|
![](/i/fill.gif) |
Ross wrote:
> "Alain" <aze### [at] qwerty gov> wrote in message
> news:41d5efe6@news.povray.org...
>
>
>>You hit a hardware limitation, that of the FPU that is "only" 64 bits
>>(double presision). To be able to use such whide ranges, you'd need quad
>>(128 bits) or octuple (256 bits) presision math, with a processor
>>equiped with a FPU capable of such.
>>Maybe some super computer have such a beast...
>
>
> If you want it fast, it's a hardware limitation. Isn't it theoretically
> possible to emulate it in software at the cost of speed? (no, i can't offer
> examples... just thought it is possible)
>
> if someone really cared, it seems they could probably substitute math
> subroutines in povray with a "big math" library.
>
> i had a course on this in college. i forget every word about it, except that
> I rendered a scene of the intersection of earth with a fictional comet as
> part of a word problem for a homework assignment. bah
>
It's absolutely possible to have virtualy infinite presision maths. But it get
sllooooooooooooooowwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww. How do you think they find new,
bilions digits
long, prime numbers? And why they find them less than once a year using some super
computers full
time? You have to simulate pen and paper operations using numerous sub results,
calculating each
digit individualy.
Alain
Post a reply to this message
|
![](/i/fill.gif) |