Path: utzoo!attcan!uunet!mitel!sce!cognos!rayt
From: rayt@cognos.uucp (R.)
Newsgroups: comp.lang.c
Subject: Re: Yet another pseudo-random number generator
Message-ID: <5313@cognos.UUCP>
Date: 17 Feb 89 18:37:53 GMT
References: <7383@batcomputer.tn.cornell.edu> <5269@cognos.UUCP> <9653@smoke.BRL.MIL>
Reply-To: rayt@cognos.UUCP (R.)
Organization: Cognos Inc., Ottawa, Canada
Lines: 18

In article <9653@smoke.BRL.MIL> gwyn@brl.arpa (Doug Gwyn) writes
in response to my mention of the recent CACM article on random
generators, that:

"Virtually all PRNG schemes in common use have provable
 statistical behavior.  That doesn't mean they're always
 suitable.  In fact I'm not enthised about that CACM article."

Are you refering to the need to have skewed or otherwise non-normal
distributions? Also I'm interested in knowing what you found
wanting in the CACM article.

					Ray Tigg
-- 
Ray Tigg                          |  Cognos Incorporated
                                  |  P.O. Box 9707
(613) 738-1440 x5013              |  3755 Riverside Dr.
UUCP: uunet!mitel!sce!cognos!rayt |  Ottawa, Ontario CANADA K1G 3Z4