Procedural Terrain: Chunks as Actors or Components?

Hi,

I want to implement a chunk-based procedural terrain using the Runtime Mesh Plugin. Since I don’t have to much experience, I’m not sure what the best approach is to implement the chunks. I assume the two options are to either have one big Actor that generates new chunks as components or to have multiple actors and a separate object class that manages the actor generation.
(The general idea was to simply split the world in chunks and then create new chunks and unload old chunks based on the distance to the player)

The factors I have in mind are loading / creation performance, collision performance and rendering. I assume that components are not rendered when not in view but I’m not sure how the collision works.

Can somebody give me a few tips on this topic?

1 Like