The technique is called "LOD" or "Level of Detail". It's used by a majority of commercial 3D games to one extent or another, and usually consists of 3 different levels of detail for each model.
This is essentially the same thing as mip-mapping, only on 3D models instead of textures. It's a pretty easy thing to do (theoretically) - you measure the distance from the view camera to each object in the scene, and the distance dictates which LOD you need to use.
As WickedRush says, you can probably do it with Blitz3D as is - my description above is if you want to code it yourself... :)
This is essentially the same thing as mip-mapping, only on 3D models instead of textures. It's a pretty easy thing to do (theoretically) - you measure the distance from the view camera to each object in the scene, and the distance dictates which LOD you need to use.
As WickedRush says, you can probably do it with Blitz3D as is - my description above is if you want to code it yourself... :)