Hello
I’ve developed an algorithm to create procedurally generated dungeons from premade geometry. Here is a video for demonstration purposes:
My problem is that there are way too many draw calls in the scene when walking through the dungeon in a first-person camera. I am looking for a way to optimize the geometry so that only the what is in front of the player would be seen.
My first two thoughts were either to combine the entire dungeon to be one mesh, or to use Occlusion Culling. Both of which need to be premade with the scene, but it won’t work for procedurally generated maps. Please correct me if I am wrong, and I would really appreciate it if someone could give me some optimization tips for procedurally generated levels.