Re: CCL:3D complex to complex FFT routine, Car-Parrinelo

[Keith Refson <Keith.Refson { *at * } earth.ox.ac.uk>]

 Dear All,
 > Before I do so, I wonder if somebody has a routine equivalent to SCFT3
 > and is willing to make it available through the net.
 Anyone looking for a good FFT could do a lot worse than to ftp
 "/pub/gpfa.tar.Z" from here "earth.ox.ac.uk".  This contains
 an in-place, self-sorting complex FFT of radix 2,3,5 with multiple
 1-D and 3-D interfaces.
 These use the recent algorithms described in
 Ref: Clive Temperton, Siam J. Sci. Stat. Comput 13(3), 676-686 (1992)
 which represent a considerable advance in speed and workspace requirements.
 In every case I have tested, these routines (in FORTRAN) are faster than
 computer suppliers' own tuned library FFTs (often in assembler).  I have
 been using them for car-parrinello type calculations for over a year
 now with considerable success.
 Keith Refson
 ------------------------------------------------------------------------------
 | Email   : keith { *at * } earth.ox.ac.uk    | Dr Keith Refson, Dept of Earth
 Sciences|
 | TEL(FAX): +44 1865 272026 (272072)| Parks Road, Oxford OX1 3PR, UK         |
 ------------------------------------------------------------------------------