Robert, the bfrj function takes a long time using this old code.
These numerical algorithms are based on Numerical Algorithms 10(1995)13-26.
B.C. Carlson Numerical computation of real or complex elliptic integrals.
There is new and improved code.
The new code has been updated to the algorithms in Journal of Computational
and Applied Mathematics 118 (2000) 71-85
Reduction theorems for elliptic integrands with the square root of
two quadratic factors
B.C. Carlson, James FitzSimons
I have made more changes since 2000.
Regards, Jim FitzSimons
-----Original Message-----
From: maxima-admin at math.utexas.edu [mailto:maxima-admin at math.utexas.edu] On
Behalf Of Robert Dodier
Sent: Sunday, July 02, 2006 8:58 PM
To: Jim FitzSimons
Cc: maxima at math.utexas.edu
Subject: Re: [Maxima] bfloat elliptic integrals
Jim,
Thanks a lot for working on the bigfloat elliptic integrals.
I am running the test script and it seems to be OK
although some parts take a long time. Also I am not
familiar with elliptic integrals so I really can't tell if the
results are correct or not.
If you are interested maybe you can arrange the test
script so that it has pairs of expressions like this:
[sin(0), sin(%pi/2), sin(%pi)];
[0, 1, 0];
The first is some input and the second is the expected result.
Then batch("myscript.mac", test); evaluates the two
expressions and tests whether they are the same.
All of the scripts in run_testsuite are arranged like that.
Thanks again for your contribution,
Robert Dodier
_______________________________________________
Maxima mailing list
Maxima at math.utexas.edu
http://www.math.utexas.edu/mailman/listinfo/maxima