Date: Tue, 29 Sep 1998 12:02:18 -0400 (EDT)
Message-Id: <199809291602.MAA20294@indy.delorie.com>
From: DJ Delorie <dj AT delorie DOT com>
To: Kbwms AT aol DOT com
CC: djgpp-workers AT delorie DOT com
In-reply-to: <4ed1b56c.3611007e@aol.com> (Kbwms@aol.com)
Subject: Re: Proposed New Random
Precedence: bulk


> I might be able to make a contribution here.  Recently, I completed a
> paper on combined random number generators, "Quick and Portable Random
> Number Generators," with Dr. Gerald P. Dwyer, formerly with Clemson
> University and now at the Atlanta Fed.  The supporting random number
> generator is very well behaved and has a period of about 2^62.  This
> is certainly long enough to outlive the expected life of the universe.

rand() is a sufficiently good RNG with a long period.  The random()
and srandom() functions use a particular algorithm, and we'd need to
replicate *that* algorithm.