Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!csd4.milw.wisc.edu!cs.utexas.edu!ut-emx!bill From: bill@ut-emx.UUCP (Bill Jefferys) Newsgroups: comp.sys.mac Subject: Re: Bug in EXCEL v. 2.2, and a flame Message-ID: <16497@ut-emx.UUCP> Date: 2 Aug 89 19:25:58 GMT References: <959@rex.cs.tulane.edu> <3531@mentor.cc.purdue.edu> Reply-To: bill@emx.UUCP (Bill Jefferys) Distribution: usa Organization: UTexas Computation Center, Austin, Texas Lines: 13 In article <3531@mentor.cc.purdue.edu> ags@mentor.cc.purdue.edu (Dave Seaman) writes: #First of all, a simple Taylor series expansion of f(x) = sqrt(x) about x=1 #should show you that the square root of .999975 is .9999875 with an error on #the order of 10^(-10). But since it was claimed that Excel 2.2 on a Mac II #did not agree, I decided to check it out. This is NOT what I claimed. I said that the result on the Mac II is correct (when you widen the field to 16 digits). Another poster showed that on the SE/30 the answer is correct. APPARENTLY THE BUG ONLY SHOWS UP ON A MACHINE WITHOUT THE COPROCESSOR. IT IS DEFINITELY THERE ON A MAC+. Bill Jefferys