Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!rutgers!cbmvax!vu-vlsi!perry
From: perry@vu-vlsi.UUCP (Rick Perry)
Newsgroups: comp.arch
Subject: Re: random number generator in hardware
Message-ID: <420@vu-vlsi.UUCP>
Date: Sat, 15-Nov-86 15:55:13 EST
Article-I.D.: vu-vlsi.420
Posted: Sat Nov 15 15:55:13 1986
Date-Received: Sun, 16-Nov-86 01:42:38 EST
References: <317@zuring.mcvax.UUCP> <267@bath63.UUCP> <2490@phri.UUCP> <2500@phri.UUCP>
Reply-To: perry@vu-vlsi.UUCP (Rick Perry)
Organization: Villanova Univ. EE Dept.
Lines: 9

I think the idea of hardware random number generators is an excellent one.
It's all too easy for the average joe (who didn't do his doctoral thesis
on super-duper pseudo-random number generation) to implement or use exisiting
pseudo-random number generators that are not very good.  Other problems
with the pseudo methods (like the common linear congruential method) is that
the least significant bits might not be very random, certain operations
on the generated numbers (like using MOD's) might destroy the randomness,
and successive samples may be highly correlated.  It would be great to have
an easy to use, dependable, fast hardware generator of really random numbers.