• darcmage@lemmy.dbzer0.com
    link
    fedilink
    arrow-up
    6
    ·
    edit-2
    4 months ago

    What I’m really waiting for someone to figure out is what makes the 13th/14th gen 7/9 series processors more prone to these failures compared to the 1/4/6 series and why the 12th gen chips remain unaffected given the minor architecture changes.

    • Contramuffin@lemmy.world
      link
      fedilink
      arrow-up
      11
      ·
      4 months ago

      Not sure if you saw Level1Tech’s recent video on the topic, but he speculated that it could be the area connecting the cache to the cores, as that was apparently changed to accommodate for more cores in the 13th/14th gen parts. The change was speculated to have made the connection weaker and more prone to degradation, especially when the connection was expected to communicate with a lot of cores (hence why this occurs mainly on high core count parts)

      • darcmage@lemmy.dbzer0.com
        link
        fedilink
        arrow-up
        4
        ·
        4 months ago

        Thanks, I watched it but I must’ve missed that part. If it does turn out that the 900mhz boost to the compute fabric is at fault, Wendell seems to be implying it might not be possible to solve with a microcode update. I hope that’s not the case but I guess we’ll find out soon enough.