• just_another_person@lemmy.world
    link
    fedilink
    English
    arrow-up
    11
    ·
    10 hours ago

    There have been rumblings about this for a year or so, and it’s interesting because Nvidia doesn’t have any large hardware inclusions in the gaming space, aside from Switch 2. Nvidia has NVM as an API abstraction for their hardware, so I assume this job is about working on improving performance or support on that end.

    Now, the big question is…why? It’s unlikely they would do this to squeeze about more performance out of their existing hardware for PC and Switch users alone. I think the general speculation would be that they are planning a product launch of some sort, and from the sound of the posting, I’m guessing they’ve got something to compete with AMD in the handheld market that isn’t Tegra.

    They’ve also been saying for years they intend to hop into the ARM space, but keep fumbling it with delays. Their N1X(?) is supposed to finally launch “soon”, but with component prices the way they are right now, I’d expect them to delay again. Their Grace chips were kind of a joke, and they didn’t come anywhere near the power efficiency of other SoC options in the market.

    All this to say: I think they’re hiring a small dedicated team to improving mobile or ARM gaming chips for…something, and that’s what these jobs are for. They need to squeeze some more juice from these chips to make them attractive at market.

    • folekaule@lemmy.world
      link
      fedilink
      English
      arrow-up
      4
      ·
      9 hours ago

      I agree with you, and in particular the general “they’re cooking something” sentiment. The rest of this is just my personal take, so take it with a grain of salt. I do not have any inside knowledge of this.

      Nvidia currently rules the AI space, but they don’t have much of a “moat”. In theory, anyone can swoop in and make their gold-rush shovels obsolete–or at least too expensive. They probably also look to what Apple has done with vertical integration and want some of that stability.

      I predict they will focus on:

      1. AI (as they already are)
      2. Cloud workloads, like cloud gaming, but also business workloads (beyond what has already been moving to the cloud at a dizzying clip), e.g., workstations. Basically, we’re back to thin clients (again).
      3. Mobile gaming. Maybe mobile CPU/SoC.
      4. Handheld gaming. These last two are important since the desktop gaming market will be stagnating with component prices being so ridiculous.
      5. Desktop gaming. This isn’t a priority for them right now, but If/when desktop gaming snaps back, there will be a huge backlog of demand.

      Linux runs the majority of items 1-4 and is gaining in 5. Windows will probably remain the platform of choice for business desktops for a while. But, Linux runs great on ARM. Windows can run on ARM and should handle business desktop workloads just fine.

      Most likely they’re hiring for mobile gaming first. That would be the quickest bang for the buck, in my opinion. However, that work (improving performance on Linux) will be useful for all of the points above.

      We’re entering interesting times in computing. It will be interesting to see how Apple fares since they had a lead in ARM development, but don’t have much of a hold on gaming.