When creating static meshes, is it necessary (or strongly suggested) to use a continuous, unified mesh? Or is it also OK to use intersecting faces? For example, on a fire hydrant, there are large hex bolts. They could be added like one of the following:
The first way is much easier to model as I just create the bolts and push them through the rest of the mesh. The second way is much more tedious as I have to cut out a hole in the larger mesh and figure out how to fix the topology (This example mesh is just for demonstration, so don’t judge the crappy topo!)
I was under the impression that intersecting faces was a big no-no in UE. However, I have recently imported several models with this kind of intersection and see no noticeable artifacts or anything like that. If it is perfectly fine to just intersect like that, then I will save a LOT of time by doing it that way.
It’s fine, the only thing that can happen is that the lightmap can end up with shadow on the parts that aren’t visible, which can bleed over to the visible parts of the mesh. But you have to test that out and see if there’s an issue after building your lights.
OK, thank you @darthviper107 , I will just test then and make sure I’m not getting any weird shadows. There are certain models (such as those with a lot of tiny dials and switches) that will be much easier to create with intersecting geometry.
I’d think poly count would be lower on models like that too.
I’ve been wondering this myself.
You may be able to do merge the objects much easier if you do CSG though. In 3DS max I’d merge those two objects in a matter of seconds using proboolean. Looks like you’re using blender. I’m pretty sure blender has boolean ops as well.
Keep them intersecting - it’s rare that you’d get light-map bleed on such a small object, and you save on poly/vert count which is a lot more important.
