Discussion about the split surface component - SrfSplit

Dear all,

I have came up with a issue with the (relatively) new component - SrfSplit.

I tried to search for some post about it but couldn't find much about its story. It is definitely a great function which was a PIA before. I do not know how David or his team managed to write this component, (I heard that) the SDK do not have a function to perform this.

However, I noticed that the behavior of the function could be strange when the surface is non-planar.

Test1 :

All input planar. When the curve input is rotated out of XY plane, the curve is projected onto the surface along the Z axis and then split the surface. OK, no problem.

Test2 :

All input planar. When the surface input is rotated out of XY plane, the curve is projected onto the surface along the normal of the surface (projected onto the surface plane) and then split the surface. OK, no problem.

Test3 :

Surface planar, curve non planar. Similar logic, the curve is projected to the plane of the surface. No Problem too.

Test4:

Surface is non-planar. The trouble comes, the outcome is pretty unpredictable.

I wonder if it would be better if a the SrfSplit function allows a plane input to define the projection plane, to give a better control to it. Or maybe I am mis-understanding the situation.

Replies to This Discussion

Permalink Reply by Gabe Krause on February 8, 2011 at 11:55am

If I understand your tests correctly, it looks to me like it's behaving exactly like Rhino's "Pull Curve to Surface" Command. It projects the non-planar curve to the non-planar surface along whatever surface normals that would intersect the non-planar curve. (I'm sure there's a better way to explain that). You're may be expecting SrfSplit to behave more like Rhino's "Project to Surface" command.