From owner-sc22wg5@open-std.org Tue Nov 1 14:36:57 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 5C1D5149B5; Tue, 1 Nov 2005 14:36:57 +0100 (CET) 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 D816811263 for ; Tue, 1 Nov 2005 14:36:50 +0100 (CET) Received: from mx1.liv.ac.uk (mx1.liv.ac.uk [138.253.100.179]) by dkuug.dk (8.12.10/8.9.2) with ESMTP id jA1DaUZb076639 for ; Tue, 1 Nov 2005 14:36:41 +0100 (CET) (envelope-from j.l.schonfelder@liverpool.ac.uk) Received: from mailhub5.liv.ac.uk ([138.253.100.157]) by mx1.liv.ac.uk with esmtp (Exim 4.52) id 1EWurj-0005ly-Ss for sc22wg5@dkuug.dk; Tue, 01 Nov 2005 12:03:47 +0000 Received: from localhost ([127.0.0.1] helo=mailhub5.liv.ac.uk) by mailhub5.liv.ac.uk with esmtp (Exim 4.52) id 1EWurj-0005TT-Pr for sc22wg5@dkuug.dk; Tue, 01 Nov 2005 12:03:47 +0000 Received: from host217-42-70-85.range217-42.btcentralplus.com ([217.42.70.85] helo=[192.168.0.2]) by mailhub5.liv.ac.uk with esmtpsa (TLSv1:DES-CBC3-SHA:168) (Exim 4.52) id 1EWurj-0005TF-J1 for sc22wg5@dkuug.dk; Tue, 01 Nov 2005 12:03:47 +0000 Date: Tue, 01 Nov 2005 12:03:47 +0000 From: "J.L.Schonfelder" To: WG5 Subject: Re: (j3.2005) (SC22WG5.3327) Re: [ukfortran] SC22 meeting Message-ID: <4860AD579F0E6E5DFF820D8F@[192.168.0.2]> In-Reply-To: <20051101113255.E5F4611583@open-std.org> References: <20051101113255.E5F4611583@open-std.org> X-Mailer: Mulberry/3.1.6 (Win32) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline X-Spam-Score: 0 () Sender: owner-sc22wg5@open-std.org Precedence: bulk For all the reasons stated I think support for decimal arithmetic is highly desirable. I remember Kahan being very keen on a packed (10bit=3D) decimal representation back in the mid 70s. His view then, and on such matters Valvo was usually right, that it would be possible to implement an arithmetic based on this representation in hardware that was only marginally less efficient than a binary system a la IEEE. Direct support for decimal in a revised IEEE and the major language specs would perhaps provide an incentive for the chip builders to produce the hardware. --On 01 November 2005 11:35 +0000 John Reid wrote: > > > -------- Original Message -------- > > From: "Ian Chivers" > To: , "'WG5'" > Cc: "'Stevens, Charles C'" > Subject: RE: [ukfortran] (SC22WG5.3324) SC22 meeting > Date: Tue, 1 Nov 2005 11:21:02 -0000 > MIME-Version: 1.0 > Content-Type: text/plain; > charset="us-ascii" > Content-Transfer-Encoding: 7bit > X-Mailer: Microsoft Office Outlook, Build 11.0.5510 > Thread-Index: AcXa3mJ+sl5ajZGtSumAaPTGG03FnAD9Iy3g > In-Reply-To: <20051027095851.711BA1152F@open-std.org> > X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2527 > X-Spam-Score: 1.015 (*) MISSING_OUTLOOK_NAME,MSGID_FROM_MTA_HEADER > > >> The other resolutions that affect us are >> >> 1. Resolution 05-08: Revision of IEEE 754 Standard for Floating-Point >> Arithmetic >> > > Stuff deleted > > It has also been proposed that decimal arithmetic be removed altogether. I > think that would be a shame. We all use decimal arithmetic for our > thinking > and hand calculations. It would be great if our computers did, too - for > example, no rounding would be needed for formatted output. Does anyone > disagree with this position? > > Of course, the Cobol people strongly favour retention of decimal > arithmetic > in the new standard. > > Cheers, > > John. > > _______________________________________________ > ukfortran mailing list > http://www.accu.org/mailman/listinfo/ukfortran > > I would not like to see decimal arithmetic removed totally. > I do not have strong views however about the packing method. > > SQL has exact decimal and has had this for a long time. > Microsoft > Visual Basic has decimal data type > C# has system.decimal. This is the same as the vb one. > Microsoft Java has access to the same decimal type. > The ISO JTC1/SC22/WG14 C and WG21 C++ committees are preparing > Technical > Reports which add the decimal types to those languages > Java (through the Sun or IBM BigDecimal class) > Decimal arithmetic packages are available for Eiffel and Python > > A good coverage is at > > http://www2.hursley.ibm.com/decimal/decifaq1.html > > And is where I located some of the above information. > > I think that is important that Fortran be able to address > Problems in this area. > -- Dr. Lawrie Schonfelder Honorary Senior Fellow, University of Liverpool Home: 1 Marine Park, West Kirby, Wirral, UK, CH48 5HN Phone: +44 (151) 625 6986