Linux? · Issue #2 · amd/RyzenAI-SW
github.com
https://www.phoronix.com/news/AMD-Ryzen-AI-Open-Source-Demo AMD Has Open-Source Ryzen AI Demo Code - But Only For Windows

Please create a comment or react with an emoji there.

(IMO, they should’ve limited comments,and gone with reaction count there, its looks mess right now )

  • tal@lemmy.today
    15·
    1 year ago

    if it can use all the system RAM it might provide medium-fast inference of decent models

    Yeah, I get what you mean – if I can throw 128GB or 256GB of system memory and parallel compute hardware together, that’d enable use of large models, which would let you do some things that can’t currently be done other than (a) slowly, on a CPU or (b) with far-more-expensive GPU or GPU-like hardware. Like, you could run a huge model with parallel compute hardware in a middle ground for performance that doesn’t exist today.

    It doesn’t really sound to me like that’s the goal, though.

    https://www.tomshardware.com/news/amd-demoes-ryzen-ai-at-computex-2023

    AMD Demoes Ryzen AI at Computex 2023

    AI for the masses.

    The goal for the XDNA AI engine is to execute lower-intensity AI inference workloads, like audio, photo, and video processing, at lower power than you could achieve on a CPU or GPU while delivering faster response times than online services, thus boosting performance and saving battery power.

    Much of the advantage of having an inbuilt AI engine resides in power efficiency, a must in power-constrained devices like laptops, but that might not be as meaningful in an unconstrained desktop PC that can use a more powerful dedicated GPU or CPU for inference workloads – but without the battery life concerns.

    I asked McAfee if those factors could impact AMD’s decision on whether or not it would bring XDNA to desktop PCs, and he responded that it will boil down to whether or not the feature delivers enough value that it would make sense to dedicate valuable die area to the engine. AMD is still evaluating the impact, particularly as Ryzen 7040 works its way into the market.

    That sounds like the goal is providing low-power parallel compute capability. I’m guessing stuff like local speech recognition on laptops would be a useful local, low-power application that could take advantage of parallel compute.

    The demo has it doing facial recognition, though I don’t really know where there’s a lot of demand for doing that with limited power use today.

    • tal@lemmy.today
      4·
      1 year ago

      I can imagine that there would be people who do want cheap, low-power parallel compute, but speaking for myself, I’ve got no particular use for that today. Personally, if they have available resources for Linux, I’d rather that they go towards improving support for beefier systems like their GPUs, doing parallel compute on Radeons. That’s definitely an area that I’ve seen people complain about being under-resourced on the dev side.

      I have no idea if it makes business sense for them, but if they can do something like a 80GB GPU (well, compute accelerator, whatever) that costs a lot less than $43k, that’d probably do more to enable the kind of thing that @fhein@lemmy.world is talking about.

    • CeeBee@lemmy.world
      2·
      1 year ago

      The demo has it doing facial recognition

      The demo is just face detection and not recognition. But usually if something can run one, then it can run the other.

    • bouh@lemmy.world
      1·
      1 year ago

      IMO they’d be idiot not to go hard on this. More efficiency on the computing needed for AI can quickly scale to a’y application that will be developed in the future.

      We are going in a future of limited resources and expensive energy. That’s a short term problem.