Re: [Help-gsl] bessel performance

[Brian Gough]

At Sat, 4 Apr 2009 20:09:36 -0400,
Gideon Simpson wrote:
> If I have an array of points at which I want to evaluate a bessel  
> function, I should expect better performance by using  
> gsl_sf_bessel_sequence_Jnu_e, as opposed to gsl_sf_bessel_Jnu, right?

Yes, that's the intent.

> Also, why does the former include a precision mode flag, while the  
> latter does not?

Typically only the more specialised functions, like
gsl_sf_bessel_sequence_Jnu, support it.  I think originally the plan
was to offer it on all functions but that was too much work.

-- 
Brian Gough
(GSL Maintainer)

Support freedom by joining the FSF!
http://www.fsf.org/associate/support_freedom/join_fsf?referrer=37