[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#477806: ITP: djbfft -- extremely fast library for floating-point convolution



Package: wnpp
Severity: wishlist

Package Name:	djbfft
Version:	0.76
Upstream Author:	D. J. Bernstein
URL:	http://cr.yp.to/djbfft.html
License:	Public Domain
Programming Language:	C

 djbfft is an extremely fast library for floating-point convolution.
 The current version holds most of the speed records for
 double-precision FFTs on general-purpose computers.
 .
 djbfft provides power-of-2 complex FFTs, real FFTs at twice the
 speed, and fast multiplication of complex arrays. Single precision
 and double precision are equally supported.


Paul Wise schrieb:
On Sat, Mar 22, 2008 at 9:19 PM, Fabian Greffrath
<fabian.greffrath@lycos.de> wrote:
Am Samstag, den 22.03.2008, 08:47 +0100 schrieb Sebastian Dröge:

Yes, it does exactly that (and uses xine's own memalign function).
 > Unless you build the package with -DLIBA52_DJBFFT and get djbfft from
 > somewhere we won't have any accelerated IMDCT transform.
 >
 > No idea what djbfft exactly is though :)

 I've never heard of this before, too. However, the source code can be
 found at <http://cr.yp.to/djbfft.html>

 It is Copyright 1999, D. J. Bernstein but does not contain a license. I
 doubt that it is suitable (or even useful) for Debian.

DJB has since placed all his future and past software in the public
domain. Perhaps ping him if you need clarification of this.



--
Dipl.-Phys. Fabian Greffrath

Ruhr-Universität Bochum
Lehrstuhl für Energieanlagen und Energieprozesstechnik (LEAT)
Universitätsstr. 150, IB 3/134
D-44780 Bochum

Telefon: +49 (0)234 / 32-26334
Fax:     +49 (0)234 / 32-14227
E-Mail:  greffrath@leat.ruhr-uni-bochum.de



Reply to: