Path: utzoo!attcan!uunet!wuarchive!udel!princeton!siemens!xydeco!vlo
From: vlo@xydeco.siemens.com (John Vlontzos)
Newsgroups: comp.dsp
Subject: Re: Mystery FFT Code...
Message-ID: <32040@siemens.siemens.com>
Date: 11 Jun 90 14:24:05 GMT
References: <30678@cup.portal.com>
Sender: news@siemens.siemens.com
Reply-To: vlo@xydeco.siemens.com (John Vlontzos)
Organization: Siemens Corporate Research, Inc.
Lines: 9

Looking at the code:
nn is the log(N) (base 2) that is for 1024 points, nn=10
isign controls the sign of exp(2kp/N) (forward/inverse fft)
ifft=(1/N)fft(X*) where X*=complex conjugate.
I didn't look too hard but I think the routine doesn't
divide by N when computing the inverse fft. 

John Vlontzos
Siemens