Why 3D Doesn't Work and Never Will (2011)

I don't even understand the 24fps logic, as all it causes is tearing and motion blur. We need better framerates, but across the board, not just in action spots, as 48/60 give smoother motion overall. The Hobbit was 48fps; I bet most people either didn't notice, or thought it looked better. Certainly, there was no appreciable motion blur or tearing in a film that would have been full of it at 24.

>2) Brighter projectors - don't know why this isn't the case already.

Already done, along with more reflective screens, but yes, still needs improvement.

>3) Actually shot in stereo. There's a very good chance that the last 3D movie you saw was depth-ified in post process. Shooting in 3D is expensive and requires more editing, calibration etc, so people don't like doing it.

Yep. This is the main problem. Having movies shot in 2D and made 3D in postproduction is like shooting in black and white and having a 6 year old colour them in with crayons. Native 3D shooting is easier than it was thanks to James Cameron et al but still requires more investment in skills, equipment, time, calibration, etc, and better ongoing reviewing and monitoring during production. Some people just don't like to spend money where they should, but still ant to reap the benefits.

The simple --too simple-- view is that screens have to reproduce reality, that the world is 3D and thus that 3D will eventually win. But this has been proven false, at least for audio (which I know better).

Some people think they hear left or right by doing some triangulation between the two ears. Nothing more wrong: with only two ear we would not perceive height, and people deaf of one ear certainly do not "hear in 1D".

In fact we localise sound because of

- The shape of our ears. (See how complex are the ears of some animals)

- Tiny movements of the head.

- Past experience (learning) of the shape of reverberation and reflections in common rooms.

A full "real" simulation of sound localisation, which has been experimented and works, requires:

- Sounds recorded in an anechoic chamber (these are very small and expensive, you won't get a philharmonic in it, and playing music in this echo-less room is extremely painful).

- Microphone must be perfect, a thing that do not exist.

- Synthetic room reflections computed on the fly according to where the listener sits when listening (shape and texture of the room and where are the two ears in the room)

- A polar reflection model of the ear shapes of the listener.

- An helmet detecting tiny head movements and adjusting all the computation above accordingly.

- Perfect earphones inside the ears of the listener.

So this all works in theory and has been tested experimentally, but it has not crossed anyone's mind that we really need this to enjoy a properly spatialized concerto. We can approximate a soundscape enough with the very crude left-right localization provided by stereophony, and this is quite enough to enjoy good music.

It is certainly different for the visual field, but I would bet it will be ressembling in the big strokes: music, movies, books, painting, all these create illusions, automomous worlds that do not need to match reality perfectly. It needs to be realistic enough and based on accepted conventions: When we see the image of a plane taking off, we accept that our hero is likely inside, and that it is related to the story, e.g. not a random plane talking off as we would see from our window.

But it doesn't need to be "pixel-perfect", as exemplified by the many great black and white movies.

I don't know how cumbersome the lenses would have to be, or how insanely fast the system would need to operate, but it's at least possible in theory.

The main clue the brain uses for depth perception on top of the binocular input (of instead of the binocular input for those who don't have it) is parallax effects gleaned from moving objects within the view passing behind each other. On top of that comes knowledge of the world (how big things should be relative to each other) though that is very easily fooled.

On top of that the perception of parallax exploited further by moving your head. Ever seen a cat size up a jump is isn't entirely confident with? You seem them bob their heap up and down a bit giving them extra parallax clues effectively extended the effect of multi-eye vision into a second degree of freedom. This effect can be used by single ey eby moving the head more.

Binocular vision can help 3D perception a lot, which is why most creatures use it, but it is far from the be-all and end-all of it.