Trying to make sense of TerrainPage

Firstly, sorry for so many posts to the boards but I’m on fire now but struggling through all this which is completely new to me.

I have looked at the cvs TestTerrainPage code and also the javadoc page. I think I might be getting confused but I gather that this is for rendering huge terrains efficiently?

Lets say I have generated a world (ala Civilization) like this:

which is 512x384. Do the red lines correctly depict how I might split up that map into TerrainBlocks? And to meet the requirements for quad trees I could add a row of black (renderless) TerrainBlocks along the top or bottom? So I would end up with:

Page (World)
    |--- Page (Top Left) ---< (4 Blocks from top left)
    |--- Page (Top Right) --< (4 Blocks from top right)
    |--- Page (Bot. Left) --< (2 Blocks from bot. left + 2 blanks)
    |--- Page (Bot. Right) -< (2 Blocks from bot. right + 2 blanks)

Also, just want to confirm that any objects rendered on a particular Block are put into the scene graph as children of that Block?



When building a TerrainPage, you don’t have to worry about HOW to split it up, only HOW MUCH to split it up. Basically, each pass the terrain page with split the terrain in half vertically and horizontally (creating 4 pages), and recursively do that on the pages themselves until it reaches the level you told it.

I see, that makes things easy. So I guess other objects in the scene graph are shown/not shown based on thier own bounding boxes and therefor attached to the root node?