I see what you mean now. I think you have established that
the maxima documentation
http://maxima.sourceforge.net/docs/manual/en/maxima_16.html#SEC52
is wrong. I

've tested other J-Bessel values and some Y-Bessel
values based on
http://idlastro.gsfc.nasa.gov/idl_html_help/BESELJ.html
The pari version of the J-Bessel fcn seems fine but the maxima
version seems backwards. The Y Bessel function seems fine.
The maxima version of the I-Bessel fcn is fine but the pari version
seems backwards, based on
http://idlastro.gsfc.nasa.gov/idl_html_help/BESELI.html
That graph indicates the K-Bessel is backwards.

Thanks for pointing this out. Before making changes, I want to
email the maxima people and see what they say.

+++++++++++++++++++++++++++++++++++=

On 10/27/06, dt <[EMAIL PROTECTED]> wrote:
>
> First,
> thank you for the quick replies.
>
> I did read the document at
> http://modular.math.washington.edu/sage/doc/html/ref/module-sage.functions.special.html
> before posting, but it didn't make it clear to me.  I understood that
> pari was the default, and that is fine with me.  I realize now that
> conventions were confusing me.
>
> OK, I see what is going on.  I was wondering if argument order was
> playing a role and I was instead calculating J(nu,x) nu=0..5 instead of
> J(nu,x) x=0..5
>
> I did indead do the plots you showed above that show zeroes.  Why is
> the position of x different in your last  exampe?
>
> What is very confusing and non-intuitive to me at least is why the
> conventions of the default bessel/pari and the 'maxima' option bessel
> are reversed and are contradicted by the documentation. Bear with me:
>
> We know from tables or independant sources the Bessel J of order n=3.0
> has a value of 0 at z=0 and a value of 0.128 and z=2.0
>
> The documentation cited above says:
> bessel_J(       nu, z, [alg=53], [prec=pari]) where alg='maxima' or 'pari'
>
> Now look at the following values from sage:
> bessel_J(2.,3) #default is pari
>   0.12894324947440206
> bessel_J(2.,3,'maxima')
>   0.48609126058589119
> bessel_J(2.,3,'pari')
>   0.12894324947440206 #this verfies default is pari
> Note that 0.128.. is the expected value for J(nu=2,z=3.) so the maxima
> value is 'wrong' see below
>
> bessel_J(3,2.,)
>   0.48609126058589108 #same as maxima answer with reversed args
> #i.e. this is bessel J (nu=2, z=3)
> bessel_J(3,2.,'maxima')
>   0.12894324947440211  #now we have the right answer
> Therefore the argument order is reversed for 'maxima' option and if
> fact, the default bessel has an order differenct from the documentation
> or am I missing something elsewhere?
>
> bessel_J(0.,3)
>   0.00000000000000000 #right
> bessel_J(0.,3,'maxima')
>   -0.26005195490193339  #'wrong'
> bessel_J(3,0.,'maxima')
> 0.00000000000000000 #right
>
> So, the argument order given in the linked documentation is consistent
> with maxima form but
> not with the default or with 'pari' option.  In fact, the examples in
> that documentation:
> "
> sage: bessel_J(2,1.1)  # last few digits are random
> 0.136564153956658000
> sage: bessel_J(0,1.1)  # last few digits are random
> 0.719622018527510801
> sage: bessel_J(0,1)    # last few digits are random
> 0.765197686557966605
> "
>
> Do not work unless you reverse the order of the arguments or add the
> "maxima" option.
>
> This is in fact what led me to believe a bug had been introduced.
>
> So the question is, is it intended that the default(Pari) bessel
> functions (I've only looked at J in detail, I don't know if this is a
> problem with the other bessel functions) have a different argument
> convention than the ''maxima" option, and if so, the documentation
> should describe this differences.  As it stands now, something is
> inconsistent.
>
> Sorry for the long post, I hope I've at least made my confusion clear.
>
>
> -john
>
>
> >
>

--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~----------~----~----~----~------~----~------~--~---

Reply via email to