This is great this one.
so we have elipses down a infinitely tall cone and their hyperbolic and parabolic subsection and we have non elliptic parabolic curves.
if you offset horizontally the tip of the cone to the bottom of the cone and use lines to line up your offset cone instead of circular sections then you get convex 0 knot loops or eliptiods with an infinite amount of possible feature dynamic through iterative cone engineering in fact all closed 0 knot 2d curves with zero concavity can be described. Also all open loops smaller than half an elliptoid in elliptical dimension can be described from the not elliptic hyperbolic sections. If you needed concavity than integrating wave dimension into the cone engineering would allow this.
so what about equivalence in 3d
elliptoid bubbles with any number of holes derived from 4d iterative cone engineering.
I like this because It shows how the cone geometry in one dimension up can describe a reasonable curved surface feature set of the dimension below.
this certainly would lead to more advance dynamic in 3d curve surface modelling and physics this type of dynamic cone transformation function set.
You could also do a better job simulating rocket trajectory.