I've taken on a personal project to make a map of the town I live in with some friends. Here is what I've done so far: I've taken DEM data derived from LiDAR point clouds for a 8.939 km x 8.939 km square that covers the city. I've used my GIS software (I do GIS for a living, so this part is the bit I do understand) to clip out only the roads from the entire DEM heightmap. The result is basically a heightmap with only the roads. My thought was that this would make the resulting mesh (which I'm making from the DEM heightmap in WorldMachine) a smaller size so I could work with it in blender to smooth out the roads, and then later combine it all back into one heightmap or .dae file or something. My problem is that it's about 10.25 GB which is too big for me to work on in blender without considerable load times and lag. The attached is the heightmap for the entire area, and then the heightmap for just the roads. The .obj files that I get from these are each over 10GB. Is there any sort of best practice for this workflow? Am I doing something totally wrong, or am I on the right track? How much should be done in the BeamNG editor vs. in blender?
