Just when I thought I was beginning to better understanding some of the inherit behaviour of NURBS, I wasn't really expecting to see the results of this experiment:

First draw a multi-point (>4) 'Through points' curve, then draw a 'Control points' curve by sequentially picking the start+intermediate+end control points of the first curve.

Below: 'Through points' curve (blue), 'Control points' curve (red). The control points of both are turned on and are coincident.


I don't understand why I would want - or need - the weightings of each curve type to be different. AFAIU, different weightings for NURBS DO exist, (much the same as Rho value alters a conic segment). This page has some interesting reading on the topic: http://www.pilot3d.com/NurbSecrets.htm

Any enlightening info appreciated.

Thanks,
Hamish