Re: Well DUH ! AI People Finally Realize They Can Ditch Most Floating-Point for Big Ints

On 10/18/24 12:34 PM, Richard Kettlewell wrote:

"186282@ud0s4.net" <186283@ud0s4.net> writes:

On 10/16/24 6:56 AM, Richard Kettlewell wrote:

"186282@ud0s4.net" <186283@ud0s4.net> writes:

    BUT ... as said, even a 32-bit int can handle fairly
    large vals. Mult little vals by 100 or 1000 and you can
    throw away the need for decimal points - and the POWER
    required to do such calx. Accuracy should be more than
    adequate.

You’re talking about fixed-point arithmetic, which is already used
where appropriate (although the scale is a power of 2 so you can
shift products down into the right place rather than dividing).
>

    In any case, I'm happy SOMEONE finally realized this.
>
    TOOK a really LONG time though ......

>
It’s obvious that you’ve not actually read or understood the paper
that this thread is about.

>
   Maybe I understood it better than you ... and from
   4+ decades of experiences.

 Perhaps you could explain why you keep talking about integer arithmetic
when the paper is about floating point arithmetic, then.

   Umm ... because the idea of swapping FP for ints in
   order to save lots of power was introduced ?
   This issue is getting to be *poitical* now - the
   ultra-greenies freaking about how much power the
   'AI' computing centers require.