Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!helios!bcm!rice!hsdndev!wuarchive!sdd.hp.com!hplabs!hpfcso!hplvec!kwb From: kwb@hplvec.LVLD.HP.COM (Keith Blackwell) Newsgroups: comp.graphics Subject: Re: Finding nearest point Message-ID: <640004@hplvec.LVLD.HP.COM> Date: 1 Mar 91 00:37:23 GMT References: Organization: Hewlett-Packard Co., Loveland, CO Lines: 19 | This is correct for the 2-dimensional case, and almost so for the | 3-dimensional case. The general answer is that for a D-dimensional minimum | distance search, step 3 should be: Thanks for the correction to my correction. I should have pointed out that I didn't even attempt to expand my analysis to 3 or more dimensions since my interest in this whole issue was a two-dimensional application I am currently dealing with. Funny thing, though, my first thought upon reading the original post was "should'nt it be cube-root instead of square-root?", which of course was also wrong, but it was based on the right intuitive hunch. When I got around to analyzing the thing, though, I found the problem was (at least) in something else. Math usually just isn't so intuitive anyway :-). | David Seal | dseal@acorn.co.uk -- Keith Blackwell (my employer has nothing to do with this)