Hi there,
I´m very new in Unity but already amazed on its resources. I have to build a full cave based on scan data. So my questions in terms of performance is do I use displacement (already tested and not very high quality) or decimated geometry where I could have pieces up to 1M polys?
Also using Amplify texture and my meshes have several UVtiles. I´m not much interested on the game play and the project itself is just the cave and the camera to move around to show how the real cave looks form inside.
… Im not incredibly experienced, but a tip i have is, if you dont already know, youd maybe want to cut the mesh into many separate pieces of geometry, so the computer only loads what you can see, instead of the whole thing at once as one piece… like if you put your face looking close to a wall, it will only process that one mesh instead of the whole cave system
Even if you’re not interested in game play you should keep it in mind. You may want to do some work with your cave… generate a high resolution mesh from your scan data, and then retoplogize it by creating one or many lower poly meshes to form fit as closely as you need it for the larger details. Then use the higher resolution mesh and bake textures. You can bake the base texture if you need with Blender and can use a tool like Blender or XNormal to bake Normal and Ambient Occlusion maps from you high poly model to your low and then use the low poly model in Unity. You’d be amazed at how nice it still looks as long as you’re not viewing point blank and at grazing angles.
It’s hard to give specific advice without specific information (i.e what format the scan data is in, what the target hardware is, what your use case is). It sounds like if displacement mapping is too low quality for you, you’ve already got your answer. What issues are you running into with displacement vs decimated geometry?