Hello, everyone. My first post. I'd like to discuss about large terrains i.e. how are they being designed and loaded during the game play. An example for this is Gamebryo engine, where games like Oblivion and Fallout 3 are realized.
From my humble assumption, there is the terrain split into a horizontal 2D matrix of cubes. Each cube has at least two variants: (1) the most detailed, and (2) one or more that are less detailed. Which variant will be rendered, depends whether the player is inside the cube.
Loading would be another topic to discuss here, but lets keep this post shorter. How much right or wrong I am about this? How is it actually being done, perhaps in other games?
Designing and loading large terrains
This is a huge topic but in essence you could:
-> generate the terrain procedurally at runtime, and only generate the part of the terrain that might be seen or can be seen
-> store the big terrain on disk and load in background chunks of it that will be seen, and be careful to load it before those chunks will be visible, so there's a seamless transition and the player will fail to see what's really happening there.
-> generate the terrain procedurally at runtime, and only generate the part of the terrain that might be seen or can be seen
-> store the big terrain on disk and load in background chunks of it that will be seen, and be careful to load it before those chunks will be visible, so there's a seamless transition and the player will fail to see what's really happening there.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement