From owner-sc22wg5@open-std.org Mon Oct 10 21:38:58 2005 Return-Path: X-Original-To: sc22wg5-domo2 Delivered-To: sc22wg5-domo2@open-std.org Received: by open-std.org (Postfix, from userid 521) id 31D0119361; Mon, 10 Oct 2005 21:38:58 +0200 (CET DST) X-Original-To: sc22wg5@open-std.org Delivered-To: sc22wg5@open-std.org Received: from dkuug.dk (ptah.dkuug.dk [195.215.30.66]) by open-std.org (Postfix) with ESMTP id DB8BE1147C for ; Mon, 10 Oct 2005 21:38:54 +0200 (CET DST) Received: from mfbak.jpl.nasa.gov (mfbak.jpl.nasa.gov [137.78.160.179]) by dkuug.dk (8.12.10/8.9.2) with ESMTP id j9AJchbU074868 for ; Mon, 10 Oct 2005 21:38:48 +0200 (CEST) (envelope-from van.snyder@jpl.nasa.gov) Received: from nmta3.jpl.nasa.gov (nmta3.jpl.nasa.gov [137.78.160.108]) by mfbak.jpl.nasa.gov (Switch-3.1.7/Switch-3.1.7) with ESMTP id j9AJIBav017279 for ; Mon, 10 Oct 2005 12:18:11 -0700 Received: from xmta1.jpl.nasa.gov (xmta1.jpl.nasa.gov [137.78.160.144]) by nmta3.jpl.nasa.gov (Switch-3.1.7/Switch-3.1.7) with ESMTP id j9AJI2sH029353 for ; Mon, 10 Oct 2005 12:18:02 -0700 Received: from mls.jpl.nasa.gov (mls.jpl.nasa.gov [137.79.7.84]) by xmta1.jpl.nasa.gov (Switch-3.1.7/Switch-3.1.7) with ESMTP id j9AJI105011954 for ; Mon, 10 Oct 2005 12:18:01 -0700 Received: from math.jpl.nasa.gov (math.jpl.nasa.gov [137.79.7.57]) by mls.jpl.nasa.gov (8.12.10+Sun/8.12.9) with ESMTP id j9AJGkjo000190 for ; Mon, 10 Oct 2005 12:16:46 -0700 (PDT) Received: from math.jpl.nasa.gov (localhost.localdomain [127.0.0.1]) by math.jpl.nasa.gov (8.13.1/8.12.11) with ESMTP id j9AJI1a4006239 for ; Mon, 10 Oct 2005 12:18:01 -0700 Received: (from vsnyder@localhost) by math.jpl.nasa.gov (8.13.1/8.12.11/Submit) id j9AJI1ua006238 for sc22wg5@dkuug.dk; Mon, 10 Oct 2005 12:18:01 -0700 X-Authentication-Warning: math.jpl.nasa.gov: vsnyder set sender to van.snyder@jpl.nasa.gov using -f Subject: Re: (j3.2005) (SC22WG5.3321) SC22 meeting From: Van Snyder Reply-To: van.snyder@jpl.nasa.gov To: sc22wg5@dkuug.dk In-Reply-To: <20051010132555.2FFA116941@open-std.org> References: <20051007143249.EA5CB13E6D@open-std.org> <20051008145352.35B9912FB2@open-std.org> <20051010132555.2FFA116941@open-std.org> Content-Type: text/plain Content-Transfer-Encoding: 7bit Date: Mon, 10 Oct 2005 12:18:00 -0700 Message-Id: <1128971880.5653.25.camel@math.jpl.nasa.gov> Mime-Version: 1.0 X-Mailer: Evolution 2.0.2 (2.0.2-22) X-Source-IP: mls.jpl.nasa.gov [137.79.7.84] X-Source-Sender: van.snyder@jpl.nasa.gov X-AUTH: Internal IP X-Spam-Score: 0 () Sender: owner-sc22wg5@open-std.org Precedence: bulk On Mon, 2005-10-10 at 09:18 -0400, Aleksandar Donev wrote: > J.L.Schonfelder wrote: > > >> I said that we had no strong views on this and would probably > >> support whatever the standard finally says. Does anyone disagree? > > The probable solution would be to operate with a base 10 > > normalisation but to use a representation that effectively packed 3D > > into each 10 bits. > Thanks Lawrie for the edification on the issue. I however still agree > with John that the Fortran's committee does not have a vested interest > in this and can go with whatever the external standard's says. I can > understand that Cobol people would care, but I don't think there is > vested Fortran interest eitherway. Wearing my math software hat, I prefer that the exponent is a base-10 exponent rather than a base-1000 exponent, for exactly the reasons Lawrie outlined. I don't care what the radix or representation of the mantissa is, but a 10-bit-digit base-1000 representation is attractive for exactly the reasons Lawrie outlined. If the mantissa is base 1000 and the exponent is base 10, the ALU probably has to convert between them, for correct normalization as Lawrie outlined (i.e., the leading digit of the mantissa, in base 1000, should not be less than 100), but that's "simply a matter of more transistors" and, unfortunately, probably more gate delays. I read a paper, probably in IBM System Journal, probably late '60's or early '70's, about a 10-bit base-1000 ALU, so this is not a radically new topic. -- Van Snyder | What fraction of Americans believe Van.Snyder@jpl.nasa.gov | Wrestling is real and NASA is fake? Any alleged opinions are my own and have not been approved or disapproved by JPL, CalTech, NASA, Frederick Gregory, George Bush, or anybody else.