home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.std.c
- Path: in2.uu.net!jyacc!usenet
- From: jtrigg@marks.jyacc.com (Jim Trigg)
- Subject: Re: Help, best way to compare doubles
- In-Reply-To: tanmoy@qcd.lanl.gov's message of 15 Jan 1996 16:13:22 GMT
- X-Nntp-Posting-Host: marks
- Message-ID: <9livibg6p4.fsf@marks.jyacc.com>
- Sender: jtrigg@marks.jyacc.com
- Organization: JYACC, Inc.
- X-Newsreader: Gnus v5.1
- References: <4d1k09$aqq@mercury.IntNet.net> <4dbos6$o7q@umbc9.umbc.edu>
- <4ddev4$1vg@stingray.mcnc.org> <TANMOY.96Jan15091322@qcd.lanl.gov>
- Date: Tue, 16 Jan 1996 21:39:35 GMT
-
- In article <TANMOY.96Jan15091322@qcd.lanl.gov> tanmoy@qcd.lanl.gov
- (Tanmoy Bhattacharya) writes:
-
- > In article <4ddev4$1vg@stingray.mcnc.org> coats@mcnc.org (Carlie
- > Coats) writes:
- > <snip>
- > or (equvalently, but in decently-efficient C):
-
- > ...though undefined according to the ANSI C standard...
-
- > ( t = a - b )*t < esquared * ( a*a + b*b + delta )
-
- I believe a standard-compliant version would be:
-
- ( t = a - b , t * t) < esquared * ( a * a + b * b + delta )
-
- Just a Thought,
- Jim Trigg
-