I’ve been working on a custom optimization tool for Unity, and recently ran into an interesting architectural issue regarding procedural generation and dynamic objects.
Traditionally, heavily optimized scenes rely on pre-baked data in the editor. However, I wanted my system to support dynamically spawned objects without being restricted to static scenes.
My current approach is implementing a runtime scanning system. Instead of automated continuous tracking the system requires an explicit method call right after the procedural level generation is complete. This scans the current scene state and applies the optimizations instantly.
What approach should I take for this?
I'm open to suggestions.