Karim BELABAS on Fri, 15 Nov 2002 17:44:01 +0100 (MET)


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: Bug rnfconductor


On Fri, 15 Nov 2002, Igor Schein wrote:

> On Fri, Nov 15, 2002 at 02:31:13PM +0100, Karim BELABAS wrote:
> > On 1 Oct 2002, I wrote:
> [snip]
> > I've implemented these changes in both conductor() and discrayrelall(). This
> [snip]
> > Anything broken so far ? Igor ?
> ? lindep([I,1]);
>   ***   impossible assignment t_COMPLEX --> t_REAL.

Typo related to a series of LLL patches I've not mentioned yet (still
tuning). Compare

   m = 1/mathilbert(30);
   norml2(m * qflll(m))
   norml2(m * qflll(m, 1))

in current CVS and the stable version. For dense matrices, the integral LLL
is now superseded by the floating point one (much faster, same reduction
properties).

I'm investigating the other ones.

Thanks,

    Karim.
-- 
Karim Belabas                    Tel: (+33) (0)1 69 15 57 48
Dép. de Mathématiques, Bât. 425  Fax: (+33) (0)1 69 15 60 19
Université Paris-Sud             Email: Karim.Belabas@math.u-psud.fr
F-91405 Orsay (France)           http://www.math.u-psud.fr/~belabas/
--
PARI/GP Home Page: http://www.parigp-home.de/