1… Is a Heightmap Impostor in any way more performant than a MeshImpostor
2… Can a Heightmap Impostor be used WITHOUT creating the mesh with CreateGroundFromHeightmap. Basically i have authored mesh tiles already dont need to actually create a ground mesh. But i can also export the heightmap image i used at design time to generate the mesh in unity. But i cant see how you use a separate heightmap image to create a heightmap impostor for an already authored terrain mesh.
3… Is it better to have several small chunks of terrain tile mesh. And what is considered too many. Should i allow you break a large terrain down into 16 tiles, 32, 64, 128 etc… how much is too many terrain tiles. Same for collision meshes (if using mesh impostors instead of heightmap impostors)
basically in unity i can generate the LARGE terrain or segment them out into tiles… I can also generate seperate collision meshes and generate those to even more tiles than the underlying terrain visual meshes… Allow you you have several more smaller collision mesh chunks.
Or i can generate LARGE heightmap terrain images representing the WHOLE terrain and CHOP that image up into TILES to use as the heightmap and CREATE EVERYTHING at runtime in babylon using CreateGroundFromHeightmap and possibly also generating LOD for each segment and i suppose since i am using the ground terrain no need of a separate collision mesh for the impostor since it is using the underlying image the ground was created from… I think.
Anyways… I need your advice on creating the BEST possible terrain exporter for the toolkit. Not just what is quicker, but what will be the BEST DEFAULT terrain export system for the UnityExporter… Thanks again for your input