[Scilab-users] rand("seed",s)

[Stéphane Mottelet]

Le 29/04/2019 à 19:03, Jens Simon Strom a écrit :
>
> Hallo Scilab équipe,
>
> In my application it is vital   -  in addition to the normal focus 
> when using seeds -  that /different/ seeds in rand("seed",seed) 
> produce /different/ random number series. (1) Is that the case?
>
Yes. However with "rand" you have a periodicit of 2^31 (this is 
explained in the help page).

Please use "grand" instead of "rand", as grand produces sequences with 
better properties, moreover you have the choice of the basic generator. 
The default one "Mersenne-Twister" has period 2^199937. See the help 
page of grand for more details.

S.

> I am aware that  seed has to be a non negative integer. (2) Are there 
> any further restrictions?  (3) What is the upper bound?
>
> The minimal example below is as expected.
>
> Kind regards
>
> Jens
>
> s1=1e3;
> s2=1e10;
> rand("seed",s1);
> r1=rand(3,1);
> rand("seed",s2);
> r2=rand(3,1);
> //
> rand("seed",s1)
> r3=rand(3,1);
> rand("seed",s2);
> r4=rand(3,1);
> r1234=[r1  r2  r3  r4] ----------------------------------------------------
>
>
>
>
>
> _______________________________________________
> users mailing list
> users at lists.scilab.org
> https://antispam.utc.fr/proxy/1/c3RlcGhhbmUubW90dGVsZXRAdXRjLmZy/lists.scilab.org/mailman/listinfo/users


-- 
Stéphane Mottelet
Ingénieur de recherche
EA 4297 Transformations Intégrées de la Matière Renouvelable
Département Génie des Procédés Industriels
Sorbonne Universités - Université de Technologie de Compiègne
CS 60319, 60203 Compiègne cedex
Tel : +33(0)344234688
http://www.utc.fr/~mottelet

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20190430/f3d2207b/attachment.htm>