Article View: net.games.emp
Article #50rand problems
From: utzoodecvaxharpoeaglemhtsaihnssi...
Date: Tue, 11 May 1982 16:58
Date: Tue, 11 May 1982 16:58
15 lines
923 bytes
923 bytes
Looking at the C source, rand() returns an shift right of the iterated random variable so its low order digits should be fine. Try digging up the source on your system, its one line of code. The only bias to taking rand() mod a value is that the values 0 through 0x7fff mod modvalue are slightly more likely to occur than the values (0x7fff mod modvalue) plus one through modvalue minus one. In the case of an even divisor of 0x10000 ( this includes 8 ) there should be no bias at all. rand() gets blamed for all kinds of things, I think you should check this out further. Lew Mammel, Jr. - BTL Indian Hill
Technical Details
Message-ID:
<bnews.ihuxv.121>Path:
arc1-ca.pugleaf.net!unknown!not-for-mailShow Full Headers
Complete Article Headers
From: utzoo!decvax!harpo!eagle!mhtsa!ihnss!ihuxv!lew Newsgroups: net.games.emp Title: rand problems Article-I.D.: ihuxv.121 Posted: Tue May 11 16:58:21 1982 Received: Wed May 12 03:30:35 1982 Subject: rand problems Date: Tue, 11 May 1982 16:58:21 +0000 Message-ID: <bnews.ihuxv.121>