Closed PrimalOrB closed 11 months ago
From the docs:
All brush geometry must be two-manifold - or water tight with no triangle interpenetration.
You are using open-ended shapes in your example which is not valid for CSG.
Thank you for the response, I will update my build tomorrow with closed geometry to compare results
My understanding of the two-manifold geometry was in reference to full polygons, and not the entire mesh. In the example photos, the polygons/form which are being clipped are entirely on both sides of the clipping geometry with no gaps
@gkjohnson I have updated the codepen link with a toggle for closed geometry, and yes that does do it!
Greatly appreciated! It will take some work for me to implement this in my other project, but this proof of concept certainly helps brighten the outlook!
Thank you again
Good day,
I am looking for feedback or suggestions regarding the following issue.
I have a project which I cannot share, but have created a codepen which will duplicate the general way geometry is being built / the issue it is currently having with resulting geometry.
My current project is using the
three-csg-ts
however it has major performance implications on heavier models. I created a version usingthree-bvh-csg
and was amazed by the speed, but the stability / predictability of output isn't quite what I need yetThe codepen has a toggle between
three-bvh-csg
andthree-csg-ts
so the results can be compared.https://codepen.io/primalorb-the-reactor/pen/abQMjWe
To describe what is being done:
There are two generated geometries, arbitrarily called
These have settings which can be adjusted in the dat.gui
The issue I come across is holes in the result using
compared to the ![image](https://github.com/gkjohnson/three-bvh-csg/assets/69044956/2f9f4157-3bfc-4039-a118-bc300cf02b9d)
three-bvh-csg
like this:three-csg-ts
like this:This issue may be eligible for bounty reward, please reach out to discuss