hypergeometric([1.3, 1.84-%i], [1.8], 5) using SBCL + i386-apple-darwin11.4.0?



On 3/11/13 5:54 PM, Barton Willis wrote:
> My student using i386-apple-darwin11.4.0 + SBCL",
> lisp_version="1.0.55.0-abb03f9" gets a noun form for
> 
>   hypergeometric([1.3,1.84-%i],[1.8],5)
> 
> Using either SBCL",lisp_version="1.0.55.0-abb03f9" or 64 bit Clozure
> 1.9, I get
>  
>   (%i2) hypergeometric([1.3,1.84-%i],[1.8],5);
>   (%o2)            - 0.0090589676864236 %i - 0.0493215929836587
>  
> I've verified that this value is correct. Does anybody know what is
> going on?
> 
> Oh, I should try the (new) 32 bit CCL 1.9 to see if the bugs have gone
> away....
> 
>  
> 
> --Barton
> ------------------------------------------------------------------------
> 
> 
> _______________________________________________
> Maxima mailing list
> Maxima at math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima
> 

Using Maxima-5.29.1 (release), I get the right answer for the cases that
I've tried:

i386-apple-darwin10.8.0, SBCL-1.1.5 (release)
x86_64-apple-darwin12.2, SBCL-1.1.5 (release)
powerpc-apple-darwin9.8, SBCL-1.0.58 (release)

(%i1) hypergeometric([1.3,1.84-%i],[1.8],5);
STYLE-WARNING: redefining MAXIMA::SIMP-HYPERGEOMETRIC in DEFUN
STYLE-WARNING: redefining MAXIMA::FLOAT-OR-RATIONAL-P in DEFUN
(%o1)            - .009058967686423613 %i - .04932159298365872

-- 
Alexander Hansen, Ph.D.
Fink User Liaison
My package updates: http://finkakh.wordpress.com/