Path: utzoo!mnetor!uunet!lll-winken!lll-lcc!ames!mailrus!umix!umich!mibte!gamma!sabre!sword!faline!bellcore!spectral!sjs
From: sjs@spectral.ctt.bellcore.com (Stan Switzer)
Newsgroups: comp.lang.c
Subject: Re: Why NULL is 0
Message-ID: <6594@bellcore.bellcore.com>
Date: 7 Apr 88 12:52:43 GMT
References: <2550049@hpisod2.HP.COM> <7412@brl-smoke.ARPA> <3351@chinet.UUCP> <10229@steinmetz.steinmetz.ge.com> <1323@PT.CS.CMU.EDU>
Sender: news@bellcore.bellcore.com
Reply-To: sjs@spectral.UUCP (Stan Switzer)
Organization: Computer Technology Transfer, Bellcore
Lines: 17

In reference to:
> > People (usually ;>) have no problem with the idea the while assigning a
> > zero to a float gives it the vaule 0.0, in most implementations the
> > float value does not have all bits set to zero.

And:
>   I do believe IEEE and Vax's version of float is a bit pattern
> consisting of all zeros.  So what are these implementations that don't
> use zero bit pattern?

The Honeywell 6000 Series and GE 625/635 machines represented NORMALIZED
float zero as o400000000000 (36 bits, msb ON).  This may date back to its
grand-uncle the 709.

BTW, all 0 bits WAS a float zero, but it wasn't normalized.  The results of
most float operations on unnormalized values was undefined.  I suspect that
additions and subtractions would lose precision in the process
of alligning the points.