2008/4/23 David Harvey <[EMAIL PROTECTED]>:
>
>
> On Apr 23, 2008, at 5:38 PM, Martin Albrecht wrote:
>
> > 3) search for a tri- or pentanomial with some code similar to the
> > one in blog
> > post
>
> You might want to check the NTL code for whether NTL "auto-detects"
> that you have suppl
Thanks for the debugging. I am trying it out (sage -br takes many
minutes with this for some reason). If that works then I'll follow
your suggestion to move this polynomial selection somewhere else.
Should have a patch ready tomorrow...
John
2008/4/23 Martin Albrecht <[EMAIL PROTECTED]>:
>
>
On Apr 23, 2008, at 5:38 PM, Martin Albrecht wrote:
> 3) search for a tri- or pentanomial with some code similar to the
> one in blog
> post
You might want to check the NTL code for whether NTL "auto-detects"
that you have supplied a sparse polynomial (and hence uses faster
code for arith
On Wednesday 23 April 2008, John Cremona wrote:
> Thanks, Martin. I agree with your comments. All I am really talking
> about here is the time taken to construct the field -- I know that a
> lot more work will be needed to get a really good finite field setup
> in Sage (including arbitrary coerc
Thanks, Martin. I agree with your comments. All I am really talking
about here is the time taken to construct the field -- I know that a
lot more work will be needed to get a really good finite field setup
in Sage (including arbitrary coercions into extension fields, as
previously discussed).
I
On Wednesday 23 April 2008, John Cremona wrote:
> There's a serious inefficiency in the construction of finite fields
> GF(2^n) in Sage at the moment.
> For n<=15 these are constructed of type
> sage.rings.finite_field_givaro.FiniteField_givaro, about which I have
> nothing to say now; for n>=16
