Re: Computing big numbers




Pubkeybreaker wrote:
> toorandom@xxxxxxxxx wrote:
> > Hi, im new in this group and i hope my question can be new too (i doubt
> > that) , im implementing a big number library , i have made sums and
> > multiplications with big numbers.
> > i need to do this for a microcontroler to have the ability of
> > public-key algorithms.
> > maybe discret log , i would like to implement modulus , i can implement
> > my current work without modulus... but it will be better and ill
> > restrict my bandwidth if i make this. but im having problems , does
> > anybody know a good starting website or how can i search in google for
> > big number arithmetic in computers, i have search , but with no
> > success, GMP is a little complex to understand , and if i made
> > multuplication and sum smaller than GMP i think modulus will be much
> > smaller
>
> Read Knuth, Vol 2.

Knuth vol2 isn't exactly complete for what you need to do crypto math
on say a microcontroller. Actually iirc it doesn't cover a lot of what
you need to know like fast reduction, or multiplications [other than
interpolation based stuff].

vol2 is a good starting point I agree but not what an implementor
needs. It's more of what a student needs.

Tom

.



Relevant Pages

  • Re: Where is SSLeay?
    ... Seems that Math::Pari handles the smaller multiplications better than ... Wrt to the factorial subroutine you supplied I find that Math::GMP takes ... over from Math::Pari as the preferred module at around factorial ... I've also written a module that wraps most of the GMP ...
    (comp.lang.perl.modules)
  • Re: Fibonacci Recursion, What??
    ... the number of additions and multiplications in fib-lucas and fib-11 ... yet under sbcl there is this huge difference ... use gmp for bignums. ... At least that's the only explanation that I can come up with off the ...
    (comp.lang.lisp)