CVS testsuite failure erfi(-0.15) = 0.17



Windows + git head + Clozure 1.9 gives -0.170..... This agrees with Wolfram Alpha  - I * Erf[- 0.15 * I] .

Maxima branch_5_30_base_98_g29f9239_dirty http://maxima.sourceforge.net
using Lisp Clozure Common Lisp Version 1.9-r15764  (WindowsX8632)

(%i1) erfi(- 0.15);
(%o1) -0.17053491636155

--Barton

________________________________________
From: maxima-bounces at math.utexas.edu [maxima-bounces at math.utexas.edu] on behalf of Raymond Toy [toy.raymond at gmail.com]
Sent: Tuesday, June 04, 2013 11:14
To: maxima at math.utexas.edu
Subject: Re: [Maxima] CVS testsuite failure erfi(-0.15) = 0.17

>>>>> "David" == David Billinghurst <dbmaxima at gmail.com> writes:

    David> On 4/06/2013 1:43 PM, Raymond Toy wrote:
    >> On 6/2/13 8:04 AM, David Billinghurst wrote:
    >>> I get a testsuite failure with CVS maxima under gcl/windows.
    >>>
    >>> test is rtest_gamma problem 552 and it reduces to
    >>>
    >>> (%i1) erfi(- 0.15);
    >>> (%o1)                          0.17053491636155
    >>>
    >>> With maxima-5.30.0 and mma I get the opposite sign.
    >> Interesting.  With a pretty recent build, I get -0.17 using gcl on a
    >> sparc box. (The only machine I have that will run gcl.)
    >>
    >> This is going to be difficult to work out.  Did you get this on with
    >> your windows installer?  If so, I'll have to try it there.
    >>
    >> Ray
    >>

    David> I don't see these with my windows installer for 5.30.0 - only with git head.

    David> I can do a binary search through the git tree to see when it started,
    David> and do some debugging.

That would be great.  I did a quick diff for changes between head and
5.30.0.  The erf routines haven't changed.

Ray

_______________________________________________
Maxima mailing list
Maxima at math.utexas.edu
http://www.math.utexas.edu/mailman/listinfo/maxima