Jamis Buck wrote:
> 
> NURBS stands for Non-Uniform Rational Bezier Splines, right?  How are
> these different from the implementation of splines provided with
> Skarda's patch?

It is NonUniform Rational B-Spline, a somewhat different critter.  The
advantage and problem lie largely in the "nonuniform" part.  Advantage
because it improves accuracy and allows you e.g. to represent an
accurate circular shape, which Bezier splines simply cannot do, and
gives you generally more flexibilith and fluidity because the "knots"
are not necessarily uniformly spaced.  The drawback is that this same
nonuniformity prevents the simple matrix representation used for Bezier,
Catmull-Rom etc. splines and requires a more sophisticated (probably
slower, harder to code) method of calculating surface points.  See Foley
and vanDam for more info.  This is a very limited "close enough for
government work" once-over-lightly, but it should convey the idea.

Jerry Anning
cle### [at] dholcom

Post a reply to this message