Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: $Revision: 1.6.2.16 $; site ima.UUCP
Path: utzoo!watmath!clyde!cbosgd!ukma!psuvm.bitnet!psuvax1!burdvax!sdcrdcf!ucla-cs!ucbvax!decvax!cca!ima!johnl
From: johnl@ima.UUCP
Newsgroups: net.arch
Subject: Re: ECLIPSE Floating Point Format
Message-ID: <100000004@ima.UUCP>
Date: Wed, 11-Dec-85 12:58:00 EST
Article-I.D.: ima.100000004
Posted: Wed Dec 11 12:58:00 1985
Date-Received: Sat, 14-Dec-85 03:35:39 EST
References: <973@loral.UUCP>
Lines: 14
Nf-ID: #R:loral:-97300:ima:100000004:000:701
Nf-From: ima!johnl    Dec 11 12:58:00 1985


/* Written  6:41 pm  Dec  9, 1985 by miller@loral in ima:net.micro */
> I need some information on the floating point format of the Data General
> ECLIPSE type 32-bit (MV-10000) Computers.  I tried contacting the local
> Data General office but all they seemed interested in was a P.O.  so they
> could answer my question.

Your local DG office must be woefully underinformed.  DG has always used
exactly the same floating point format as the IBM 360/370 series, that is
1-bit sign, 7-bit exponent, and 6 (float) or 14 (double) hex digits of
fraction.  The exponent is excess-64, and is powers of 16 not of 2.  The 360
floating point format is well-known if not well-liked.

John Levine, ima!johnl