This is the mail archive of the gsl-discuss@sources.redhat.com mailing list for the GSL project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: bug in specfunc/gamma_inc.c


Jason Hooper Stover writes:
 > gsl_sf_gamma_inc_Q (185.0, 200.0) gives the value 0.13594954199834325,
 > whereas a gp script gives 0.13594954213327904. The disparity
 > between gsl and gp grows as the second argument increases. (I guess
 > gp could be wrong, 

Hmmm... GSL-1.4 produces the correct answer to full double-precision
accuracy for gsl_sf_gamma_inc_Q(185,200) (for integer "a" this can be
verified analytically).

The gp front-end prints too many digits by default:

$ gp
? incgam(185,200)/gamma(185)
%1 = 0.1359495421332790409768941685
? default(realprecision,300)
? incgam(185,200)/gamma(185)
%2 = 0.13594954199834326027261113221577.....

 > but since it has almost-arbitrary precision, I figured
 > it's correct.)

I can't believe you said that ;-)

-- 
Brian Gough


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]