Depth Perception is the operative phrase here.
In humans, visual depth perception emerges from a combination of cues:
Traditional "2D" movies already have most of the monocular cues necessary for depth perception. Without them, you wouldn't be able to tell whether a car was driving towards or away from the camera. However, some monocular cues are missing. Within ~6 feet, "accommodation" can be used to tell you how near or far you are focusing, based on how your eye muscles are shaping your lens.
What most people consider "3D" is just stereopsis -- presenting a different image to each retina. This gets you much closer to "real" depth perception, and is enough of an improvement to be "worth it" for a lot of applications (CAD, Hollywood movies, etc). However, it is missing "convergence," which is how much your eyes have to point towards each other to see an object.
Read the Wikipedia article for the full list.