Your Browser does not support this javascript.
Please wait, loading...

Reflections Done Right

By Admin | 09/22/2016 | 5 Comments


Reflections done incorrectly seem to be a pet peeve of mine. Doubtless you've seen examples of them in magazines, websites, and TV.


I believe some designers continue to get reflections wrong because they have the mistaken belief that reflections behave the same way as shadows. While shadows are the absence of light particles hiting a surface due to an object obstructing them, reflections are quite different.

I suppose I could go into all the mathematics and the compound angles involved—but I think it will be easier just to say that reflections are simply a copy of the object shown, using the object's opposite angle(s)—in a 3D space.

Since most reflective surfaces depicted in art are simply a flat, level plane of some kind—a table, a lake, a river, etc. this should be somewhat familiar. On a level surface there will be no need to calculate more than 2 compound angles. For simplicity's sake, I have reduced these two diagrams to highlight only one angle (the angle the object has in relation to the reflective surface).

(fig. 1) Object shown resting at a 90° angle, reflection has an inverted angle of -90°.

(fig. 2) Object shown just above surface at a 45° angle, reflection has an inverted angle of -45°.

However, if we take a closer look at the two diagrams we can see that not only is there the angle relative to the surface, but also the angle relative to the viewer (this provides objects in the scene with depth and vanishing points). In fig. 1, the viewer is slightly above the first object since we can see the top of the cylinder. The reflection is an inverted copy of the original, but you'll notice that eventhough the object's reflection is opposite—the top of the cylinder is still only visible on the original. This is due to the fact that the verticle vanishining point is still dependent on the viewer's location relative to the object being reflected.

In fig. 2, yet another angle is introduced, as the cylinder is itself rotated slightly toward the viewer so as to make the bottom of the object visable. In this instance, the object has a reflection that has the opposite angle of approach as well as an opposite angle of rotation.

Thus we see that when dealing with level reflective surfaces the angle of the reflected objects are exactly inverted—and, since vanishing points are always in relation to the viewer's position relative to the object being reflected—the verticle vanishing point is inverted or flipped, but the vanishing point(s) along the horizon line is left unchanged.

Unlevel Reflective Surfaces

But what about when the reflective surface isn't level? This is where things get crazy. In these cases you must also take the surface angle into consideration when deciding how the object will be reflected. You can quickly be in need of calculating 3 or more compound angles in these cases.

Reflections can quickly get complicated, but, nothwithstanding the painfulness of the task, it is more important that the object is reflected properly. If you wish to maintain credibility, that is.

So, do incorrect reflections bother you? Do they make you cringe? Do they make you vurp? Tell me about it below.


  • Rembrandt | 6 day ago

    Ah, yes! You hit this one right on Brent! I can tell you know what you are doing and if I were still alive I would ask you for your autograph. People just need to focus more on looking at things reflecting in nature.

    • Picaso | 6 day ago

      Couldn't have said it better myself actually...

  • Da Vinci | 6 day ago

    Proin gravida nibh vel velit auctor aliquet. Aenean sollicitudin, lorem quis bibendum auctor, nisi elit consequat ipsum, nec sagittis sem nibh id elit.


Add a Comments