Here are my questions:

1. Are IMAX features filmed from the start for either the flat-screen version of IMAX or the dome version of IMAX, or can prints be made for either version? When I saw "Wings of Courage" years ago at the IMAX dome theatre here in Huntsville, there was severe distortion (curvature of straight lines) away from the center of the image. However, I noticed no such distortion in "Blue Planet" in that same theatre.

2. How exactly do the goggles work? I noticed that I could see reflections off the metal strips on the right wall through the left eye but not the right, which led me to believe that the glasses were polarizing filters, but the announcer before the movie said to push a button on the goggles to "activate" them, which means they're the type that are synchronized with the projector. The 3D image was fantastic, and are far better than anything I've seen with the red and blue-green types of 3D.

There were artifacts that occurred during the movie that made me wonder about some things.

1. 3 or 4 times during the movie, the 3D effect would go awry and everything would be flickery and hard to focus on, but would correct itself within seconds, or at the next scene change. Based on one post made long ago by John Pytlak, I assume this to be the shutter goggles going out of sync with the projector?

2. During the movie, what looked like specks of dirt would appear in the image, but the speck would only be visible through one eye, and would go away after a few seconds. One time when this happened and I lifted the goggles and could see the speck on the screen through both eyes. This means me that the images for each eye are projected at different times?

Evans

While a couple of films were specifically filmed for the Imax Dome with a fisheye camera lens, most are now filmed with a standard lens. In printing the film, a fisheye effect can be created to correct for the shape of the dome. I am also under the impression that some domes may have a fisheye lens for their projector so they can show a standard Imax print. Imax has recently significantly improved their dome lens.

>>> How exactly do the goggles work? <<<

There are two methods to the madness of Imax 3D. Polarized 3D uses polarized lenses in the glasses matched with polarized filters on the projector. Electronic 3D combines the polarized method with LCD shutters in the glasses that are synced with the shutters in the projector. You get a much better effect with the E3D system, as "polarized-only" does not completely block out the other eye's image and causes ghosting on high-contrast scenes.

>>> 3 or 4 times during the movie, the 3D effect would go awry and everything would be flickery and hard to focus on, but would correct itself within seconds <<<

I haven't used the E3D system (which is what you were wearing), but I know that it uses an infrared transmitter to sync the glasses to the projector. If the signal gets interrupted or blocked, the glasses will go nuts. Also, if the battery in the glasses goes dead, the shutters don't run and you either see double or "go blind".

>>> During the movie, what looked like specks of dirt would appear in the image, but the speck would only be visible through one eye, and would go away after a few seconds. <<<

Imax 3D uses two films simultaneously, one for the left and one for the right eye. They are projected at about a half frame offset (it has to do with the threading, the E3D glasses, and reducing the 24fps flicker of two 15K lamps). The film is sucked against a lens called a "field flattener" while the picture is being projected, so a piece of dust may get mashed between the film and the lens. This is why it only appears on one eye. The field flattener is able to move up and down while the film is running, so that it can be wiped with lens tissue (basically) to keep the picture clean through the show. This is why the dust disappeared after a few seconds.

>>> the images for each eye are projected at different times? <<<

Other than the offset, both images (due to persistence of vision) will appear on the screen if you take the glasses off.

Also with the glasses off, you can see how the 3D works ... objects closer to the viewer will have two images farther apart on the screen and objects back in infinity will be practically on top of each other. Try this at home: put your hand in front of your nose and point one finger to the ceiling. Move your hand away from your face and back while keeping your eyes straight ahead (not following the finger). See how the images diverge and converge? Neat, huh?

It took a while to find, but I brought another Imax 3D thread to the top of the list.

AFAIK, when the left eye and right eye images are registered without any fringing ("on top of each other"), the 3D object will appear to be at the screen plane. Most 3D movies emphasize action coming "out" of the screen, but the opposite convergence can also put objects "behind" the screen.

Lenny Lipton's company Stereographics has a website with quite a bit of information on 3D:
http://www.stereographics.com/

------------------
John P. Pytlak, Senior Technical Specialist
Worldwide Technical Services, Entertainment Imaging
Eastman Kodak Company
Research Labs, Building 69, Room 7419
Rochester, New York, 14650-1922 USA
Tel: 716-477-5325 Cell: 716-781-4036 Fax: 716-722-7243
E-Mail: john.pytlak@kodak.com
Web site: http://www.kodak.com/go/motion

The images are not supposed to 'oppositely converge', as your eyes are comfortable crossing to the center to see an object 'in your face', but they aren't comfortable uncrossing away from center. For objects in the distance, your eyes pretty much look straight ahead, but your focus adjusts for the depth.

When we run 2D trailers on a 3D feature, we run the same trailer through both "eyes". The image you see with the glasses off is doubled at 2.5 inches apart, but with the glasses on, the images are fused to a flat 2D image which appears to be slightly in front of the screen.

When the polarizers are reversed opposite of the glasses (ie, left image to the right eye and vice versa), your eyes look away from each other to try and fuse the image and you get a result anywhere from a raging headache to nausea and protein spills.

A good example of stuff moving from up close back to infinity is the opening credits to "Cirque du Soleil: Journey of Man". You are flying backwards through a galaxy with lots of stuff around you and a deep, deep focal depth.

That's about the extent of my polarized stereographic knowledge, so I'm off to visit John's link!

Evans

As strange as it may sound for an object that never gets touched by anything but air or the occasional lens tissue, polarizers tend to wear out. Depending on cooling and lamp focus, polarizers can develop discoloration or streaks and should be replaced at least annually in order to prevent poor extinction rates.

In a theater with just polarizers and not electronic glasses, the signal-to-noise ratio can be as low as 20:1 (at the sides of the auditorium), with 5% of the image being cross-talk from the opposite eye. Obviously the center of the auditorium will provide a higher SNR (150:1) and more enjoyable experience. Electronic glasses bring the SNR up to an average of 500:1. These are real-world numbers. Our auditorium runs at about 150:1 or a little higher.

The glasses can also wear out, due primarily to the chemicals they are subjected to at each cleaning. We have to occasionally check the glasses for proper extinction and toss worn out pairs.

I think the amount of light being used can make ghosting more prominent in Imax. Our measurements drop from 24fL to something like 16fL per eye for 3D. If the ratio is the same for 35mm, 16fL drops to 11fL and then reduce again if it's over/under or side-by-side and it makes the signal hard to see and noise even harder to see.

And I agree with you about the dirt specks. The operator should be watching, whether or not they've seen the movie 300 times.

Evans, for some reason it took me three searches to find it. And I only found it because I knew I posted in it!

What is the shape of the opening in the aperture plate for dome IMAX theatres? Does the entire rectangular area of the frame get projected and "mapped" onto the dome screen, or just part of it?

Evans

The aperature on an IMAX works differently than on a conventional projector.IMAX film is also struck differently. First it is a full 70mm frame sound is sycnronized on a seperate reel to reel magnetic tape deck. Secondly The frames sit side by side instead of top of one another and film is pulled through the projector from side to side in a revolving shutter that act's as the aperature and shutter also, therefor the aperature would be the size of 1 frame of 8 perf 70mm film. Film is projected in a running loop as a opposed to stopping the frame in an intermitten movement so film at no point is stopped for even a tenth of a millisecond. The IMAX projector also has no way of being framed if it is started out of frame so threading on of these giants properly is an all time absolute must. One trick IMAX uses when it is out of fram is to put black electrical tape on the window on the side that is out of frame. Thats right if IMAX is out of frame the picture shifts to the left or the right instead of top to bottom. Hope this helps give you an idea of the IMAX system in general.

I'd like to be able to understand it well enough, given a point (x, y) on the rectangular 15-perf 70mm print frame, to be able to calculate the point (rho, theta, phi) on the dome where that point would be projected, where rho is the radius of the dome (which should be the distance to the projector), theta is the angle measured from vertical (like latitude except going the opposite direction, starting at 0 at the "north pole"), and phi is the horizontal angle "longitude" from the center of the screen. I'm using spherical coordinates for the dome rather than rectangular coordinates because spherical coordinates will probably be a lot easier.

Evans A Criswell

I am afraid you have a few facts wrong.

1) IMAX uses 15perf 70mm not 8 perf

2) The current sound system employed in IMAX theatres is the DTAC system. This uses DVD soundtracks loaded onto a computer harddrive and synchronised with the film either by a SMPTE code generated on the projector (IMAX SR) or through an encoder and interface box.

35mm full coat is still used in some of the older theatres or as back up.

There is also a system known as the DDP which uses 3 interlocked CD's to reproduce the 6 track sound.

3) The film is carried(not pulled) through the projector using a system known as the Rolling Loop. John Pytlak has already mentioned the SMPTE journal articles outlining it's invention and development.

the rotor ,the circular rotating part of the projector,incorporates the pull down(pull across) shutter and the flicker shutter.

4) The frame stops in the gate just like any other film projector.

5) You are right there is no framing knob but if the film is out of rack you should stop and rethread . The idea of putting electrical tape on the port thus blocking out 1/15th of the screen width (up to 7 feet) on one side is crap presentation.

You may have seen the edge of the image masked on the port hole to prevent the picture spilling off the edge of the screen on a 2D projector. this is a permanent feature as there is no screen masking and the old 2D projectors didn't have aperture plates.

Also Omnimax/IMAX dome projectors alll incorporate a fisheye lens.

The horizon line is also different between "flat" IMAX and Dome IMAX and prints made specifically for Dome are/have been printed with the "non image areas " masked out.

For "OMNIMAX" dome theatres, the projectable area is "approximately elliptical", with the ellipse having a maximum width of 2.74 inches (69.60mm)and a height of 2.00 inches (50.80mm). For dome projection, the fisheye lens centerline is offset from the film (image) centerline by 0.37 inches (94.0mm). Films shot in the "OMNIMAX" format using a fisheye lens on the camera have an elliptical image area, surrounded by black.

"Film Dynamics of a Rolling-Loop Film-Transport System" and "New Large-Screen and Multi-Image Motion-Picture System" by William C. Shaw were published in the September 1970 SMPTE Journal.

The original paper "The Rolling Loop -- A New Concept of Film Transport" by P.R.W. Jones was published in the January 1968 SMPTE Journal.

------------------
John P. Pytlak, Senior Technical Specialist
Worldwide Technical Services, Entertainment Imaging
Eastman Kodak Company
Research Labs, Building 69, Room 7419
Rochester, New York, 14650-1922 USA
Tel: 716-477-5325 Cell: 716-781-4036 Fax: 716-722-7243
E-Mail: john.pytlak@kodak.com
Web site: http://www.kodak.com/go/motion

Here's what the image area looks like on a print for the Imax Dome. Green area is the approximate shape of the picture, the black is the hard-matte.

I don't run a dome, so I'm not sure whether the image on the film is printed distorted or not. The dome film that I held in my hands was Beavers (The biggest dam movie you've ever seen!) and it was printed with a fish-eye distortion, but it may have been filmed like that for effect in that one scene and not specifically printed like that for the dome. I have also seen 8/70 for a dome and it looks pretty much the same, but it was the same darned scene (!) so my confusion of printed distortion/lens correction continues.

Evans