Brain's fusion of two slightly offset eye images into depth perception — the neurological foundation of stereoscopic 3D. Film 3D triggers this innate mechanism; without proper stereopsis, the illusion collapses.
The human eye sits about 6.5 centimeters apart. Each captures the world from a slightly different perspective — this tiny difference is all you need to see depth. Your brain quickly calculates these two images into a spatial perception. That is stereopsis. On set and in the edit, 3D cinema works according to this exact biological principle: you need two cameras (or one with a mirror rig), mount them with the correct interocular distance, and the viewer activates their natural vision — if everything is done correctly.
Practice quickly reveals where the pitfalls lie. The interpupillary distance in humans — the so-called interocular distance — must be consistently maintained. If you deviate, the image either appears flat or causes eye strain for the viewer. Most 3D productions work with an approximate camera separation of 65 millimeters for normal scenes. Even more important is the convergence plane — the point on which both cameras are sharply focused. If it is too close to the lens, a difficult-to-bear "pseudoskopy" occurs, where the perception of depth flips and the viewer is constantly strained. In the edit and during grading, you notice this immediately: too much divergence between the views makes seeing strenuous, too little completely destroys the 3D effect.
A common mistake: beginners confuse stereopsis with pure technical separation — two different images side-by-side or superimposed. That is superficial. Stereopsis requires your brain to be able to truly fuse these images without strain. This means: sharpness in both cameras, exact geometry, no chromatic aberrations. During acquisition, you therefore work either with synchronized dual cameras or with a mirror system (beamsplitter) that guarantees both views share identical focal lengths, apertures, and focus planes. Then, in the Digital Intermediate: alignment must be sub-pixel accurate. A shift of three pixels destroys the stereopsis experience for the entire scene.
The technology varies — polarization methods, anaglyph, active shutter, autostereoscopy — but the foundation always remains the same: two spatially offset images that activate natural binocular vision. Those who understand stereopsis know why some 3D films appear tiring and others work invisibly. It's not about chasing effects, but about the viewer using their eyes as they are biologically designed.