Karim Belabas on Fri, 27 Jul 2018 12:36:47 +0200


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

Re: Running time of bnfinit() 

* Karim Belabas [2018-07-27 03:32]:
> * Jeroen Demeyer [2018-07-26 09:38]:
> > do you happen to have a rule-of-thumb to estimate the running time of a
> > bnfinit() call, say in terms of the degree and discriminant?
> 
> Not really. An average estimate, possibly, but with *huge* deviations...
[...]
> Basically, [assuming GRH] 
> (not the defining polynomial!), a discriminant < 10^40 should be fine
[...]

For some reason, I mistyped and half a paragraph was deleted, sorry. This is
somewhat incidental but, for large discriminants, I was assuming that the input
was a nfinit structure or a pair [polynomial, bound for disc(K) divisors],
otherwise factoring the polynomial discriminant will already be a showstopper.

Cheers,

    K.B.
--
Karim Belabas, IMB (UMR 5251)  Tel: (+33) (0)5 40 00 26 17
Universite de Bordeaux         Fax: (+33) (0)5 40 00 21 23
351, cours de la Liberation    http://www.math.u-bordeaux.fr/~kbelabas/
F-33405 Talence (France)       http://pari.math.u-bordeaux.fr/  [PARI/GP]
`