say a M1
Might have to do with the fact that AMD just doesn't seem to have the resources (see the common complaints about their drivers' quality) to fully support every chip. Another reason is certainly that they simply don't need to - just like Intel's iGPU, people working with deep learning opt for discrete GPUs (either built-in or external), both just isn't an option (yet?) for M1-based systems. The audience would be a nic…
- https://developer.apple.com/documentation/mlcompute
- https://blog.tensorflow.org/2020/11/accelerating-tensorflow-...
For a more personal take on your answer - do consider the rest of world. For example, Ryzen is very popular in India. Discrete GPU are unaffordable for that college student who wants to train a non-English NLP model on GPU.
Earlier quoted context omitted.
Jetson Xavier NX, but that comes with a high price tag. It’s much more powerful however.
I'll also add a caveat that toolage for Jetson boards is extremely incomplete. They supply you with a bunch of sorely outdated models for TensorRT like Inceptionv3 and SSD-MobileNetv2 and VGG-16. WTF, it's 2021. If you want to use anything remotely state-of-the-art like EfficientDet or HRNet or Deeplab or whatever you're left in the dark. Yes you can run TensorFlow or PyTorch (thankfully they give you wheels for thos…
TVM is another alternative to get models to inference fast on nano
Earlier quoted context omitted.
I think this is more of an issue that they have Compute optimised and Graphics optimised cards and Vega is their last compute optimised card. It would be very nice for them to refresh their compute cards as well.
Going for market segmentation like that sounds like a pretty bad idea if you are already the underdog in the game.
AMD meanwhile considers GPU compute as a premium feature... which is a different approach.
Not having a shippable IR like PTX but explicitly targeting a given GPU ISA, making this unshippable outside of HPC and supporting Linux only also points in that direction.
Intel will end up being a much better option than AMD for GPU compute once they ship dGPUs... in their first gen.
Now that major frameworks finally started supporting ROCm, AMD has half-abandoned it (IIRC the last consumer cards supported were the Vega ones, cards from 2 generations ago). I hope this will change.
I work for AMD, but this comment contains exclusively my personal opinions and information that is publicly available. ROCm has not been abandoned. PyTorch is built on top of rocBLAS, rocFFT, and Tensile (among other libraries) which are all under active development. You can watch the commits roll in day-by-day on their public GitHub repositories. I can't speak about hardware support beyond what's written in the docs…
I want to buy AMD because they are more open than Nvidia. But Nvidia supports CUDA day one for all their graphic cards and AMD still don't have rocm support on most of their product even years after their release [0]
Given AMD size & budget, the reason why they don't hire a few more employee full time on making rocm work with their own graphic card is beyond me.
The worst is how they keep people in waiting. It's always vague phrases like "not currently", "may be supported in the future", " "future plan", " we cannot comment on specific model support ", etc.
AMD doesn't want rocm on consumer card ? Then say it. Stop making me check rocm repos every week to get always more disappointed.
AMD plans to support it on consumer card ? Then say it and give a release date : "In May 2021 , the RX 6800 will get rocm support, thanks for your patience and your trust in our product".
I like AMD for their openness and support of standards, but they are so unprofessional when it comes to Compute