Grasshopper Surface split modifies original surface after split

I am trying to split one original surface using two sets of curves. The two curves overlap. Using the first curve (Cutting Crv. 1) the surface splits successfully. However, the second curve (Cutting Crv. 2), modifies the original surface by introducing ‘folds’. These folds are present even after baking the geometry, which means it is more than a rendering issue. Can someone please help me to understand why this is happening, and a workaround?

Srf Split fold issue.gh (92.5 KB)

Hello,

the command "_ShrinkTrimmedSrf " will help. Nevertheless, I believe your surfaces and curves need a drastic “control point” diet. Such errors occur on heavy geometry.

Thanks Tom,

_ShrinkTrimmedSrf has the same issue.

  1. Can you help me understand why I am able to split one surface (shown in original GH file) and not another?
  2. Regarding reducing control points: would it help to cut this with a 45 degree plane and mirror, or is this still the same relative amount of control points

Victor

Srf Split fold issue_re.gh (91.4 KB)

I think you should try this.

2 Likes

If you cut a surface it needs to calculate the intersection curve. If your surface already contains lots of unnecessary cps, the intersection will have much more cps. This also affects the render mesh tessellation of your surface and may cause weird behaviour, since adjacent control point distances of your intersection are extremely low.
So what you are seeing is a render issue created by heavy geometry. Sometimes tricks like Shrink work, but not always.

You can “refit” your shape using one approach shown by @dowazura . Its just not the best way to do. It doesn’t make sense to build something with thousand cps, just to refit it in a second step. Build it clean and lightweight from beginning on. In Automotive industry most aesthetical surfaces have less then 7x7 controlpoints each. Usually we try to model a surface with 3x3 cps and increase the order as needed. -> single span of course, which means the degree is cpcount -1.

In your model I see 5 “main” surfaces and 4 fillets. If you have 6x6 on each this means your shape can be described by 324 control points. (Probably with even less, since your main surfaces are curved in one direction only)

1 Like

Thanks, will work with this.

Tom,

Will take some time to reduce the CP’s. Thanks for the knowledge.

Victor