Path: utzoo!mnetor!uunet!lll-winken!lll-lcc!ames!hao!noao!mcdsun!sunburn!gtx!edge!doug
From: doug@edge.UUCP (Doug Pardee)
Newsgroups: comp.arch
Subject: Re: conditional branches
Message-ID: <1042@edge.UUCP>
Date: 25 Feb 88 20:16:51 GMT
References: <191@telesoft.UUCP> <9782@tekecs.TEK.COM>
Organization: Edge Computer Corporation, Scottsdale, AZ
Lines: 13
Summary: Comparison tests

]I've seen this assertion a couple of times, but it ain't so.  You need
]to worry about carry when doing a subtract, but not when just doing a
]compare.  If both operands are positive, a simple bit-for-bit
]comparison is sufficient, analogous to the way that C programs compare
]strings.  When the operands can be signed, a little extra logic is
]required, but nothing as fancy as a full adder/subtractor.

Not only is bit-for-bit comparison sufficient, it should be considered
*necessary*.  Using subtraction for comparison only makes a mess of
things by introducing (unnecessarily) the possibility of overflow when
comparing a quite negative number with a quite positive one.
-- 
Doug Pardee         {ames,hplabs,sun,amdahl,ihnp4,allegra}!oliveb!edge!doug
Edge Computer Corp., Scottsdale, AZ                 uunet!ism780c!edge!doug