Karim BELABAS on Tue, 22 Oct 2002 15:11:55 +0200 (MEST)

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

Re: serreverse() on intmods

On Mon, 21 Oct 2002, Vasily Golyshev wrote:
> I seem to be in need of a version of serreverse() that would
> accept intmods as coefficients.
> Doing  Mod(serreverse(lift(*)),p) is no good since serreverse on ints is
> much slower than it would be on intmods because of the accumulation of
> magnitudes as the precision of the series grows. Which would be the
> recommended way of writing that?

Provided the modulus has no small prime divisors, I don't see anything wrong
with serreverse with intmod coefficients.

If it has, lift and use p-adic coefficients insteads: multiply by (1+O(p^n))
for a suitable n: 1 + ceil( log_p(number of significant terms ) )  should be
enough [untested].

Hope this helps,

Karim Belabas                    Tel: (+33) (0)1 69 15 57 48
Dép. de Mathematiques, Bat. 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/