Dan Sugalski wrote:
>
> At 10:14 AM 1/2/01 +0000, David Mitchell wrote:
> >Nick Ing-Simmons <[EMAIL PROTECTED]> wrote
> > > BigFloat could well build on BigInt for its "mantissa" and have another
> > > int-of-some-kind as its exponent. We don't need to pack it tightly
> > > so we should probably avoid IEEE-like hidden MSB. The size of exponent
> > > is one area where "known range of int" is important.
> >
> >Or we could go the really obscene route and implement bigfloat using
> >bigints for both mantissa *and* exponent. Not much danger of
> >overflow then ;-)
>
> The sick thing was, I was actually considering this... :)
reality checking:
This scheme (the bigint + bigint float) would be a problem because
1: there's no memory gain over a bigint + bigint bigrat
2: without a precision level built in and maintained everywhere
the first irrational you run into will use up all your memory
3: never mind the irrationals, it'll happen on the first repeating fraction!
right?
(still agitating for a "precision" field built into the standard numbers)
--
David Nicol 816.235.1187 [EMAIL PROTECTED]
"the toad doesn't know..."
