> Since his polygons are convex, surely a more efficient solution would
> be to use the log n method.
> 

Dan said his polygons are convex, but did he really mean that?  Convex 
polygons with lots of vertices would come about only from some sort of 
automatic approximation to a curve, methinks.

A J implementation of the O(log n) method shouldn't be hard & I'll make 
a try at it if Dan confirms that he really does have many-sided convex 
polygons.

Henry Rich
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to