Hello, I have static meshes that are very plain, they do not have many details by themselves. The majority are walls, floor, roofs, etc.
So large areas. I was wondering, if I want to use for example a brick texture, and I want to simulate the volume of bricks, is it better to create one wall mesh by stacking up individual brick meshes or just create a plain wall and add a displacement map in the shaders?
I am interesterd in the performance cost of each option.
First of all with many bricks I meant to have just one mesh but made of many groups of vertices (bricks). It’s still one mesh so 1 draw call.
Displacement in the mesh is simply to modify the mesh itself, which would increase triangle count right (so what I said about simulating brick texture in the mesh)? My question was about increasing triangle count vs using flat surface and adding displacement through material.
I will use a normal map of course for each texture that I apply on the meshes. But I have been looking into PBR textures that have displacement in the texture itself. I was wondering, if I am going to use a texture array in a material in unreal (so I will not texture in Blender), would it be better to have this TArray that would include normals, displacement map, etc. instead of simulating bricks through sculping the mesh itself?
If you have a lot of these meshes in the scene, then Nanite will probably out perform displacement. But you can also have both now, not quite sure what the point of that is
Yes but it is not only the performance cost but also the cost of production. To create detailed meshes takes way more time than just using displacement maps. I’ll try displacement maps first.
Do you know any specifics in terms of draw calls, material complexity, etc. of using TArrays with displacement?