1) No glasses, no camera, small screen, 1 user
(Nintendo 3DS, devices with Sharp 3D screen or Hitachi WOOO Keitai)
How it works : number of pixels on each line is doubled, one for left eye, one for right eye, one for left eye, etc... The theoretical position of each eye MUST be known by the device. Since there is no camera to know it, and anyway, hardware can't really adapt itself to any change -except for a depth adjustment i.e your distance-, it's fixed. Your left eye must be on left side of central console axis, and the right one on right side. Globally, you have to face the console and no longer move your head... (or the console). LCD cells are thus 'oriented'. If targetted eye is not in right position you won't see a thing, or not the expected thing. (Videos showing a 3DS being turned a lot and view inside changing accordingly are fakes).
Pros: 3D effect without the need of glasses or camera. Compatible with pre-calculated movies. (good for small, cheap and large public devices)
Cons: Relative position 'console'-'your head' must be fixed. GPU load doubled for dynamic scenes. Horizontal resolution halved (but since starts high at 800, on 3DS, result is 400 which is usual on last gen very small devices.). You can't share experience with a 2nd user (unless he puts his head above yours, sharing same vertical axis).
I don't think most people will like keeping arms so static (especially kids!).
Thus, can't work with large screen because you can't predict where user will be without camera, and even with camera, the 'expected eye position' is fixed in LCD cell hardware. If console not put on static table, once your arm is tired of keeping static position, most people will turn off 3D effect...
2) No glasses, 1 camera, any screen size, 1 user
('Desktop Virtual reality with head tracking', all platforms)
How it works : camera finds out where are your eyes (webcam, pseye, wiimote, etc... can do the job), then image is calculated so the screen frame acts as a real window. You get the feeling to explore a 3D space behind that window, moving your head a lot allows you to explore different spaces of this VR. Both eyes receive same image but your brain memorizes a real 3D space (shutting one of your eye doesn't prevent your brain from understanding space if you can move your head).
Pros : Associated with a light gun-like controller can really make you feel part of a 3D virtual space. GPU load not doubled for dynamic scenes.
Cons : Not compatible with pre-calculated movies. You can't share the experience with a second user.
3) Glasses, no camera, any screen size, many users
(classic 3D glasses method)
How it works : either by fast shutters switching, or by color filtering (red/blue filters) or elecromagnetic polarization (horizontal/verical filters), each glass will let pass a sight of a full image calculated for the associated eye, alternatively. Both eyes are not shown an image at same moment.
Pros : Compatible with pre-calculated movies.
Cons : You have to bear glasses on your nose. GPU load doubled for dynamic scenes (1080p games on ps3 or 360 will become 720p if 3D glasses are activated). Since device doesn't know where you are, effect may be odd in your brain. Just like if you go to a theatre to see a 3D movie and get a seat completely on the side of the room. Logically images are calculated for a central expected user location. However several users not too far on sides, can share a good experience, close to perfect, especially if screen is very large.
4) No glasses, no camera, any screen size, many users
(2D hologram method)
How it works : Lasers can be used to impress different views on a glass material, but that specific view is rendered only if your eye looks at that matter from the same angle.
Pros: Similar to 'Desktop VR with head tracking' (but just passive matter)
Cons: Only renders static images. Dynamic holograms not invented yet (I think), in our reality. Not easy to do more than monochromatic static images.
5) No glasses, no camera, special cubic space, many users
(3D hologram method)
How it works : It doesn't. Doesn't exists yet. See Star Wars I movie. Matter is affected in a cubic space.
Pros : The best method, I guess, if it existed. You can share perfect 3D experience with everyone around.
Cons : Can't show horizon or landscape, just well defined objects in a specific cubic space. Not invented yet, in our reality.
(However some famous japanese company did show a giant 15 meters tall 3D hologram located above water in Tokyo bay. Technology details are unknown but it may be something that works only on a large cubic space with viewers located very far from it. Video : http://www.youtube.c...player_embedded )
My favorite si Desktop VR (no glasses, no eye stress, no static head)
Becareful if you plan to offer a 3DS to a kid that can't really remain static in one place! This console is best played put on a static table for players staying static with their nose aligned with console axis (prepare yourself mentally to have 3D effects turned off in many circumstances).
Also, note that Hitachi officialy suggests its 3D mobile users to not activate 3D effect for too long (seems to be a lots of stress for eyes)... So I wonder what will happen to kids playing on nintendo 3DS with 3D effect activated for long... (will nintendo allow 3D only during pre-calculated scenes?)
The trouble is that we fall into a nice paradox : Yes, you can predict where user is in front of a small mobile device. But, also, a mobile user, very often, moves...
Edited by openxdkman, 18 June 2010 - 01:29 PM.