Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site sdcrdcf.UUCP
Path: utzoo!watmath!clyde!bonnie!akgua!sdcsvax!sdcrdcf!pmontgom
From: pmontgom@sdcrdcf.UUCP (Peter Montgomery)
Newsgroups: net.math
Subject: Re: random bignums
Message-ID: <1839@sdcrdcf.UUCP>
Date: Sat, 16-Mar-85 21:48:48 EST
Article-I.D.: sdcrdcf.1839
Posted: Sat Mar 16 21:48:48 1985
Date-Received: Mon, 18-Mar-85 07:38:33 EST
References: <2630@mcnc.UUCP>
Reply-To: pmontgom@sdcrdcf.UUCP (Peter Montgomery)
Organization: System Development Corp. R+D, Santa Monica
Lines: 24
Summary: 

In article <2630@mcnc.UUCP> version B 2.10.2 9/18/84; site sdcrdcf.UUCP version B 2.10.1 6/24/83; site mcnc.UUCP sdcrdcf!hplabs!tektronix!uw-beaver!cornell!vax135!houxm!mhuxj!mhuxr!ulysses!unc!mcnc!bts bts@mcnc.UUCP (Bruce T. Smith) writes:
>Does anyone have suggestions for generating random bignums?
>Currently, we're using several calls to random-- shifting
>and adding each time-- to generate a number bigger than we
>need.  Then we mod this by the desired upper bound.
>_____________________________
>Bruce T. Smith, MCNC           Microelectronics Center of N.C.
>decvax!mcnc!bts      (USENET)  P.O. Box 12889, 3021 Cornwallis Rd.
>bts.mcnc@CSnet-Relay (others)  Research Triangle Park, NC 27709

Suppose we need to generate a random number between 0 and 123456789, and
our multi-precision routines work in base 1000.  I generate the bottom
three digits by taking one random integer between 0 and 999 inclusive, the
middle three digits likewise, and the upper three digits by taking a random
integer from 0 to 123.  The result will be a random number between 0 and
123999999.  If the result exceeds 123456789, then I reject it and start
afresh.  Otherwise I use this as my answer.
-- 
			Peter Montgomery

	{aero,allegra,bmcg,burdvax,hplabs,
	 ihnp4,psivax,randvax,sdcsvax,trwrb}!sdcrdcf!pmontgom

Don't blame me for the crowded freeways - I don't drive.