Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: Notesfiles $Revision: 1.7.0.10 $; site ccvaxa
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!ihnp4!inuxc!pur-ee!uiucdcs!ccvaxa!aglew
From: aglew@ccvaxa.UUCP
Newsgroups: net.arch
Subject: Re: Representations of Real Numbers
Message-ID: <5100006@ccvaxa>
Date: Wed, 12-Feb-86 00:02:00 EST
Article-I.D.: ccvaxa.5100006
Posted: Wed Feb 12 00:02:00 1986
Date-Received: Fri, 14-Feb-86 01:14:09 EST
References: <931@houxa.UUCP>
Lines: 11
Nf-ID: #R:houxa.UUCP:931:ccvaxa:5100006:000:617
Nf-From: ccvaxa.UUCP!aglew    Feb 11 23:02:00 1986


I saw a paper in the not-too-distant past about a real number representation
that had three fields (four if you count the sign): fraction, exponent, and 
meta-exponent. I am not sure that I remember exactly what the meta-exponent 
was - it may have been something like the number of tens in 10**10**10**...
exponent - but it was used to greatly increase range. I believe that you
could express the loss in precision of large numbers as a fraction of the log
of the number?

Anybody remember this paper? Unfortunately, I have changed cities, and I 
remember where to find this paper only by the position in the stacks.