I’m making an Actor to procedurally place objects around the camera at runtime. It’s mostly as an exercise to see if I can make it work.
I made a base class that places a tiled grid centered around the camera. The tiles then overlap when you move in such a way that the players always stays at the center.
Currently I’m focusing on an actor that then places instance meshes in each tile when they are updated.
I made it so you can select the size of the perlin noise. Having two actors with the same noise size and inverted min/max for spawning makes for interesting effects.
I’m just toying around learning about the editor. I’m not planning on releasing this, at least not for now.
It’s on github so anyone can play with the code and copy it.
I like 3D, large environments and I suck as a designer.
So I’m trying to make something that would look like it was designed/natural, but generated dynamically.
I see… That’d be quite unfortunate if this project won’t develop into a proper plugin or BP system Populating open worlds (or even not-so-open worlds) is a royal pita and I’ve been searching for such system for a while. Would be an insta-buy for me on the Marketplace
Awesome game! Wish it was available for PC… Been getting inspired a lot for Ghost Recon: Wildlands lately, it looks absolutely amazing.
Here is another demo I did. Testing out how easy it is to get ok results. Having the ability to layer multiple noise layers opens up a lot of possibilities.
Added an actor with 80 % spawn chance.
Added a small noise layer with min/max 0.2/1.0
Added a larger noise layer with min/max 0.3/1.0
Added a huge noise layer with min/max 0.2/1.0
Added another actor with the huge noise layer min/max 0.0/0.2 and another with 0.3/1.0 to thin it out.
I’m all about using procedural and user-mediated content generation. Grabbed myself a copy of Terra Nova Dynamic in-game environment builder w/ GUI editor. I can see potential here. What is your ultimate goal with this system? To add Architecture, Roads, Bodies of Water, etc. Perhaps even an GUI?
So far I’m going for 100 % run time generated logic. I don’t really have a ultimate goal here, just playing around and see what I can make.
It means you can’t remove one and one tree. But the goal is for it to be good enough that you don’t feel the need to manually paint/remove.
I am however adding a blocking volume that you can use to exclude areas. Just need to figure out how I can find all my blocking volumes in the world. The API Documentation is hard sometimes.
I would encourage adding a GUI and Multiplayer support which would instantly transform this system into Application. This is what I’m doing with my own in-game multiuser Constructions applications: GOLEMCraft, GLAMOUR, GLAIVERave, WIZMO, EyeManga3D, Labyrinth. When its all said done, I will a MAKE GAME Button. Keep the good work!
There are a lot of seemingly trivial stuff I find quite hard to do. Managed to make it possible to show the result in editor (default off), and the ability to add a volume to exclude stuff.
Added blocking volumes which can contain multiple layer types. Each foliage actor can have a given layer name.
Added ability to run the code while in the editor. Loading, unloading and reloading was quiet a challenge to get right. Here you can also see the blocking volumes in effect.
Here is a demo of my latest improvements.
You can toggle the actor to render in the viewport. And it now loads the closest tiles first, and only one tile per tick.
It loads smooth and super fast.
Populating open worlds (or even not-so-open worlds) is a royal pita and I’ve been searching for such system for a while. Would be an insta-buy for me on the Marketplace 
