Say I have a simple cube, and I apply one material to one face, and another material to the remaining faces. It took me a while to realize that I could effectively generate TWO UV maps for that object, one per material. So I can select the one face, apply say a brick image, resize that face in its own UV map, then do the same for the remaining faces with a different image and a totally different UV map. Which, for me, is awesome.
So it’s kinda like UDIM’s, where you can set up multiple UV maps, but I’m still unclear on exactly what’s the difference.
Does anyone know exactly what’s going on under the hood in Blender in this regard?
One difference that comes to mind is that a UDIM spreads one set of UVs over multiple tiles, whereas UV map per material has one set of UVs per map. In other words, a UDIM tile can contain UV data for a small part of the mesh, whereas UV map per material will contain the UV data for the whole mesh for each map.
Whether that’s an issue depends on how much geometry you have. Many UV maps means more redundant data (the UVs for the faces not using that material) and potentially slower texture painting compared to UDIM.
Thanks. I’ve never seen specific mention of this ability to make separate UV’s per material in Blender, so I’m wondering if everyone knows you can make multiple UV’s per object and I’m the dummy? For the longest time I was going under the assumption that for every object you basically have to make a single UV and cram all textures, etc, into it. I guess that’s what they do for games, but for others it seems like a huge benefit to be able to do it per material.
But it seems that any Blender tutorial you see kind of assumes you’re gonna make one UV per object.
gorion103 I’m not sure what you’re talking about. First, for some, memory usage is somewhat irrelevant. But otherwise, are you saying it makes no sense to have two UV’s and two materials for an object?
No… I think they’re saying that you seem to have a fundamental misunderstanding… UV maps don’t belong to materials, they belong to meshes… so what you’re saying doesn’t make sense. Based on your description, you’re actually still re-using the same UV map.
Unless you’re adding new UV Maps in the mesh properties:
… and unless you’re using the UV Map Node in the material editor:
… then you’re not using multiple UVs… you’re just mapping different materials through the same UVs.
Ahh, okay, thanks. So there is one “definition” set (UV map) for all the vertices in an object, which for each vertex maps it to a UV coordinate. Even if you’ve taken one of the faces and scaled it or moved it, all the vertices in the object still share a single definition set (UV map). And that’s independent of what material/texture you end up applying, since all the UV map says is, for each vertex, how does that map to the UV space.
I guess I was thinking that if you applied a separate material to one face it saved it as a separate UV map just for that face…
I think the challenge I had was if, for example, you look at the attached, which is an unwrapped cube where I took one of the faces and scaled it up. So it seems like (to me at least) the single scaled face is in its own UV map since it’s overlapping the others, which is kinda confusing (I thought) for the unwrapping algorithm. But at the end of the day, it’s just vertex 1 maps to UV coordinate A.
