Subject: GiNaC as the symbolic manipulation engine in Sage
From: Robert Dodier
Date: Tue, 12 Aug 2008 08:38:59 -0700 (PDT)
On Aug 11, 9:43 pm, rjf <fate... at gmail.com> wrote:
> 1. There are fixes to the general simplifier using hash coding that
> make collection and ordering of terms run much faster.
> In essence, replacing sorting that probably takes O(n^3) time by O(n)
> time. I have provided these, but there are some nuances to
> integrating these fixes that other people should look at.
Yeah, I remember that, thanks for the reminder.
I will take a look at merging that code.
> 2. The fixes are, to some extent, irrelevant for the following reason:
> If the problem is large and consists of polynomial arithmetic, you
> should use rat() form.
> If the problem is large and is NOT obviously polynomial arithmetic,
> you should make some substitutions and
> convert it to polynomial form if at all possible.
Well, after trying it, it appears rat doesn't help in the case at
hand.
In any event, a naive formulation should scale in a reasonable
manner, even if there are special methods to make it faster.
FWIW
Robert Dodier