Newsgroups: comp.archives
Path: utzoo!utgpu!news-server.csri.toronto.edu!math.lsa.umich.edu!math.lsa.umich.edu!emv
From: krukar@pprg.unm.edu (Richard Krukar [CHTM])
Subject: [sci.math...] FFTs in C available
Message-ID: <1990Aug28.233946.25669@math.lsa.umich.edu>
Followup-To: sci.math,sci.math.num-analysis
Sender: emv@math.lsa.umich.edu (Edward Vielmetti)
Reply-To: krukar@pprg.unm.edu (Richard Krukar [CHTM])
Organization: University of Michigan, Department of Mathematics
Date: Tue, 28 Aug 90 23:39:46 GMT
Approved: emv@math.lsa.umich.edu (Edward Vielmetti)
X-Original-Newsgroups: sci.math,sci.math.num-analysis
Lines: 21

Archive-name: fast-fourier-transforms/28-Aug-90
Original-posting-by: krukar@pprg.unm.edu (Richard Krukar [CHTM])
Original-subject: FFTs in C available
Archive-site: pprg.unm.edu [129.24.13.10]
Archive-directory: /pub/misc
Reposted-by: emv@math.lsa.umich.edu (Edward Vielmetti)


	There is now a set of fft routines, written in C, available
for anonymous ftp from pprg.unm.edu ( 129.24.13.10 ).  Log in as
'anonymous' and supply username@hostname as passwd. Actually, any
string with an embedded @ will work as passwd.  You want pub/misc/ffts.tar.Z
so don't forget to go into binary mode.

	There are some rather long subroutines due to loop unrolling and
a large weight table, but there is a significant speed improvement over
the IEEE standard routines written in Fortran.  I am quite interested in
how these things do on different platforms, so mail me your 'constructive
critism.'

	Richard Krukar  (krukar@pprg.unm.edu)