Re: your mail

["Prof. John P. Ertel (wizard)" <jpe@nadn.navy.mil>]

If you get a chance, take a look at Hamming's book or Matmatical Recipies,
both do a very good job of covering the topic of both the concept and the
pit-falls of the FFT algorythm.  Basically, you really have only to
determine
1. if your data has 2^n points or not (yes -> skip to 3.)
2. "pad the data to obtain 2^n points
   2.a. pad data evenly from both ends with zeros
   2.b. pad data from one end or the other with zeros
3. write in the branching algorythm to calculate the "FFT tree"
4. normalize as you find appropriate
5. look for the aliases that have been generated by padding or from the
(usually incorrect) assumption that the time series started at -infinity
and goes to +infinity.

Using only these rules Hamming's algorythm, I can usually make a fully
branched FFT in and language or even spread sheet in less than 1 hour.
Try it and you'll see that it's really quite easy.

Now as far as "canned code" goes, I believe that Mathematica, MAPLE,
MatLab, and MathCad all have a true Hamming branched code built in
functions.  Certainly Mathmatical Recipies has this as listed code either
in FORTRAN or C or PASCAL.  Additionally, the IMSL libraries all have this
routine.  One last APP that includes a true FFT is "IGOR" which is a very
nice signal processing APP.  In fact, you'll probably have to search long
and hard to find a signal processing routine that is important but not
found in IGOR.

All of the above work well on any Mac (including those that do not really 
support hardware floating point but rather simple emulate via the SANE).
Of course, the faster the better but a PowerMac is hardly necessary.



On Fri, 15 Nov 1996, Quist, Oren               Phys wrote:

> I have FFT software and information available for PC's, but not Mac's.
> If you would like it, let me know.
>
> Oren Quist
> SDSU
>
> quisto@mg.sdstate.edu

ERTEL SENDS.         _____________________
                    / Prof. John P. Ertel \
                   /   jpe@nadn.navy.mil   \    
       +================================================+
       | Physics Department, 9C   Office : 410-293-6657 |
       |  U. S. Naval Academy        FAX : 410-293-3729 |
       |                            Home : 410-757-6618 |
       +================================================+