On Dec 19, 2007, at 7:19 PM, Bill Hart wrote:
> > I get about 7us per loop on sage.math for Pari for the exponentiation. > So perhaps this is all architecture dependent. This would not surprise > me in the slightest. > > At any rate, I suspect the algorithms used for factorisation are > implemented quite differently between NTL and Pari. Since both Pari > and NTL use GMP mpn's for underlying integer arithmetic in SAGE, I > think the algorithms being used for factorisation are much more > relevant than the speed of basic arithmetic, which should be the same > for both. > > The other point is that both Pari and NTL use finite field stuff to > factor polynomials over the integers. So the speed of integer > arithmetic is almost irrelevant. > > Having said all that, it is surprising that NTL is behind Pari for > polynomial factorisation, given the amount of work Victor put into > this. Can you try your example problems on sage.math? On the other hand, Shoup's research is mostly about the asymptotics for very large degree problems, so perhaps he didn't bother trying to optimise the small polynomial case very much (?) david --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/ -~----------~----~----~----~------~----~------~--~---
