From owner-sc22wg5@open-std.org Thu Oct 27 22:22:48 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 04D8B168D9; Thu, 27 Oct 2005 22:22:48 +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 0329C11593 for ; Thu, 27 Oct 2005 22:22:44 +0200 (CET DST) Received: from brmea-mail-4.sun.com (brmea-mail-4.Sun.COM [192.18.98.36]) by dkuug.dk (8.12.10/8.9.2) with ESMTP id j9RKMSZb064257 for ; Thu, 27 Oct 2005 22:22:31 +0200 (CEST) (envelope-from Keith.Bierman@Sun.COM) Received: from fe-amer-10.sun.com ([192.18.108.184]) by brmea-mail-4.sun.com (8.12.10/8.12.9) with ESMTP id j9RJjkD7016904 for ; Thu, 27 Oct 2005 13:45:47 -0600 (MDT) Received: from conversion-daemon.mail-amer.sun.com by mail-amer.sun.com (Sun Java System Messaging Server 6.1 HotFix 0.11 (built Jan 28 2005)) id <0IP100401ARO3Z00@mail-amer.sun.com> (original mail from Keith.Bierman@Sun.COM) for sc22wg5@dkuug.dk; Thu, 27 Oct 2005 13:45:46 -0600 (MDT) Received: from [192.168.100.108] ([129.150.30.70]) by mail-amer.sun.com (Sun Java System Messaging Server 6.1 HotFix 0.11 (built Jan 28 2005)) with ESMTPSA id <0IP1008BSAW9J677@mail-amer.sun.com>; Thu, 27 Oct 2005 13:45:46 -0600 (MDT) Date: Thu, 27 Oct 2005 12:45:44 -0700 From: Keith Bierman Subject: Re: (j3.2005) (SC22WG5.3325) SC22 meeting In-reply-to: <20051027193948.635481496E@open-std.org> To: longb@cray.com Cc: j.k.reid@rl.ac.uk, WG5 , "Stevens, Charles C" Message-id: <43612E68.9000301@sun.com> MIME-version: 1.0 Content-type: text/plain; format=flowed; charset=ISO-8859-1 Content-transfer-encoding: 7BIT X-Accept-Language: en-us, en References: <20051027095851.711BA1152F@open-std.org> <20051027193948.635481496E@open-std.org> User-Agent: Mozilla Thunderbird 1.0.6 (Macintosh/20050716) X-Spam-Score: 0 () Sender: owner-sc22wg5@open-std.org Precedence: bulk Bill Long wrote: > ... Cobol, and IBM. They will have to live with whatever they > decide. It seems outside the scope of WG5. > I agree with that. > In terms of the arguments for an against having decimal floating point > at all, I see the human comfort argument as of secondary importance. > If the hardware implementation of decimal floating point results in > faster execution than binary floating point, then decimal may be > adopted in practice. Otherwise, it will not. For Cobol programs, > where there is a lot of format conversion, decimal will probably be > faster. For Fortran, the issue is computation speed, and the case is > a lot less clear. I think speed is missing the point (I doubt COBOL is faster because they decimal. I think that folks writing programs like spreadsheet programs care a bit less about speed, and more about providing answers which match naive user expectations. Some such programs could be (and should be) written in Fortran if a decimal type was easy to use, and /fast enough/. Having this standardized is likely to be a GoodThing. Clearly, not everything will be computed that way, and anyone who replaced their binary format with a decimal format would probably not be very successful. But there is no reason why a vendor should do that; Fortran is already flexible enough to have more than one floating point format ;> -- Keith H. Bierman keith.bierman@Sun.COM| Sun Microsystems PAE | khbkhb@gmail.com 12 Network Circle UMPK 12-325 | 650-352-4432 voice+fax Menlo Park, California 94025 | sun internal 68207 Copyright 2005 ------------------------------------------------------------ NOTICE: This email message is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message.