home *** CD-ROM | disk | FTP | other *** search
- Path: alterdial.uu.net!not-for-mail
- From: rex@aussie.com
- Newsgroups: comp.std.c
- Subject: Re: int's and zero
- Date: 9 Jan 1996 18:09:46 GMT
- Message-ID: <4cub1a$jbl@alterdial.UU.NET>
- References: <4cth4e$4q@odin.funcom.no>
- NNTP-Posting-Host: pool041.max1.ffx1.va.dynip.alter.net
- X-Newsreader: SPRY News 3.03 (SPRY, Inc.)
-
- > eivind@odin.funcom.com (Eivind Eklund) writes:
- > Are zero in an integer required to be represented by binary zeros?
-
-
- There was considerable debate about whether or not integers could be stored
- as BCD, Grey code, or some such, but in the end, we decided that they had to
- be stored as binary numbers. And, of course, 2's comp, 1's comp, and signed
- magnitude all use all-bits-zero for zero. The only thing that could cause a
- problem is for machines/compilers with signed zeros; I think that negative
- zero might have some other bit pattern.
-
- Rex
-
- ----------------------------------------------------------------------------
- +1 703 860-0091 | Rex Jaeschke | C and C++ Seminars
- FAX +1 703 860-3008 | 2051 Swans Neck Way | and Consulting
- rex@aussie.com | Reston, Virginia 22091, USA |
- ----------------------------------------------------------------------------
- Chair of X3J11 (ANSI C) and participant in SC22/WG14 (ISO C). Columnist for
- NT Developer, VC++ Professional, and Enterprise NT.
- ----------------------------------------------------------------------------
-
-