I've stumbled upon this question on a CG exam:

Which of the following techniques/optimizations doesn't apply to ray tracing?

A: Back-face culling

B: "Shadow-feeler" rays

C: Recursive tracing of rays

D: Calculation of refraction rays

E: None of the above

B, C and D are obviously false, but what about A? Here, is says that "with ray tracing, this feature is not as useful. Generally with ray tracing, we want geometry in the scene to cast shadows for example, regardless of the orientation of the object's surface with respect to the ray direction however the backface culling option might still be desired for primary rays".

I feel that the question is a bit ambiguous, since back-face culling can be applied to ray tracing but it's not (correct me if I'm wrong) a "native" feature as the other techniques.

  • 1
    $\begingroup$ Depends heavily on what you trace, but this isnt a good fit for the site in my opinion written in this form. Edit the post to be about backface culling in raytracing. $\endgroup$
    – joojaa
    Jan 18 '16 at 16:08

I would exclude (A) for the following reasons:

  1. In any modern ray-tracer, the acceleration structure (BVH, k-d tree, etc) holding your geometry is the primary source of culling. The final list of primitives to trace for a given ray may include non-visible triangles, but ideally there should be very few with good acceleration structures.

  2. Ray tracers support many more primitives than just triangles. (How do you back-face cull a sphere?)

The following problems with back-face culling in general are also more relevant to ray tracing:

  1. Transparent materials (such as glass) will make certain objects render incorrectly with back-face culling applied.

  2. Non-closed meshes (like plants) need to render the back-facing triangles.


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.