d@0: d@0: d@0: Acknowledgments - FFTW 3.2.1 d@0: d@0: d@0: d@0: d@0: d@0: d@0: d@0: d@0: d@0: d@0: d@0: d@0:
d@0:

d@0: d@0: Next: , d@0: Previous: Installation and Customization, d@0: Up: Top d@0:


d@0:
d@0: d@0:

10 Acknowledgments

d@0: d@0:

Matteo Frigo was supported in part by the Special Research Program SFB d@0: F011 “AURORA” of the Austrian Science Fund FWF and by MIT Lincoln d@0: Laboratory. For previous versions of FFTW, he was supported in part by the d@0: Defense Advanced Research Projects Agency (DARPA), under Grants d@0: N00014-94-1-0985 and F30602-97-1-0270, and by a Digital Equipment d@0: Corporation Fellowship. d@0: d@0:

Steven G. Johnson was supported in part by a Dept. of Defense NDSEG d@0: Fellowship, an MIT Karl Taylor Compton Fellowship, and by the Materials d@0: Research Science and Engineering Center program of the National Science d@0: Foundation under award DMR-9400334. d@0: d@0:

Code for the Cell Broadband Engine was graciously donated to the FFTW d@0: project by the IBM Austin Research Lab. d@0: d@0:

Code for the MIPS paired-single SIMD support was graciously donated to d@0: the FFTW project by CodeSourcery, Inc. d@0: d@0:

We are grateful to Sun Microsystems Inc. for its donation of a d@0: cluster of 9 8-processor Ultra HPC 5000 SMPs (24 Gflops peak). These d@0: machines served as the primary platform for the development of early d@0: versions of FFTW. d@0: d@0:

We thank Intel Corporation for donating a four-processor Pentium Pro d@0: machine. We thank the GNU/Linux community for giving us a decent OS to d@0: run on that machine. d@0: d@0:

We are thankful to the AMD corporation for donating an AMD Athlon XP 1700+ d@0: computer to the FFTW project. d@0: d@0:

We thank the Compaq/HP testdrive program and VA Software Corporation d@0: (SourceForge.net) for providing remote access to machines that were used d@0: to test FFTW. d@0: d@0:

The genfft suite of code generators was written using Objective d@0: Caml, a dialect of ML. Objective Caml is a small and elegant language d@0: developed by Xavier Leroy. The implementation is available from d@0: http://caml.inria.fr/. In previous d@0: releases of FFTW, genfft was written in Caml Light, by the same d@0: authors. An even earlier implementation of genfft was written in d@0: Scheme, but Caml is definitely better for this kind of application. d@0: d@0: FFTW uses many tools from the GNU project, including automake, d@0: texinfo, and libtool. d@0: d@0:

Prof. Charles E. Leiserson of MIT provided continuous support and d@0: encouragement. This program would not exist without him. Charles also d@0: proposed the name “codelets” for the basic FFT blocks. d@0: d@0: Prof. John D. Joannopoulos of MIT demonstrated continuing tolerance of d@0: Steven's “extra-curricular” computer-science activities, as well as d@0: remarkable creativity in working them into his grant proposals. d@0: Steven's physics degree would not exist without him. d@0: d@0:

Franz Franchetti wrote SIMD extensions to FFTW 2, which eventually d@0: led to the SIMD support in FFTW 3. d@0: d@0:

Stefan Kral wrote most of the K7 code generator distributed with FFTW d@0: 3.0.x and 3.1.x. d@0: d@0:

Andrew Sterian contributed the Windows timing code in FFTW 2. d@0: d@0:

Didier Miras reported a bug in the test procedure used in FFTW 1.2. We d@0: now use a completely different test algorithm by Funda Ergun that does d@0: not require a separate FFT program to compare against. d@0: d@0:

Wolfgang Reimer contributed the Pentium cycle counter and a few fixes d@0: that help portability. d@0: d@0:

Ming-Chang Liu uncovered a well-hidden bug in the complex transforms of d@0: FFTW 2.0 and supplied a patch to correct it. d@0: d@0:

The FFTW FAQ was written in bfnn (Bizarre Format With No Name) d@0: and formatted using the tools developed by Ian Jackson for the Linux d@0: FAQ. d@0: d@0:

We are especially thankful to all of our users for their d@0: continuing support, feedback, and interest during our development of d@0: FFTW. d@0: d@0: d@0: d@0: