On Thu, 2 Dec 2004, Lars V. Nielsen (HVM) wrote:

> Eh? If you pick "any" point, you'll end up with different shapes, depending
> on how your points are distributed.

Well, well... Right you are! This method doesn't result in a unique
solution. But I'm not sure that there *is* a unique solution for a 
"concave polygon" made from a distribution of points. But it does give you 
a polygon that includes all the points.

However, there are choices in some situations you can make that create a
pathological polygon (one where more than two edges intersect a single
point.)  So let me retreat from my earlier sweeping claim about picking
"any" point and say instead that you pick for the starting point the one
with the smallest Y-value, and if there are multiple choices, then refine
it by choosing from those the one with the smallest X value.

- Bill Thoen



---------------------------------------------------------------------
List hosting provided by Directions Magazine | www.directionsmag.com |
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Message number: 14342

Reply via email to