Home - 3-D Revolution Productions

Stereoscopic 3-D Home

Our Work

Moonridge 5

Directors

Clients

Press / Publications

Contact

Site Map


Stereo 3-D
Our Work
Moonridge 5
Clients
Press - Publications
Contact
 

Contact 3-D Revolution Productions


How a 3-D movie is made – from Home to Hollywood 3-D

Jaws 3-D shoot
ArriVision & StereoVision 3-D Lens Adapters -
Hollywood standards of 1983
IMAX' heavyweight
The 3-D IMAX Camera -
70MM 3-D film recording

Click to find out how to build this and other 3-D Cameras
One Simple 8MM Film Camera plus two mirrors -
D.I.Y. 3-D


A step-by-step guide that will take you from 3-D Script to 3-D Screen

Before you start filming in 3-D

A script should be written with 3-D in mind. There are three axes to work with so action can take place horizontally, vertically, into the screen and out of the screen. While it may be the task of the storyboard(er) to visualise the story in 3-Dimensional mise-en-scene, the script should allow room for this movement. Plot twists, character actions and object importance can be emphasized by using placement into the available depth. Camera moves become a different tool and using cinematic language like choice of lens, light, cuts and transitions changes dramatically in a 3-D film.

Ideally, you will need a subject and a story that will look better in 3-D than in 2-D. A story that translates a unique emotion because of its 3-Dimensionality. One could theorise about what the third dimension means for the language of film - and at The 3-D Revolution we do just that. Check out the pages on stereoscopic cinematography and the reasons for filming in 3-D.

Friday the 13th 3-D
Still from Friday the 13th - Part III - 3-D
In 3-D, we are literally closer to this helpless victim of Jason Voorhees. The horror envelops us -
but does the visual spectacle take us out of the cinematic experience or enhance it?


A 3-D Camera


Clearly, you will need a camera of some sorts to film your subject in 3-D. A DV camera or a regular camcorder or a film camera on its own is not enough. Please note that shooting 3-D is totally separate from presenting 3-D because you can treat any source material in post production to meet your presentation needs.
You will either need a Second camera or a 3-D Lens adapter.

As you may have figured out, to produce quality like you see in an IMAX theatre or in a release like Spy Kids 3-D, you need a film or HD video camera with sufficient resolution to fill a huge theatre screen. In brief, and generically speaking, 3-D IMAX works with a resolution of about 3840 x 2880 pixels, a 3-D HD camera can have a resolution of 1920 x 1080, while a 3-D consumer DV camera works with 720 x 480 pixels. You can see that blowing up DV to IMAX would be quite a painful visual experience.

IMAX HD DV comparison
To rule out any confusion:


DV is Digital Video - the consumer video standard that has replaced VHS and S-VHS.
HD is High Definition - a new professional video format trying to replace 35MM and 65MM film.
DLP (Digital Line Projection) or D-Cinema (Digital Cinema) is a way of projecting high definition video (not neccecarily HD) on a cinema screen.
This is not DVD (Digital Versitile Disc) - the consumer video standard that has replaced VHS.
Although the words 'resolution' and 'pixels' are used here, film like IMAX or 35MM does not have any pixels because it is an optical medium that works on atomic level. However, experts have agreed on a pixel resolution of film so that it can be compared numerically to video.

Blowing up HD to IMAX resolution (DMR) can produce grainy artefacts that will damage 3-D. To read more about this, check out the 21st Century 3-D section.


Alternative techiques

There are two other techniques that allow for shooting with one camera and no camera attachment. These are the Pulfrich method and the ChromaDepth method and they are totally unrelated to any 3-D camera setup.

Pulfrich only works when the camera and / or the objects around it are spinning (moving horizontally), so your choice of subject matter is rather limited. Objects moving from right to left in relation to other objects will appear to be in front of those other objects. Objects moving from left to right in relationship to other objects will appear to be behind those other objects.

This technique works with all forms of presentation and all storing formats. Your audience will have to wear glasses with one clear (left eye) and one darkened (right eye) glass.

Chromadepth requires coloured lights from red, via orange, yellow, green to blue to light your scene. All red objects will stick out of the screen and the other colours go further into the screen. Blue objects are in the back.
This technique works with all forms of presentation and all storing formats. Your audience will have to wear Chromadepth glasses.

Click here to see an example of a Chromadepth movie
Chromadepth 3-D
ChromaDepth 3D Rendering of a Teapot

3-D Setup with a second camera

Twin video camera 3-D setup
Two identical DV cameras on a slidebar by John Billingham
Ideally, the two cameras are of the same make. This, to minimize differences in image quality, look, feel and frame rate. For example: a DV camera and a film camera won’t be a workable setup because video records at 25 frames per second (25 fps) and film records at 24 fps. Also, a lot of DV cameras can film in progressive mode instead of interlaced mode, at which regular camcorders will always record. So watch out for progressive/interlaced settings on your DV cameras.

Interlaced vs. Progressive

In the case of PAL video, recording interlaced means recording an image every 50th of a second, but storing those images on the even lines of video and storing the 2nd 50th of a second on the odd lines of video. These images are then called fields in stead of frames, so while video runs at 25 fames per second, every frame consists or 2 fields and thus video runs at 50 fields per second. Why? It makes for smoother motion and the human eye is too lazy or too indifferent to see the difference between 576 lines or image and 288 lines of image.

Shooting progressive means forgetting about the interlaced technique and just recording the whole frame at 25 fps, using the full 576 lines. This gives a filmic look to your video because that’s what happens in film (which records a whole frame at 24 fps), but fast motion will appear to strobe because of the reduced frequency.

For NTSC, the story is the same, but the fps is 29.97 and the field frequency will thus be 60 fields per second. NTSC has a height of 480 lines, so 240 lines will be used in interlaced mode.

Twin video 3-D rig
Two identical video cameras on a slidebar by John Billingham

 

Setup with a 3-D Lens Adapter (a.k.a. Beamsplitter)

Using a 3-D lens adapter means being able to shoot 3-D within seconds of fixing it to the lens of your camera. There are many different variations of the lens adapter design and each one has its merits and pitfalls.

For film cameras, the choice is abundant thanks to the second Hollywood-driven boom of 3-D in 1983.
With a bit of luck and a 35mm converter these adapters can be used on an SD or HD video camera.

 

Over-under 3-D film

Over-Under '8 Perf Dynavision' Print
on 70MM Film.
The bigger gaps are the divisions between frames,
while the smaller gaps are the divisions
between left-eye and right-eye image.

3-D photo camera beamsplitter
Beamsplitter on photocamera
SIS 3-D video adapter
StereoImageSystems 3-D Video lens adapter on a camcorder

The principle concept of 3-D film lens adapters
is to direct light from two separate points (left eye and right eye points) through mirrors to two separate parts of the same frame of film. This can be done side-by-side to generate two half-width images on one frame, and it is clearly a good idea to include a 2x anamorphic lens in this adapter so that the projected 3-D image will again be full-width. Many variations to this idea exist, like over-and-under and rotated setups, and you can be pretty sure that anything you could come up with for different configurations has already been invented and implemented.

With both film and video lens adapters control knobs may be present that allow you to control convergence and in rare occasions even interaxial distance (not on the StereoVision adapter depicted below, though).


The StereoVision Lens Adapter (Over-Under) as used on part of the Jaws 3-D shoot.

Side-By-Side recording using an Anamorphic lens to squeeze and unsqueeze images


The most popular 3-D lens adapter systems for video employ field sequential 3-D encoding. This means that the left eye image will be recorded onto the even fields and the right eye image to the odd fields (this can be inverted).
Please note the section on video fields and progressive recording. As you may understand, DV and HD/HDV cameras operating in progressive mode are not suitable for field-sequential 3-D recording.

NuView video camera adapter
The NuView Video Lens Adapter
for Interlaced 3-D Recording
The concept of field sequential 3-D videoThe concept of field sequential 3-D video
3-D Field Sequential Image as provided to the Television or CRT Monitor

 

Camera and lens distance

In the case of two cameras, they need to be fixed next to each other on a bar that you (or your cameraman) can hold and / or fits onto a tripod or even a Steadicam.

So how far do the lenses of your cameras need to be apart? Well, this depends on the contents of your scene and desired 3-D depth perception, but as a good starting point and in most occasions the distance should that 65 mm or 2.5 inches. This is the distance between your eyes, so the cameras see the world the way you see it. And changing this distance means manipulating the way your audience sees your scene in 3-D.

This distance is called interaxial or interocular distance in 3-D filmmaking terms.


Steadicam 3-D
3-D legend Daniel Symmes holding a camera
with his 3-D lens adapter - on a Steadicam

3-D cardboarding
Typical example of cardboarding. In this case, the creator of these images was trying to generate
3-D pictures that look good both with and without 3-D glasses. But something's gotta give.

Most lenses, placed next to each other, will create
a larger than natural interaxial of approximately 70-75 mm


So what happens when you change this interaxial value? In short, when you make the interaxial smaller, the 3-D image will look flatter. At a certain point, objects will look so flat, they’ll look like cardboard cut-outs. Hence the term ‘cardboarding’ is used. Inversely, when you increase the interaxial, the 3-D image will look smaller in proportion. At a certain point, objects will look like they are miniatures, or the audience feels like a giant. This is called hyper-stereoscopy.

hyper stereo setup
Murray Lerner setting up a hyper-stereoscopic shot for 'Magic Journeys'
Twin 3-D camera setup
Dual small HD Cameras on a slidebar

What if your cameras are too bulky to have their lenses next to each other at a distance of 65 mm? A setup with mirrors or a special rig will have to do the trick. You can use a simple setup with cameras facing each other, filming off 45-degree mirrors or you can use any other creative mirror solution.

NaturalVision 3-D camera
Warner's NaturalVision camera used on House of Wax
Simple 45 Degree mirror 3-D filming
3-D Mirror setup
Double 45 degree mirrors creating a larger interaxial
Different mirror layouts
Paracam 3-D camera rig

 

Shooting forward or converging

Now you can choose to point your cameras (or mirrors) straight forward or make them toe-in. On 3-D lens adapters you can use control knobs for this. If you don’t fix the cameras to your rig, but attach them with rotating parts, you can control the shooting angle between shots or even whilst shooting!

Shooting straight forward results in the better, calmer 3-D shots but more work needs to be done in post to get the parts of the shot out of the screen and inside of the screen as needed. Part of this work will require Stereo Base shifting, which will result in cropped imagery. Therefore, parallel-shot 3-D is always shot with excess space on the sides.

The work in post to shift parallel footage in depth is not necessarily more work that needs to be done on converging camera shots. Because converged material may well need image distortion correction to counteract keystone distrotion - in other words: keystone correction. Every 3-D shot will need post-productional review as shots may need to be set back or come forward more, have image distortion correction applied and be scaled or clipped on the sides to prevent retinal rivalry.

For more information on getting post work done on your 3-D imagery, contact 3-D Revolution Productions.

Moonridge 5 3-D parallel camera setup
Moonridge 5 3-D still: Parallel cameras used
Stereo Base correction applied in post
Moonridge 5 3-D converging camera setup
Same shot, different setup: Converging cameras used
Keystone correction applied in post

Converging cameras

Toeing-in means crossing the cameras’ view paths and it can result in both spectacular and plain bad 3-D results. The easy thing to understand about convergence is that any object in front of the point where the cameras’ viewpoints cross, or converge, will stick out of the screen when viewed in 3-D.

This out-of-screen, or 'negative screen parallax', also enhances the stereoscopic effect of your scene because of the exaggerated difference between the left-eye and right-eye angle on objects in your scene.

However, when using converging cameras, the depth of your scene can only be limited. Because the cameras’ view paths are crossed, so will the eyes of the audience be and any points on the horizon can be further apart than the eyes can turns outwards. This outward turning of the eyes is called divergence and generally speaking 3-D experts agree that 1 degree of divergence is acceptable but best to be avoided if possible. Needless to say, this creates discomfort with the audience and any available popcorn may be expected to be hitting the screen in abundance.

Divergence and keystoning

Second of all, toeing-in generates distortion of the combined 3-D image. This distortion is called keystoning and it is equal to a vertical skewing in opposite ways for left- and right eye images. This then results in vertical parallax – 3-D’s biggest enemy. It means having an image or image points that are vertically displaced in comparison to the other eye’s image. Since the muscles around your eyes can’t normally do this (one eye looking up, the other down), eyestrain occurs with your audience. Some of this distortion can be corrected in post-production.

Contrary to popular belief, modern technology cannot fix this problem. Even the most expensive and most advanced rig in the world will face the same problems when crossing the cameras' paths. One cannot bend the laws of physics...

Keystone example
Shot in CGI with two converging cameras,
this title shot shows non-disruptive keystoning
Keystone example
With a close up of the upper-corners the vertical parallax
resulting from the keystone distortion is clearly visible

3-D convergence

Converging or toeing-in
Browsing the image forces the eyes to go
from converging (near) to diverging (far)
3-D parallel

Parallel or Straight-forward
Browsing through the image
keeps the eyes converged (near and far)

All these bad side effects can be avoided by employing a stereographer who knows his 3-D. A lot of stereographers use formulae to determine what's the best 3-D setup to be used for particular shots. Hollywood of the 1953-D boom even came up with a spy-on-the-fly 3-D camera value disc (see illustration below). As every 3-D shot requires a different creative approach, simply and only relying on 3-D calculators may not be the best practise. Such a table of values can prevent eyestrain, but it can't prevent bad 3-D movies from being made. 3-D shots need to be thought through from every angle, which is something that a stereographer can help you do.

Surfer Girls 3-D
StereoVision lens converging on wild man with spear for
The Surfer Girls (a.k.a. Fantasy in Hawaii)
3-D film calculator
Motion Picture Research council 3-D calculator
Used from 1953 onwards

Filming in 3-Dimensions

When planning for a 3-D scene it is important to realize that we as humans perceive the 3-Dness world not just because we have two eyes. Things like depth cues, texture, light and shadow and even sound levels tell our brain a lot about our distance to objects around us. Use these elements to create powerful 3-D, or choose to confuse your audience by leaving these elements away. They are important tools of the 3-D filmmaker, besides the equally important camera variables of aperture size, focal length, depth of field, use or absence of colour, and all other tools available to filmmakers.

There are also some basic filming tips to stick to so that your audience will have a pleasant 3-D viewing experience. Of course you may want to discomfort your audience on purpose, but some things just don’t work in 3-D cinema.

Horopter-dictated limitation in 3-D


Convergence, interaxial distance, horopter limitation...

Stereoscopic 3-D is intrinsically complicated and far from a point-and-shoot medium. If you are looking to have all this 3-D information applied to your 3-D production, and you simply don't have the budget and time to investigate it all for yourself, why not contact 3-D Revolution Productions and get your 3-D project sorted quickly, easily and professionally.

Contact 3-D Revolution Productions

Contact 3-D Revolution Productions
Tel +44 1179 441 449


For 3-D without the headaches

 

Camera moves

A very fast moving camera results in 3-D that cannot be taken in properly. A violently shaking camera results in impossible to see 3-D. Eyes observing 3-D need more time than they need for 2-D shots. However, this is all but true for the object the viewer is focussing on. When the viewer is trying to focus on a spinning background, it’s not going to work and he will close his eyes. But when there is an actor of object standing or floating still before that spinning background your audience has a good point of focus and you can pull off such a shot. This also goes for horizontal driving shots; the audience will have the furthest planes of background to focus on.

Otherwise, keep it slow and don’t throw stuff at the camera. Oh, alright then, if you must… just remember what André de Toth, director of House of Wax said: “It’s not to throw things at you but to involve the audience... In stead of showing the story to an audience, make them part of it; the feeling, the experience.” And Rupert Hitzig, producer of Jaws 3-D: “If an off-screen effect serves to stop the action and remind viewers they’re at a movie and not watching a real-life story unfold, it fails”.

Andre de Toth a.o.
André de Toth - wearing his eye-patch - with
3-D professionals Dan Symmes, John Astin and
Chuck Henry. Not being able to see the 3-D effect
for himself still made de Toth a better
stereographer than any of the ones that did.
Jaws 3-D floating arm
Rupert Hitzig may have had a point, but in practice Jaws 3-D features
plenty of these off-the-screen shots of objects resulting from a rampaging shark.

Frame boundaries

If vertical parallax is 3-D’s biggest enemy, objects that are cut off by the frame edge are its second biggest nemesis. In order for the eyes to combine the left eye and the right eye image, the objects in both those images must not be cut off by the edge of the frame. Otherwise, retinal rivalry will occur and again this equals eyestrain.

Right, but every image has such cut-off objects; it is an unavoidable fact of photography. So the best thing to do is to make sure any object your shot is focussing on is never cut off. As long as the audience is looking at one object, they will not diverge their eyes to background objects that may be causing eyestrain because of retinal rivalry. At least, that’s what you aim for.Pirates! 3-D bee

The most memorable shots in 3-D cinema are the ones that contain out-of-the-screen objects. But there is a tricky bit about these objects. As you just read, objects should not be cut off by the edge of the frame. Especially when they are the objects of attention. So you have to somehow stick something out of the screen, close to the camera and thus close to the audience, but keep it small enough so that it stays within the frame boundaries. As you may be figuring out, making a car drive out of the screen and come to a rubber-burning halt to within an inch of your audience is not going to work unless you are using a mini model car. So you will find yourself restricted to using objects like hands, balls, poles, ladders, flying killer robots, and small fluffy aliens.


Contrast


High contrast should be avoided. Pure blacks cannot be distinguished in 3-D and pure whites create ghosting. So night scenes are bad for 3-D (unless well lit). Equally bad are bright lights. Both destroy the 3-D effect.

No matter how good your method of projection is, there will always be some crosstalk (double images) of left eye and right eye images. Sometimes it is simply the brain failing to completely combine stereoscopic images; sometimes it is actually the less than perfect quality of the 3-D glasses, silver screen or colour representation of a projector. Besides these factors, the extra-stereoscopic depth cues are still a very important factor in successful 3-D. Keep even the darkest scenes well lit so depth and object position can actually be distinguished. The more objects visible in space and the more texture, the more depth.

Amityville 3-D still
Still of Amityville 3-D: too dark and that flashlight is another really bad idea for 3-D

Depth of Field

Most 3-D photographers and filmmakers proclaim that the full depth of field should be in focus with 3-D. This, they argue, because the audience’s eyes scan the image and should be able to examine all the objects within the image in all its sharpness and detail – just like in real life. But this is not what happens in real life. When you hold a finger close to your eyes and focus on the room around you, your finger will be out of focus. When you focus on your finger, the surrounding room goes out of focus. Just like we sometimes see happening in film. And so in 3-D cinematography having a limited depth of field should not be in any way prohibiting.

The same goes for using zoom, but in 3-D using zoom also means decreasing the interaxial distance when toeing-in is employed. And this will decrease the depth of the scene.

depth of field in 3-D
Moonridge 5 shot in 3-D with full Depth of Field.
No need to discard one of cinematography's biggest visual impacts
.


Particular to a 3-D video lens adapter


A drawback of filming with a field-sequential lens adapter occurs with any motion in the scenery or camera motion. Because the two fields representing left eye and right eye images are not recorded at the same time but sequentially, so after each other, a 1/50th (PAL) or 1/60th (NTSC) of a second lag occurs between these two images. This can seriously reduce the stereoscopic value of the image or even cause eyestrain. So this goes when horizontal or vertical movement is filmed or a camera pan or tilt is employed.


Editing & Post production


Although for the editing process on its own you can choose any available and workable 3-D display technique, the final edit will have to be post processed to produce a form of image separation. Please note that shooting 3-D is totally separate from presenting 3-D because you can treat any source material in post production to meet your presentation needs.

As with filming 3-D, editing 3-D has some practical limitations – or should we say guidelines. Because the eyes of the audience will ‘scan’ every new 3-D shot, they need a little bit more time per shot. However, shots should not last for too long otherwise the eyes of the audience will start wandering off into the depth of the scene, analyzing the 3-D shot. That will take them out of the movie, so a certain limit should be applied unless you want the audience to do this – when there is just atmosphere and no protagonist on screen for example.

Another good idea is to use slow(er) fades. Again, this gives the audience a chance to adapt to the new scene’s depth. One should most definitely be careful with cutting between different camera setups; close-ups and wide shots will be of different interaxials and convergence values. So to give the eyes a chance to adjust to these differences where an in-between cut of a medium close-up could be used between a wide shot and a close-up. Altenatively, when convergence points are matched from shot to shot, faster edits can be achieved.


Projecting 3-D Film or Video


The biggest budgetary choices are made with the projection part of your 3-D production. A short line-up can be made of the choices, from cheapest to most expensive; the required extras, equalling extra costs, are written underneath the options:

Pulfrich projection - click to watch example movie
Pulfrich glasses

ChromaDepth projection - click to watch example movie
ChromaDepth glasses

Anaglyphic/Anaglyph, ColorCode or Trioviz projection from one source
Anaglyphic or ColorCode glasses – the latter being more expensive with an equal effect

Anaglyphic/Anaglyph, ColorCode or Trioviz projection from one source with beamsplitter
Anaglyphic or ColorCode glasses and beamsplitter

Anaglyphic/Anaglyph, ColorCode or Trioviz projection from two sources
Anaglyphic or ColorCode glasses, red & blue / red&green+blue / yellow&blue colour filters, two projectors, two reels or videos and interlocking mechanism

Polarised projection from one source with beamsplitter
Polarised glasses (linear or circular), beamsplitter, polarising beamsplitter lenses, silver screen

Polarised projection from two sources: Small venue, IMAX
Polarised glasses, two projectors, two reels or videos and interlocking mechanism, polarising projector lenses, silver screen

bulletPolarised projection from DLP source with Real-D projection
bulletPolarised glasses (linear or circular), digital frame server, digital projector, polarizing Real-D adapter, silver screen

3-D projection angle

Colour wavelength separation with Dolby 3-D projection
Dolby 3-D colour wavelength glasses, double speed film / double speed video or HD / field sequential video, sequential wavelength separating Dolby 3-D adapter, regular screen

Sequential shutter projection (active): Xpand, IMAX
Electronic shutter glasses, double speed film / double speed video or HD / field sequential video, synchronised projection shutter, infrared emitter

shutterglasses

Ivanov Free view (glassless) projection
Two screen-size rasters of thin panels, two synchronised projectors, translucent screen

Lenticular or Fresnel Lens system
Lenticular or Fresnel plastic or glass slab

 

Anaglyphic/Anaglyph projection

The most famous form of presented 3-D is the red & blue 3-D. However, this is also one of the least common forms of theatrical presentation because its limitations in quality. In print, however, it is the most common form of 3-D presentation.

The premise of anaglyphic (or anaglyph) stereoscopics is separation of the images through a red image for the left eye and a blue image for the right eye, overlapping on the screen (a grey, purple or yellowish image with red and blue edges resulting). This means the film is going to be perceived as purplish grey by the viewer. Full colour images are not entirely possible. There is a way of using a ‘regularly’ coloured image, with an added 3D effect for viewers who wear anaglyphic glasses.

Spy Kids 3-D still
Anaglyphic still from Spy Kids 3-D using Red, Green and Blue
Spy Kids 3-D still
The same still using just Red and Cyan

Take the cinematic 3-D release of 'Journey to the Center of the Earth': the choice for murky, muted colours has been the same as for what was seen in Sharkboy & Lavagirl: the anaglyph DVD release is to work well colour-wise and not have conflicting colours that produce ghosting with the anaglyphs. So that means no red and no blues or cyans, while oranges and greens are sort-of OK but not really super, leaving browns, pinks and purples besides whites, blacks and greys. This is a bit of a technical story, but it certainly needen't be like this. Using the right colour correction for the anaglyph encoding, 3-D film production can use all colours it wants without restraint, resulting in a gorgeous colour-bursting Real-D (and possibly IMAX 3-D) experience and an equally colour-true anaglyph experience - without the ghosting associated with the use of reds and blues.

Correct colours with anaglyph 3-D
Original Stereo Pair image from Wikimedia, anaglyph created by 3-D Revolution Productions - Original Image from this link

The trick with the whole process is that viewers are not going to watch the 3-D version of the film without anaglyph glasses on, so they only care about what the movie looks like through their 3-D glasses. The 2D version can then have all the correct colours you like. There is, practically speaking, no reason to keep all colours intact for an anaglyph 3-D version of a film. What matters is the viewer's experience and the quality of the 3-D plus colour correctness plays a very important part in that. Anaglyph colour correction results in the best possible end results for both these factors.

If you are a producer or director engaged in a 3-D film production, don't hesitate to contact 3-D Revolution Productions to learn about this anaglyph colour correction. It's time to shoot your 3-D project in all the colours you imagined! Contact us: info@the3drevolution.com

Spy Kids 3-D correct anaglyph 3D colors
2D still from Spy Kids 3-D - lush colours, but a potential nightmare for anaglyphs.
No longer so
with anaglyph colour correction


Black & White anaglyphs

The lack of colour in anaglyphic 3-D doesn't have to be a problem when your 3-D film is black & white anyway - like a 50ties style movie, a movie focussing more on content, or an art film of some sort. Use the look as a strength of your film.

Frustratingly, anaglyphic encoding doesn’t work well with video and TV since its YCbCr colour space is very limited. This includes 4:2:2 DLP projection but excludes 4:4:4 DLP projection. More about this in the 3-D TV section.

The anaglyphic option works without issues for theatre and with computer monitors. This is due to the ability of film and RGB monitors to keep colours intact and correctly separated.


Still from 'Bride of Fankenstein' in 3-D
Converted 3-D still from The Bride of Frankenstein

ColorCode

ColorCode is a patented spin-off of anaglyphics, using yellow and blue to separate the left eye and right eye image. RGB channel wise, yellow is red plus green. The resulting image is in colour in the same way semi-anaglyphics uses all of the RGB colours. The presentation limitations of this system are pretty much the same as anaglyphics.

Still from 'encounter in the 3rd Dimension'
Still from 3-D Mania in ColorCode
Still from 'Dial M' in 3-D
Still from Dial M for Murder in ColorCode


Polarised projection

Polarised projection is quite simply the best way to present 3-D stereoscopic imagery. It enables full colour, full frame, full frame rate 3-D with an audience wearing lightweight, unobtrusive polarised glasses.

This technique works through the premise of polarisation of light, so it can only be used with projection rather than with a single television or computer screen.
Polarising the projected light means allowing only horizontal or vertical light waves to hit the silver screen. Silver, because a normal screen does not bounce all the light to the audience and that would mean a lot of the polarisation getting lost. The horizontally travelling light waves are then allowed into one eye but not the other by rotating the polarising filter in the glasses 90 degrees. This means that one glass only allows horizontal light wave to get through and the other glass only allows vertical light waves to get through.

Polarization explained
How Polarizing light and polarised image selection works

So you can use either one of two projector setups. The first has two projectors: one has a lens with a polarising filter in the horizontal position and the other projector has a lens with a polarising filter in the vertical position. The second setup has on projector with a beamsplitter, projecting over-and-under, side-by-side or frame-sequential 3-D, directing the images through a beamsplitter fitter with polarising filters.

Besides the need for a silver screen, the screen should also be tilted forward slightly to allow the light to bounce straight into the eyes of the audience. Opposite to the gun-ho approach of anaglyphic projection, polarised projection is quite dependent on precise configuration of prime quality materials and a projector’s eye for detail.


The 3-D Glasses themselves

3D glasses come in different quality and prices. As with the different systems, the better the quality, the more expensive the goods. Limited-time projections like festivals will include the cheap cardboard glasses with plastic filters, whereas a permanent show like in a theme park will invest in hard plastic glasses with hard plastic or semi-glass filters.

polarized glassesA very important factor in this budgetary consideration is the issue of ghosting; when using cheap glasses (of poor quality), separation of the left and right image is not going to work as well as with decent glasses. Hence, the left eye will see some of the image meant for the right eye, and the right eye will see some of the image meant for the left eye. This is exactly what should never happen with 3-D. Also, bad filtering means a failing of objects that come out of the screen in very close proximity to the viewer. The viewer's eyes will lose focus, the brain won't combine the two images anymore and two separate images will appear in front of the viewer.


Contact The 3-D Revolution for your selection of 3-D glasses

info@the3drevolution.com


Freeview (glassless) projection

Freeview (free view) 3-D systems, or 'glassless' 3-D systems, are generally speaking not very practical or realistic systems for mass display. They are normally found at advertising spaces, trade shows, theme parks and other exhibition spaces with limited viewpoints. Because these systems work with lenses and their redirection of image parts to the different eyes of the viewer, normally only one to five people can see these freeview 3-D images at the same time. Where there is one perfect (or sweet-) spot that allows for undisturbed 3-D viewing.

Ivanov 3-D Screen

As early as 1940 the Russians had already devised an autostereoscopic system that has been mostly forgotten about. D.V. Surenskii and S.P. Ivanov created an ingenious system for Soyuzdetfilm, using two film projectors, a translucent cinema screen and a raster of fine panels in front of that screen - both on the side of the projectors as on the side of the audience, to make people see 3-D without glasses in a theatre.

The principle of this system is that the beams of light projected from the two projectors are either blocked or let though by the narrow gaps between the raster panels. The positions of the projectors and the subsequent angle of the projected image makes for the partitioning of the left eye and right eye image on the screen. The exact same thing happens with the viewer's eyes looking at the raster blocking the back-projected screen. The left eye will only see the strips of imagery that reached the screen through the raster from the left projector and the right eye will only see the imagery from the right projector.

The only drawback, and the historic downfall of this system is the very limited possibility of positioning of audience – a thin strip of a few seats.

Ivanov raster screen 3-D
Explanation of the communist Ivanov freeview system using ultimate capitalist camera fodder:
The tiltle sequence for Metalstorm - The Destruction of Jared Syn


Lenticular freeview

Lenticular 3-D explained
Explanation of the lenticular system
by Boris Starosta


The mother of all freeview systems is the lenticular solution.

It uses a 3-D image that has left and right eye image partitioned into thin vertical strips onto the back of a print or screen that has a plastic or glass layer on the viewer's side. This layer consists of semi-spherical lenses that are at least half the number of the vertical strips of image projected, left and right next to each other per lens. This way, the viewer's left eye will only see the part of the lenses that lets through the left eye image and the right eye only sees the right eye image part.

A raster of vertical lens strips sits over an image, which is split into strips of left-eye and right-eye views - just like the Ivanov system. However, because of the lenses the left-eye and right-eye strips of the image reach the correct eye without need of a raster screen. The lenses refract, or bend, direct, the images to the correct eye.

This system can also be used for still print. Like specialty postcards, big merchandising property lunch boxes, trading cards and blockbuster advertising posters.

Fresnel freeview

Another system, popping up its re-releasing head time and again, is one that uses a fresnel lens where the television screen or computer monitor would normally be to direct the light from the relevant two projectors into the left eye and right eye. As figures, the angle of seating will be very limited with this system. The difference between lenticular and fresnel is mainly the fact that fresnel lenslets exist both horizontally and verticallt across the screen, while lenticualr lenses only feature in vertical strips across the screen.

Fresnel lenslets

This system is also detailed in the 3-D TV section.

Fresnel lenslets explained
The concept of Fresnel lenslets
A whole screen would consist of hundreds of these lenslets
Kodak 3-D system
Kodak's fresnel HD system
Basically not freeview at all

Besides a limited angle of possible seating of audience, the practise of putting up a lenticular or fresnel slab of plastic or glass on a cinema screen would be quite prohibitive. For television or computer screen both these systems can be quite effective, though. However, as figures, the resolution of any screen using this method is easily divided by four; the minimum number of lenses on a camera built for lenticular photography.



Other systems

There are a lot of other free view 3-D systems that are not suitable for mass audience presentation. These include concave and bent mirror applications, hollow screens, Viewmaster-type, cross-eyed viewing, SIRDS (Magic Eye or StereoGram), Pepper's Ghost (A.K.A. Holavision) and VisiDep (WobbleVision).

Then there are possible future systems like dry ice, holographics, electron beams and impractical freeview applications like the stereoptiplexer that will not be a consumer reality in the foreseeable future.

Therefore these systems will not be discussed here.


Contact 3-D Revolution Productions

Contact 3-D Revolution Productions
Tel +44 1179 441 449

3-D Revolution Productions | 51 Sefton Park Road | Bristol BS7 9AN | United Kingdom

© 2009 3-D Revolution Productions

 
Guaranteed 3-D Excitement!

Sitemap
Stereo 3-D
Our Work
Moonridge 5
Clients
Press - Publications
Contact
Stereoscopic 3-D