[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: n-point FFT
- Subject: Re: n-point FFT
- From: "Richard G. French" <rfrench(at)wellesley.edu>
- Date: Tue, 28 Nov 2000 02:09:24 GMT
- Newsgroups: comp.lang.idl-pvwave
- Organization: Wellesley College
- References: <3A1A7B49.FBECA944@onecert.fr>
- Xref: news.doit.wisc.edu comp.lang.idl-pvwave:22339
Has anyone tried to get an IDL interface working for the FFTW
routine that is documented on http://www.fftw.org/? For those
of you who don't know what it is, here is a brief description from their
"FFTW is a C subroutine library for computing the Discrete Fourier
Transform (DFT) in one or more dimensions, of both real and complex
data, and of arbitrary input size. We believe that FFTW, which is free
software, should become the FFT library of choice for most applications.
Our benchmarks, performed on on a variety of platforms, show that
FFTW's performance is typically superior to that of other publicly
available FFT software. Moreover, FFTW's performance is portable: the
program will perform well on most architectures without modification."
FFTW has lots of different methods of doing FFT's, and one of the novel
features of the program is that it figures out which method is fastest
for your particular problem by running a test case. Then, once it has
selected the best method, it uses that for all of the runs you need to
do. It is most useful when you have lots of transforms to do, and you
want to have them all done as fast as possible.
I've seen an fftw.so version on a SUN Ultra-5 distribution, but I have
not yet been able to get it working on my DEC (oops - Compaq) Alpha. Has
anyone else tried to figure out how to get this working?