In the linked thread, running BU on the objects produces a non-manifold surface in V5, but an invalid object (also non-manifold) in the WIP. Maybe this can be checked/improved… --Mitch
Other than little things that come up, I’ve been working on nothing but boolean operations with fuzzy overlaps for the past month. I’m getting close. With my current v6 code, If I union the objects in this file, choosing the lower one first, I get the correct result. In the other order, I get a valid object with 11 non-manifold edges, so I’m clearly not done yet.