Re: Axiom's integration non-deterministic?

On 3/24/06, Page, Bill <[email protected]> wrote:
> On Friday, March 24, 2006 5:56 PM Igor Khavkine wrote:

> > Clearly, Axiom takes two different paths through the integration
> > algorithm, even when give identical input. What is the cause of
> > the branch? Is there a non-deterministic step somewhere in the
> > algorithm?
>
> I don't know the cause of this behaviour but I get the same effect
> on the Windows version of Axiom.

So there must be something strage going on in the integration code.
I've been trying to trace the algorithm's execution to see what it
does exactly in each case. Unfortunately, I'm having little luck. I
played with a few variations of )trace command, but couldn't find the
right one.

I've identified the routine 'integrate' in the FSINT package. Knowing
that, is there a way for me to step through its code line by line like
in a usual debugger?

Thanks.

Igor