[Top][All Lists]

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

Re: Matrix benchmark

From: Dmitri A. Sergatskov
Subject: Re: Matrix benchmark
Date: Mon, 28 Jul 2003 13:35:19 -0600
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030624 Netscape/7.1

John W. Eaton wrote:
On 28-Jul-2003, address@hidden <address@hidden> wrote:

| octave-2.1.50:35> A = rand(2048,2048);
| octave-2.1.50:36> tic; B=fft(A); toc
| ans = 0.84838
| | >> A = rand(2048,2048); | >> tic; B=fft(A); toc | elapsed_time =
|     0.5116

Me too ( 0.7 / 0.5 on atlon mp 2200+)

Even if you use FFTW with Octave, there is no guarantee that it is
precisely the same as the version used by Matlab.

I think one of the differences is the version of the compiler used.
In my experience, on x86 platform,  2.95.x does a better job than
any 3.[012].* (have not tried 3.3.x yet). For both fftw and atlas.




Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:
How to fund new projects:
Subscription information:

reply via email to

[Prev in Thread] Current Thread [Next in Thread]