Archived issue #0021415
Bad triangulation of cylindrical face with large vertex tolerances
Description
In draw, display in 3D view in shading mode the attached face.
The illustration of the bug is also attached.
When I analyzed the face, I found that its vertices have tolerance about 40,
while the radius of the cylinder is about 20.
The illustration of the bug is also attached.
When I analyzed the face, I found that its vertices have tolerance about 40,
while the radius of the cylinder is about 20.
Public activity
4 archived notes
Participants are labeled by their role within this record.
Dear Pavel,
could you please review the fix at GIT branch CR21415?
The core of the problem is not a matter of vertex tolerances, but the fact that topological vertices can have different 3d position than the real end points of the edge defined by geometrical curve.
could you please review the fix at GIT branch CR21415?
The core of the problem is not a matter of vertex tolerances, but the fact that topological vertices can have different 3d position than the real end points of the edge defined by geometrical curve.
Proposed fix doesn't implement appropriate behavior as far as it refers to geometrical data only and excludes topology data from consideration.
Problem requires more detail analysis.
Problem requires more detail analysis.
The topological structure of face looks OK but 3D point of both vertices are incorrect. They are located on the cylindrical surface with PI shift along periodic iso.
So, the distance between end of curves (so called logical position of vertex) and its 3D point is diameter of cylinder.
It is proposed to put as designed.
So, the distance between end of curves (so called logical position of vertex) and its 3D point is diameter of cylinder.
It is proposed to put as designed.
This behavior is by design (mesh is adjusted to vertices), thus nothing to be fixed