From: Jens Maurer (Jens.Maurer_at_[hidden])
Date: 2001-12-04 15:45:44
nbecker_at_[hidden] wrote:
>
> I see that moving from 1.25.1 to 1.26.0 there has been some change to
> random that is causing the resulting random number sequences to be
> different. This was an unexpected surprise.
>
> I'm not sure yet which random number generator is the culprit.
I had to fix uniform_int<> because some internal computations could
possibly overflow, giving incorrecgt results.
I've added a note to the 1.26.0 release notes in the CVS.
I'm not sure whether all overflows are really fixed, though, so
an additional check would be most appreciated.
Jens Maurer