Newsgroups: sci.math.num-analysis,sci.image.processing,comp.infosystems.gis
Path: cantaloupe.srv.cs.cmu.edu!europa.chnt.gtegsc.com!howland.reston.ans.net!news.nic.surfnet.nl!sun4nl!cwi.nl!dik
From: dik@cwi.nl (Dik T. Winter)
Subject: Re: Non-IEEE Floating point
Message-ID: <DAsvJt.KG7@cwi.nl>
Sender: news@cwi.nl (The Daily Dross)
Nntp-Posting-Host: boring.cwi.nl
Organization: CWI, Amsterdam
References: <szeliga.803859767@fluke> <3sl17u$f2c@sol.sun.csd.unb.ca>
Date: Mon, 26 Jun 1995 21:55:05 GMT
Lines: 18
Xref: glinda.oz.cs.cmu.edu sci.math.num-analysis:21709 sci.image.processing:15507 comp.infosystems.gis:36044

In article <3sl17u$f2c@sol.sun.csd.unb.ca> MLeese@atlantic.cs.unb.ca writes:
 > On Thu, 22 Jun 1995 22:22:47 GMT Tim Szeliga - NWS (szeliga@fluke.forestry.umn.edu) wrote:
 > >> I have satellite data that has all the floating point data 
 > >> formatted with  (1/7/24) one sign bit, seven exponent bits and 24 mantissa
 > >> bits, instead of the IEEE standard of 1/8/23.  
 > >> The standard C floating point routines and octal dump won't read the data.
 > >> (The GOES GVAR data uses the Gould/SEL FP representation.)
 > 
 > I don't know if this helps, but the Gould/SEL FP format is the same
 > as that used on the IBM 360 mainframe.
 > 
For positive numbers; not for negative numbers.  IBM uses sign-magnitude
while Gould uses 2's complement.  However it is fairly easy to convert
2's complement to sign-magnitude.  And there are routines floating around
to convert IBM format to IEEE format.
-- 
dik t. winter, cwi, kruislaan 413, 1098 sj  amsterdam, nederland, +31205924098
home: bovenover 215, 1025 jn  amsterdam, nederland; e-mail: dik@cwi.nl
