From WSEYMOUR@email.usps.gov Tue Mar 3 13:07:38 1998 Received: from gk-east.usps.gov (gk-east.usps.gov [198.116.37.10]) by dkuug.dk (8.6.12/8.6.12) with ESMTP id NAA05181 for ; Tue, 3 Mar 1998 13:07:23 +0100 Received: from mailgatee.usps.gov (mailgatee.usps.gov [56.80.0.21]) by gk-east.usps.gov (8.8.8/8.8.8) with ESMTP id HAA11389; Tue, 3 Mar 1998 07:07:15 -0500 (EST) Received: from EMAIL.USPS.GOV (email.usps.gov [56.88.9.237]) by mailgatee.usps.gov (8.8.8/8.8.8) with SMTP id HAA17444; Tue, 3 Mar 1998 07:05:41 -0500 (EST) Received: by EMAIL.USPS.GOV (Soft-Switch LMS 2.0) with snapi via SSWHUB id 0033000001952173; Tue, 3 Mar 1998 07:05:17 -0500 From: WILLIAM SEYMOUR To: " - (052)sc22wg5(a)dkuug.dk" Cc: " - (052)tydeman(a)tybor.com" , " - (052)frank(a)farance.com" Subject: Re[2]: ULP definition Message-ID: <0033000001952173000002L032*@MHS> Date: Tue, 3 Mar 1998 07:05:17 -0500 MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable It would seem that no one has found ULP defined in any other standard, and I can verify that it's not in ISO/IEC TR 12382:1992(E). I suggest that we define the term in C9X. --Bill Seymour ______________________________ Reply Separator ________________________= _________ Subject: Re: ULP definition Author: WILLIAM A SEYMOUR at SLMO002L Date: 2/23/98 9:38 AM I've been asked as J11's V.R. whether we need to define "ULP," meaning "Unit in the Last Place", in the C standard; and I confess that I don't know the answer. (We're using the term in words that deal with floating-point accuracy.) Is this already defined in the Fortran standard? In IEC 559? (It's not in either the current or the draft ANSI dictionary.) Thanks, --Bill Seymour Programmer/Analyst, U.S. Postal Service (wseymour@email.usps.gov) Vocabulary Representative, J11 (bsey@pobox.com) ______________________________ Reply Separator ________________________= _________ Subject: ULP definition Author: tydeman@tybor.com at INTERNET Date: 2/21/98 12:55 PM Bill, As the C vocabulary rep, do any of the relavent standards define the term ULP (Unit in the Last Place)? The review committee for DR 63 (Floating-Point Accuracy) is trying to decide if we need to define ULP in the C standard, as a footnote in the C standard, or in the rationale. --- Fred J. Tydeman +1 (512) 255-8696 Tydeman Consulting 3711 Del Robles tydeman@tybor.com Programming, testing, numerics Austin, Texas 78727 Voting member of J11 (ANSI "C") USA Sample C9X+FPCE tests: ftp://jump.net/pub/tybor/ =