• dan@upvote.au
    link
    fedilink
    arrow-up
    0
    ·
    7 months ago

    Is the FPU a reference to the Pentium FDIV bug?? What a throwback.

      • dan@upvote.au
        link
        fedilink
        arrow-up
        0
        ·
        edit-2
        7 months ago

        That’s just due to IEEE754 floating point. The CPU would do it the same way if you had a software-based implementation.

        0.1 in binary doesn’t have a precise value. It’s recurring, the same way we can’t represent 1/3 (0.333333333…) precisely in decimal.

        • frezik@midwest.social
          link
          fedilink
          arrow-up
          0
          ·
          7 months ago

          In other words, “computes numbers incorrectly”.

          You don’t have to overthink it on a meme that describes a hard drive as “remembers numbers loudly”.