Re: PARI/GP support for finite fields element v2

[Bill Allombert <Bill.Allombert@math.u-bordeaux1.fr>]

On Mon, Mar 26, 2007 at 01:17:46PM +1000, Joerg Arndt wrote:
> * Bill Allombert <Bill.Allombert@math.u-bordeaux1.fr> [Mar 26. 2007 10:27]:
> 
> I assume there will not be a change in the efficiency of the
> methods, just a better interface, right?

Depends what you call the methods. It is way faster than using
Mod(x*Mod(1,n),ffinit(p,n))

> Specialized routines for GF(p^n) where p<256 would be faster
> than the existing routines (using version 2.3.1).
> For binary fields, the NTL library gives state of art peformace.

Special support for GF(2,n) is in the work.  But this will still be much
slower than NTL.

> I suggest to include (make somehow usable) the known factorizations
> of numbers p^n-1, see 
> http://www.jjj.de/pari/mersfact.inc.gp

There are some plan to do that, but there are better way to do it than
to store the whole factorisation. 

Cheers,
Bill.