Date: Sun, 4 Oct 1998 12:39:54 +0300 (IDT)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: Kbwms AT aol DOT com
cc: dj AT delorie DOT com, djgpp-workers AT delorie DOT com
Subject: Re: Proposed New Random
In-Reply-To: <9fc72b10.3614ebaf@aol.com>
Message-ID: <Pine.SUN.3.91.981004123928.1863N-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII


On Fri, 2 Oct 1998 Kbwms AT aol DOT com wrote:

> This indicates behavior that is decidedly non-random.  The tests of
> 'rand' were conducted on the lower bits of each number generated.
> Despite the fact that 'rand' returns the "middle 32-bits," Knuth
> points out on pages 12 and 14 that even these bits will suffer non-
> random behavior.  It is my experience that the upper bits will fare
> far better.

Thanks for testing this.  That's what I knew about lower bits produced
by LCG-type random generators.