Hi Bravlin, yup that is correct - with polygon booleans where there isn't any concept of an "underlying surface", things get fragmented into little bits pretty easily.

With NURBS modeling that doesn't happen or at least happens much less frequently and that's a major factor for why booleans with NURBS work well.

There are some other things too like it helps in many situations that NURBS can represent things like an exact sphere with one surface instead of lots of little pieces and then intersections between a sphere and a plane can be handled as a special case with an exact result.

The tradeoff is that trim edges do not necessarily align with the control point structure of the underlying surface and so you can't modify an object by pulling edge points around like you can in poly modeling except in special cases.

So that also means that if you're not using booleans in your NURBS modeling approach you are also not really leveraging the strongest area that it offers.

- Michael