Path: utzoo!mnetor!uunet!seismo!sundc!pitstop!sun!decwrl!decvax!ucbvax!AERO4.LARC.NASA.GOV!blbates
From: blbates@AERO4.LARC.NASA.GOV (Bates TAD/HRNAB ms294 x2601)
Newsgroups: comp.sys.sgi
Subject: Re:  IRIS floating point read error
Message-ID: <8802261626.AA03615@aero4.larc.nasa.gov>
Date: 26 Feb 88 16:26:36 GMT
Sender: usenet@ucbvax.BERKELEY.EDU
Organization: The Internet
Lines: 4

You shouldn't have to add the .5 to your sample case, it should work as is.
I have had similar problems with programs.  A set of subroutines work fine
everyplace, but on our IRIS.  I avoid doing ANY calculations on our IRIS
that require any accuracy.