Bug in Mesh evaluation or Mesh closest point parameter output

maje90 · October 21, 2018, 2:37pm

I was getting normals of all vertices of a mesh sphere… (i can workaround in countless ways, anyhow)

With Mesh sphere Ex + mesh closest point + evaluate mesh, sometime the evaluated point is in the center of the sphere and the normal vector is almost “null” (zero length)

Rhino 6.10.18288.23181
GH 1.0.0007

laurent_delrieu · March 10, 2022, 2:00pm

Hello
same bug here
I have a mesh here (from an half sphere).
Points and Normals from Mesh Eval have some strange behaviors
Normals have a length less than one

Some points are inside and aligned with the 0 of the mesh.

mesh eval error.gh (214.7 KB)

DanielPiker · March 10, 2022, 2:11pm

I confirm the bug is still present. It seems it only affects quad meshes though, so triangulating can be a way to work around it.

https://mcneel.myjetbrains.com/youtrack/issue/RH-65889

laurent_delrieu · February 21, 2023, 4:49pm

Hello
the bug is still present with quad mesh. Triangulate mesh heal the problem.
You will see on this example that some points after Mesh Eval are put on the origin with a normal with a length quite equal to zero. It happens when the first parameter of MeshPoint is 1.0 and all the others are 0.0.

mesh eval error 2.gh (16.1 KB)