Strange Degree 2 surface behaviour


#1

Hi All,

This one has bitten me before but it was a long time ago so I forgot about it and made a mess of a job.

To replicate:
Draw a sphere of say 1500mm radius
Offset sphere inwards by say 350mm
In Properties, increase isocurve density to 4

The U isocurves are now unevenly distributed. You may have to ArrayPolar a line to see it, because it’s not immediately obvious (see attached).

Can anyone please explain WHY this happens? I would make the (obviously wrong) assumption that it should remain symmetrical.

No panic on this as I will be unlikely to screw this up again, but I’d just like to know what’s going on.

Thanks, Steve


(Pascal Golay) #2

Hi Steve - just a guess, but since it looks like the actual knot lines and the mid-spans are true, then the discrepancy may be from the un-even parameterization of rational surfaces.The UV accelerates and decelerates on those - the ‘corner’ control points are weighted at less than one, ~.707, and so the isos are not quite in the middle going from the knot lines (quads) to the mid-spans.

-Pascal


#3

Thanks Pascal,

Yep that makes sense. I’ll remember to keep an eye on this one so I don’t get in trouble again.

Cheers, Steve