• @brucethemoose
    link
    English
    30
    edit-2
    5 hours ago

    Holy moly this is awesome! I am in for the 128GB SKU.

    That’s 96GB of usable VRAM! And way more CPU bandwidth than any desktop Zen chip.

    I know people are going to complain about non upgradable memory, but you can just replace the board, and in this case it’s so worth it for the speed/power efficiency. This isn’t artificial crippling, it physically has to be soldered, at least until LPCAMM catches on.

    My only ask would be a full X16 (or at least a physical X16/electrical x8) PCIe slot or breakout ribbon. X4 would be a bit of a bottleneck for some GPUs/workloads… Does Strix Halo even support that?

    • @[email protected]
      link
      fedilink
      English
      206 hours ago

      I understand the memory constraints but it does feel weird for framework, is all I have to say. But that’s also the general trajectory of computing from what it seems. I really want lpcamm to catch on!

      • @brucethemoose
        link
        English
        16
        edit-2
        5 hours ago

        Eventually most system RAM will have to be packaged anyway. Physics dictates that one pays a penalty going over pins and mobo traces, and it gets more severe with every advancement.

        It’s possible that external RAM will eventually evolve into a “2nd tier” of system memory, for background processes, spillover, inactive programs/data, things like that.

        • @[email protected]
          link
          fedilink
          English
          75 hours ago

          It’s already fourth tier after L1, L2, L3 caches.

          Maybe something like optane will make a comeback. Having 16gb of soldered RAM and 500gb of relatively slow, but inexpensive optane RAM would be great.

          • @brucethemoose
            link
            English
            23 hours ago

            DRAM is so cheap and ubiquitous that they will probably keep using that, barring any massive breakthroughs. The “persistence after power-off” is nice to have, but not strictly needed.

      • @Scholars_Mate
        link
        English
        196 hours ago

        Apparently Framework did try to get AMD to use LPCAMM, but it just didn’t work from a signal integrity standpoint at the kind of speeds they need to run the memory at.

        • @grue
          link
          English
          135 hours ago

          Sounds like it doesn’t bode well for the future of DIMMs at all, TBH.

          • @SpaceNoodle
            link
            English
            175 hours ago

            You have a DIMM view of the future.

          • Avid Amoeba
            link
            fedilink
            English
            5
            edit-2
            4 hours ago

            My AM5 system doesn’t post with 128GB of 5600 DDR5 at higher than 4400 at JEDEC timings and voltage. 2 DIMMs are fine. 4 DIMMs… rip. So I’d say the present of DIMMs is already a bit shaky. DIMMs are great for lots of cheap RAM. I paid a lot less than what I’d have to pay for the equivalent size of RAM in a Framework desktop.

    • @alleycat
      link
      English
      56 hours ago

      What’s a SKU? Google just says “Stock Keeping Unit”, but I don’t think that’s correct in this context.