Grasshopper

algorithmic modeling for Rhino

Hi guys,

Can someone explain what's happening here?

I'm trying to create a planar outline of a Brep on its plane of symmetry XY. With every method I have tried (Brep intersection with plane, Brep intersection with Brep planar surface, Brep projection on to plane) all gave the same problem. Points originated are not all z=0. They vary between z=10^-9 to z=10^-12 in offset from z=0.

Grasshopper v0.9.0076

Can I fix this?

Thanks,

Views: 369

Replies to This Discussion

Hi Aristo,

intersection operations are subject to tolerance settings, and I'm guessing 1e-9 (and, thus, 1e-12 too) are well below tolerance.

I think in this case the best solution is not to lower the tolerance, but to project the intersections back onto the plane. It may still result in not entirely planar geometry if the plane is not parallel to XY, XZ or YZ, but it'll probably flatten it as much as possible.

Hello David,

Yes, I was expecting something along those lines. It is, naturally, well below the tolerance needed, but further down in the definition I was getting awkward behaviour and I suspect it was from that. I also had previously tried re-projecting the intersection, and managed to lower the offset to 10^-15 as you predicted. I ended up not using this method to get the job done, but I suppose I could put all the points in a list with forced z=0 coordinates.

All the best,

The section component uses surface/surface intersections in Rhino, as the Rhino SDK does not support brep/plane intersections. I suppose I should project the results back onto the plane myself before outputting them, as I know the results must be on the section plane.

Why do you require them to be exactly 0? In general you should treat floating point numbers as approximates. (E.g. zero within a certain margin).

RSS

About

Translate

Search

Videos

  • Add Videos
  • View All

© 2024   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service