Re: Monte Carlo in GSL

[Fabrice Rossi <Fabrice dot Rossi at apiacoa dot org>]

Przemyslaw Sliwa wrote:
> Dear All,
> [...]
> My problem is: I want that the random numbers I generate are purelly
> random and not that the series of the numbers repeats.

Hi.

Do you have any theoretical reason to ask for "real" random numbers?

I'm not an expert in this field, but all I've read about simulation make 
me feel that there is no reason to use true random numbers. In some 
situations, it is important to have a very long period, to avoid 
correlation between successive values, etc. All of this can be done is 
pseudo-random numbers, at a price of course (generating good random 
numbers cost much more than using basic generators) and very high 
quality algorithms are included in GSL.

Fabrice