i am using blender for modeling a building with several floors to use for a UE4 visualization.
basically i model a floor and use an array modifier to make the building, and i have some simple questions please :
do i need to UV map the floor or the entire building when the modifier is applied ?
do i even need an UV map or creating various materials on each different surface of the building is good enough ?
since any way i have to recreate them in UE
i have also discovered that latests UE4 generate lightmaps UV automatically when importing the model, is this good enough or i have to make the UV for the lightmap in blender ?
You need to set up your material UV’s whatever way you need to to get the result you want. As for materials, use as few materials as possible, and try to avoid using multiple materials on a single object. For lightmaps, you get best results doing them yourself than to use an automatic method.
Sorry for droping in this thread, my question is almost identical of so I thought that It d would be better to keep all quests at the same thread for sake of overall integrity rather than posting new thread.
#1) Why do we have to avoid using multi materials in same on single objects ? What potential drawbacks does it have ? #2) Many mentors on this forums suggest to create LM manually but creating manual lightmaps is almost impossible task to achive, AFAIU, I can’t even imagine how one is supposed to cope with thousands of faces on static mesh, align, scaling, reorientate etc… them to form finall shape.
#3) Should we use contiguous LP or seperate faced LP is fine ?
Additional materials increase draw calls. There can be ways to optimize things, so like if you have an object with a lot of solid colors and you don’t want to do complex UV mapping you can make a single texture with color swatches of each color, and then just do a simple mapping to the area of the texture for that color. In more complex materials, you can use masks to use different textures on different areas, like say you want metal on one area and wood on another, you can use a mask to define where those textures go and avoid using multiple materials.
It’s actually extremely easy to do UV mapping, I do it all the time and it’s not difficult, just time consuming. Really just comes down to getting your seams in the right place, the software will do the flattening after that and usually you don’t need to do much tweaking.
This explains why daylight demo, in the UE4 docs section, used the RGB masking for metal, chrome and wooden part, I was wondering how one should go if there was more than 3 materials in static mesh, so I have to use masking instead of multi materials, please correct me if I’m wrong?
*Is it compulsory to have the seams in the mesh?
*Do that lightmaps should be correct in terms of proportional scaling?
Sorry for the mis typo, I meant LightMap ( LM).
As per the Ue4 documentation (Unwrapping UVs for Lightmap), it suggest to use “Contiguous UVs and Padding”. So I understand that all faces of static mesh should be contiguous, right
For masking, I think you could probably use an RGB map, and then use a different color for each material, I would think given how much you can do with the material editor that you’d be able to isolate colors rather than being limited to just the RGBA color channels.
The seams for UV mapping are where you split the mesh so that it can flatten. For instance if you’re flattening a character arm you would have a seam running along the underside and it would flatten out like a cylinder.
Lightmap resolution does not require being the same size across the entire mesh, sometimes I rescale small items in the UV’s so that they get higher detail, or shrink large items that don’t need much detail.
By contiguous, they mean as few seams as possible. Each separate thing in the UV’s requires padding, so the more you split up a mesh in the UV’s the more padding is used which is wasted space. Also, if you use few seams then the pixels in the lightmap can go across those seams, which means you can use a lower resolution.