Re: NEC-LIST: Processing times

At 09:40 AM 3/23/2004 -0800, Francisco Broissin wrote:
>Dear Sirs:
>
>We have found that when we use NEC Win Plus the calculation takes a few
>little seconds.
>
>On the other hand (and for the same example) by doing all the related
>integrals, etc, (within the same computer) the calculation takes around
>1 hour for 71 elements when we do that "at hand" with Maple !!
>
>That is great because we guess that NEC Win Plus works very well. But
>we are confused about the time by using Maple in a direct calculation
>by MoM versus the calculations by NEC Win Plus
>
>Does anyone have any explanation for that, having into account that is
>the same problem?
>
>By using Maple seems clear that the time should be longer but not so
>much as 1 hour for the same number of elements.
>
>Thank you very much
>Francisco Broissin and Cedric Blanchard

Some possible reasons:
Is Maple interpreted, while NEC is compiled?
Are you using the same basis functions in your maple code as NEC uses? Some
might be more compute intensive to evaluate.
Is Maple using an iterative method to solve the equations? NEC is a direct
algebraic solution in one step.

Does your model have 71 elements or 71 segments? If the latter, the work
for NEC is essentially that of building and inverting a 71x71 matrix (and
some other stuff), which is fairly quick. Maple may not be as efficient.

James Lux, P.E.
Spacecraft Telecommunications Section
Jet Propulsion Laboratory, Mail Stop 161-213
4800 Oak Grove Drive
Pasadena CA 91109
tel: (818)354-2075
fax: (818)393-6875

-- 
The NEC-List mailing list <nec-list_at_gweep.ca>
http://www.gweep.ca/mailman/listinfo.cgi/nec-list