I am trying to intersect a flat surface with a sphere. The result should be one curve theoretically. Surprisingly two curves are generated. One of the is closed. How come???
intersection problem.gh (4.2 KB)
I am trying to intersect a flat surface with a sphere. The result should be one curve theoretically. Surprisingly two curves are generated. One of the is closed. How come???
intersection problem.gh (4.2 KB)
looks like a problem that comes from the Sphere seam intersecting exactly the Surface edge:
rotating the plane on which the sphere is built solves it:
intersection problem_Re.gh (10.9 KB)
What you wrote is a workaround of a bug. The result should be good at any circumstances, I think.
100% agree with that, geometry seams are still a weak point and need to be further improved
but because that might not happen in the next week or so, trying to avoing intersecting/overlapping seams is the best weapon in the arsenal
Actually I have a pretty large code with high numbers of intersections. So it would be a huge work to fix this. Are you saying the any time I make section, first I have to check the seam (if is any) with an extra code?
no, not necessarily, I’m just saying that seams intersecting or overlapping is (at least for me) a well known possible cause of issues, so when I’m in a situation like the one you have just posted, my istinct says to check the seams as first thing
another thing I try to avoid (especially with spheres, which I use A LOT) if I have to create many spheres on a collection of points, because the Sphere component wants a Plane and not a Point, I first create random oriented not-all-XY planes on those points: this ensures that the seams of all those spheres are not all well aligned on plane XZ, and already by itself solves 99.999% of the issues
if you plug Points where Planes are desired, a XY Plane centered on that point is automagically created, and I try to avoid doing that as much as possible