Hi there guys
As an old flash developer I’m used to making platform games by loading pages as the player runs to the left or right (like a puzzle) in stead of building one massive scene. I just wanted to know if 3d games with very big scenes work on the same principle?
Tiaan
Yes, it also works like this. A Unity project is made of scenes. Each scene is composed of objects. You don’t have a timeline like Flash (if you were working with code, it won’t make a difference for you). 
so let’s say I make a rally game with a 5km special stage, can i break the stage into e.g. 5 parts (scenes) and load them with code without the player noticing this? Or would I rather load a new mesh dyna,ically?
It would ask to be a contortionist in scripting, but I guess it is possible
ok, so from what i gather there is not really a point on doing that…
Thanks
That should be fine using asynchronous load. If nobody has it here by the time I get to work, I will post a link to the documentation covering it. As for whether or not it would be worth it, it depends on the specs you are targeting. With asynchronous loading, you would have have 2km of the track active at a time.
What you may want to use is Application.LoadLevelAsync. It is in the docs, but is a little confusing. Also, you need a method to unload the last area.
In a sense, you might want to try this.
Start level by loading KM1;
At KM 1.50 starting loading KM2;
At KM 2.25 unload KM1;
At KM 2.50 start loading KM3;
At KM 3.25 unload KM2
The actual unload is up to you. You could go through and destroy stuff based on position or parent everything in each segment so as to facilitate easy recursive destruction.