terrain generatization
generatization... is that a word?
I'm just curious to those who have implemented generated terrains in their games, what is your favored method of choice. fractals/heightmaps/etc with bsp/octree/roam...
I'm just curious to those who have implemented generated terrains in their games, what is your favored method of choice. fractals/heightmaps/etc with bsp/octree/roam...
I use a very simple quadtree subdivision to simplify my geometry into "patches" which are managed by a scenegraph.
I originally implemented a sort of roam-like dynamic level of detail, but the thing is I'm using ODE for physics ( and ODE's trimesh collider for terrain ) and it was disturbing to see the ground popping in the distance but to have objects on the terrain *not* pop, since they followed a static collider mesh.
In the end, for me, I have no performance problems drawing static chunks of terrain.
Also, the word is "generation"
I originally implemented a sort of roam-like dynamic level of detail, but the thing is I'm using ODE for physics ( and ODE's trimesh collider for terrain ) and it was disturbing to see the ground popping in the distance but to have objects on the terrain *not* pop, since they followed a static collider mesh.
In the end, for me, I have no performance problems drawing static chunks of terrain.
Also, the word is "generation"
I had a fairly sophisticated LOD method in Chopper2 based on a paper I found somewhere on geomimapping.
Thankfully I have left that behind and in my current project I just do pretty much what TomorrowPlusX does. I model chunks of terrain, save them as .objs, and draw the ones that are visable.
Thankfully I have left that behind and in my current project I just do pretty much what TomorrowPlusX does. I model chunks of terrain, save them as .objs, and draw the ones that are visable.
Chopper, iSight Screensavers, DuckDuckDuck: http://majicjungle.com
After about 6 months of work it still was miles away from being a fun game, and the code was getting out of hand.
Guess I suffered the same fate as so many have before... Too ambitious projects don't get finished. At least I learnt a lot.
Guess I suffered the same fate as so many have before... Too ambitious projects don't get finished. At least I learnt a lot.
Chopper, iSight Screensavers, DuckDuckDuck: http://majicjungle.com
reubert Wrote:After about 6 months of work it still was miles away from being a fun game, and the code was getting out of hand.
Guess I suffered the same fate as so many have before... Too ambitious projects don't get finished. At least I learnt a lot.
I know that feeling. I've got all these ideas which I have to just forget about if I'm ever to finish my game.

