I am lightmapping a couple of scenes to import into Mozilla Hubs and I have been dealing with a lot of messiness when making lightmaps (diffuse baking with only direct and indirect select) and I'm getting pretty frustrated with it. I feel like I am missing something that would make this all much easier.
Typical solutions that work are...
- correcting flipped normals
- dividing a face into smaller faces
- merging verts and faces and separate parts to be more watertight
- merge by distance
- disolving as many edges as possible
- limited dissolve
- sometimes simply deleting a face and remaking it even works
The manual processes has been extremely laboreous and labor intensive. And there often is a strange issues where artifacts/incorrect lighting "moves". I'll fix it in one area, rebake, and the same sort of artifact is on an adjacent face or sometimes is some other face that isn't connected. Then I will fix that face and it will be somewhere else, etc. This especially makes the work very frustrating, because I never know how long a model will actually take. but I digress.
As I stated before, I feel like there is a better way. Any advice? Or is this just what baking light is?