help-octave
[Top][All Lists]
Advanced

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

Re: MacOSX: test_system.m in test dir


From: David Bateman
Subject: Re: MacOSX: test_system.m in test dir
Date: Wed, 21 Feb 2007 10:03:03 +0100
User-agent: Thunderbird 1.5.0.7 (X11/20060921)

Matthias Brennwald wrote:
> Here's the results from my G4 (PowerBook), with a little glitch on  
> Octave 2.1.71:
>
> 1. With Octave 2.1.71 (binary from Fink):
>
> octave:1> rand('state',20)
> error: rand: unrecognized string argument
> octave:1> rand(6)
> ans =
>
>    0.9261073  0.3423863  0.4648016  0.1680272  0.8106666  0.8564630
> <snip>
>   
Ok, these results aren't the same, but that isn't surprising as it
appears that you aren't using the octave-forge generators for 2.1.71 but
the original octave generators..

>
> 2. Octave 2.9.9 (binary from HPC):
>
> octave:1> rand('state',20)
> octave:2> rand(6)
> ans =
>
>     0.905640   0.904946   0.103248   0.908636   0.081466   0.283472
>    <snip>
These results however are the same and so we now have at least three
data points, Linux/32bit/Intel, MacOSX/32bit/PowerPC and
MacOSX/??bit/Intel. So it seems that the main generators give stable
results across platforms. We can therefore as Paul suggested test the
generators with fixed known sequences and keep the statistical tests for
special occasions.. Is this the preferred solution?

D.

-- 
David Bateman                                address@hidden
Motorola Labs - Paris                        +33 1 69 35 48 04 (Ph) 
Parc Les Algorithmes, Commune de St Aubin    +33 6 72 01 06 33 (Mob) 
91193 Gif-Sur-Yvette FRANCE                  +33 1 69 35 77 01 (Fax) 

The information contained in this communication has been classified as: 

[x] General Business Information 
[ ] Motorola Internal Use Only 
[ ] Motorola Confidential Proprietary



reply via email to

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