• Tyfud@lemmy.world
    link
    fedilink
    English
    arrow-up
    22
    ·
    3 days ago

    I’m not OP, but my guess is they’re referring to the Intel math bug that some i5’s had. I’m struggling to track it down, but it’s basically an issue with doing long division where the floating point math would produce a very wrong result.

    You can see more here at least for the bug/issue that existed in the 90’s here

    • smitten@lemmy.blahaj.zone
      link
      fedilink
      English
      arrow-up
      8
      ·
      3 days ago

      I’m not actually, just that a binary integer that overflows at 60 couldn’t exist, hence the 5.907 whatever bit length

      • Tyfud@lemmy.world
        link
        fedilink
        English
        arrow-up
        5
        ·
        3 days ago

        oh, that’s actually clever. And I’m saying that as a software engineer. I missed that possibility :)

        • smitten@lemmy.blahaj.zone
          link
          fedilink
          English
          arrow-up
          3
          ·
          3 days ago

          I should have phrased it differently, like “Ah yes, the well known 5.9068905956 bit integer.” But thanks