In many games, there is an option named: "Model Quality", which ranges from Low to High. Low model quality simply removes a lot of vertices from the model, to make it faster, while High preserves the original model vertices (I think).

So I'm wondering, how do I implement this in my game?

  • \$\begingroup\$ The simple answer is that you don't, at least not for PC. Even the cheap chipset GPUs can now handle 200000 on-screen polygons. A few hundred well placed polygons along with bump mapping can give an excellent model, but even if you spend a few thousand polygons it is not really an issue in most cases. \$\endgroup\$ Sep 25, 2013 at 11:32
  • \$\begingroup\$ Normal mapping and tessellation increase detail of given models. \$\endgroup\$
    – danijar
    Sep 25, 2013 at 14:50

1 Answer 1


Usually this is done through LODs (Level of Detail). This is just a name for having several models of the same object with different detalization. In the options you may specify the highest LOD to use.

LOD models are usually made manually by artists because programmatic simplification often results in poor quality for anything more complex than a generic sphere. Programmatic approach may work well for untextures lowest quality models (usually used at a great distances from camera).

In the game you have e.g. 3 models, High-Medium-Low. High quality model is used from 0 to 50 meters away from camera, Medium from 50 to 150 and Low for anything farther that 150 meters. In setting you can specify Medium quality and then you rig Medium model to be used from 0 to 100 meters and Low otherwise. The exact numbers may depend on your game and models quality.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .