Path: utzoo!attcan!uunet!tut.cis.ohio-state.edu!ucbvax!agate!linus!linus!bs From: bs@linus.mitre.org (Robert D. Silverman) Newsgroups: comp.arch Subject: Re: benchmark for evaluating extended precision Keywords: extended precision,multiply,benchmark,arithmetic Message-ID: <120093@linus.mitre.org> Date: 14 Sep 90 12:35:16 GMT References: <3989@bingvaxu.cc.binghamton.edu> <119858@linus.mitre.org> <2857@canisius.UUCP> Reply-To: bs@gauss.UUCP (Robert D. Silverman) Organization: The MITRE Corporation, Bedford MA Lines: 24 In article <2857@canisius.UUCP> pavlov@canisius.UUCP (Greg Pavlov) writes: :In article <119858@linus.mitre.org>, bs@linus.mitre.org (Robert D. Silverman) writes: :> ......... Suppose A > B and A > C. How would one compute :> BC/A EXACTLY without 64 bit support? ..... :> : ... and if A = PI, and B = sq.rt. of 2, how would one calculate BC/A : EXACTLY ? : :> "You can lead a horse's ass to knowledge, but you can't make him think" : : ... obviously, we are all (all the rest of us, anyway) horses' asses.... : : greg pavlov, fstrf, amherst, ny If you had been following this thread you would know that we have been discussing INTEGER arithmetic throughout. Therefore BC/A is an integer that fits within a single machine word. -- Bob Silverman #include Mitre Corporation, Bedford, MA 01730 "You can lead a horse's ass to knowledge, but you can't make him think"