Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!zaphod.mps.ohio-state.edu!uakari.primate.wisc.edu!uflorida!mephisto!ncsuvx!mcnc!ecsgate!ecsvax!urjlew From: urjlew@uncecs.edu (Rostyk Lewyckyj) Newsgroups: comp.arch Subject: Re: F.P. vs. arbitrary-precision (was: Killer Micro II) Summary: Is the set of all sets a set? Message-ID: <1990Sep10.215549.26260@uncecs.edu> Date: 10 Sep 90 21:55:49 GMT References: <3755@osc.COM> <4513@taux01.nsc.com> <119244@linus.mitre.org> Organization: UNC Educational Computing Service Lines: 30 In article <119244@linus.mitre.org>, bs@linus.mitre.org (Robert D. Silverman) writes: > The four basic operations of arithmetic are +, -, x, /. Any computer that > can't perform them on its atomic data units [whatever the word size is] > is a joke. > > -- > Bob Silverman > #include > Mitre Corporation, Bedford, MA 01730 > "You can lead a horse's ass to knowledge, but you can't make him think" If the basic unit for representing a number is a word, then to handle the product of two numbers you may well need a double word. But then you must provide operations to handle double words. So then the double word becomes a basic unit, and so on ad infinitum. If in addition to +, -, and x, you want to handle /, then you most certainly need either something like floating point, or some variant of rational fraction representation with arbitrary precision for the numerator and denominator Now if you want to handle irrationals, then even more you need something like floating point. Choose your own precision. Finally scaledd integer is really no different from floating point, except that floating point is at least standardized (IEEE) , and assisted by hardware. Imagine having to work with user's programs where everybody is doing scaled integer arithmetic in his own way. ----------------------------------------------- Reply-To: Rostyslaw Jarema Lewyckyj urjlew@ecsvax.UUCP , urjlew@unc.bitnet or urjlew@uncvm1.acs.unc.edu (ARPA,SURA,NSF etc. internet) tel. (919)-962-6501