City Builder Node Placement Question

Unity Engine
1

Hello,

I have been learning some C# recently with the hopes to use unity to develop some small fun games for myself and to generally learn what programming entails. I have been following some tutorials and found myself fascinated with the tower defense genre. Which lead me to ask this very question, how do AAA games tackle build systems? Now I don’t mean the huge complexity behind it but rather the placement system.

With the tutorials I have followed this has been using a tile system (3d). I run a for loop to generate tiles in a grid and then raycast onto the gameobject to see if I can build and if I can then I place the tower (very basic I know, but I am learning). With the maps of the fore mentioned being on such a huge scale (tropico, city skylines) do they still use a node placement or rather a raycast to check the layer and if the building can be placed providing the terrain is empty. I imagine that the tile system I am currently using would be rather heavy on system when you talk large scale. I suppose the main reason I ask is because whilst I am still a newbie I would like to understand industry practices to help refine my code as I learn.

Thank you

2

Large scale games can’t fit all graphics and objects in one scene. Usually there’s visual-less model of the game under the hood and only part in camera frustrum is actually drawn. For user intercations checks are made against that model. What you see on the screen in cities is probably few dynamically generated meshes and a few dozens of shaders for drawing it. If you try to model city like in skylines with buildings, waters, cars, birds, etc… in one scene in regular manner like it is shown in unity tutorial and example projecs, you will probably end up with scene containing many thousands of meshes, gameobject, etc, rendering 2 fps at best

3

Where what palex-nx describes pretty much comes down to MVC, which is common practice in programming. I’m guessing this also applies to AAA games.