Mersenne: Re: Possible idea for increasing GIMPS throughput

[Steinar H. Gunderson]

On Thu, Sep 30, 1999 at 11:16:27PM -0700, Mike Bandsmer wrote:
>Basically, the article gives a simple method of computing an n-point DFT
>and an n-point IDFT using only a single complex n-point DFT (and minor
>additional computations of complexity O(n)).  

[...]

>In this way, two exponents could be tested using the same number of
>FFT/IFFT operations as is currently being used to test 1 exponent.

Unless _I'm_ the one mixing up things here (very likely...), a possible
error might be that you're mixing DFT and FFT? Aren't those two different
algorithms?

/* Steinar */
-- 
Homepage: http://members.xoom.com/sneeze/
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ      -- http://www.tasam.com/~lrwiman/FAQ-mers