I know that 3D artists make a 3D model with quads. And they advise to remesh with quads only (if it's possible).

So I have some questions:

  1. Why do we use quads in CG (for games, cartoons, film, etc) if the graphic card (or render, I'm not sure) convert quads to tris all the same?
  2. Why many 3D apps (Max, Blender, Maya, etc) have a triangulate function? Eventually, the graphic card (or render) does that.

I want to clarify this question for myself since I know it incompletely.

  1. Because the geometry will be subdivided into quads. Anyway the primary reason for this is if you have a function with 2 directions it is easer for humans to think it out andcanticipate results if its a quad.

    Software renderers dont have to rely on triangles. And did not for a long time. But then if you microdice your surface who cares its not like you can see the micropolygon anyway.

    For more on this read this post

  2. You might want to control the triangulation yourself. Interpolation does not happen the same way regardless of how the triangles are laid out. So being able to triangulate and then flip offending triangles is useful.


In addition to what other people have said, for cpu / raytracing it is more memory effecient to store quads even though they may be intersected as two triangles.

When you have millions of polygons, this memory saving is useful and there are no real performance side effects.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.