http://www.filmjournal.com/filmjournal/content_display/news-and-features/features/technology/e3if3d775aa9c43698eab6fb2500f376296
March 11, 2010
-By Bill Mead
One of the most important but baffling aspects of the cinema technology market is that innovation can come from anywhere, from any direction, or any point in time. Ideas are presented, systems developed, and equipment installed, all competing to better satisfy the filmmaker’s desire to create new and better experiences for moviegoers. Such is the case with 3D systems, where there is a competitive market for digital 3D vendors. To add to the mix, there have been recent announcements of two competitive 35mm film 3D systems: Technicolor 3D, from one of cinema’s best-known brands, and Oculus3D™, a recent newcomer to the market.
The business case for both these new film 3D systems is straightforward: There is a glut of 3D content in the production pipeline and not enough 3D theatres. With the increases in admission prices, and strong audience desire to see 3D content, there is money to be made for studios, exhibitors and equipment vendors. The film 3D vendors believe that more 3D is a good thing. While converting to digital is fairly expensive, adding a relatively low-cost 3D adapter kit to an existing film projector could be a fast route to more 3D screens. This is particularly attractive to the smaller exhibitors who may not have the resources to convert to digital now. Today’s exhibitors are faced with several new choices: Jump in and go digital now or stay with film for the time being, and if it’s the latter, which of the new 3D systems to choose.
Technicolor announced their system in September 2009, followed by demonstrations to studio executives and later to exhibitors at ShowEast in October. Field trials in the Los Angeles area produced market research that showed the audiences were unaware that the movie they had been watching was film or digital. Since last fall, Technicolor has made progress in getting industry support, now claiming seven studios and plans to have 13 titles in 2010, with the first being DreamWorks’ How to Train Your Dragon in March, followed by Warner Bros.’ Clash of the Titans in April.
Technicolor believes their early lead gives them the market advantage and has over 1,000 3D lenses ready to deliver in time for the spring releases.
Oculus3D, the new contender, was founded by three industry veterans, each with impressive credentials in the 3D industry. The team consists of Lenny Lipton, who has work on 3D for 30 years and has over 40 patents issued, including the ZScreen® used by RealD. Also with Oculus3D is Albert Mayer, Jr., an award-winning camera designer formerly with Panavision, and Marty Shindler, an experienced entertainment industry executive and 3D advocate. The founders of Oculus3D believe they have a technically superior solution and while they are not the first to market, the advantage will fall to the exhibitors who wait for their system.
Although it’s easy to understand the business case that there is money to be made in 3D, sorting out the technical issues and claims over each system’s merits is not as straightforward. The issue is not whether 3D on film works—as history proves it does— but whether it works well enough. Today’s audiences have grown accustomed to enjoying full-length features without compromise in quality.
To put this in perspective, for 3D to seem natural, the two eye-images should be identical in every aspect—except for the perspective of the scene. They need to remain symmetrically balanced and must not blend or “crosstalk” before they reach the viewer’s eyes. The left-eye and right-eye images should remain isolated, balanced in color and light levels, and most importantly, presented at the same time. If these requirements are not met, the 3D experience is degraded with artifacts leading to eye-strain or headaches.
With digital 3D, there are a number of techniques for meeting the requirements of stability and eye-image isolation, proven by the fact that patrons are enjoying full-length features using equipment from various vendors. Once properly set up, digital projectors tend to stay the same. However, film wears out as it picks up damage from scratches and dirt. As with 2D film, the viewing experience degrades over time, but because 3D needs two separate and symmetrical eye images, the tolerances are more stringent.
In the early 1950s, the 3D systems used two projectors with a separate eye-image on each projector. 3D was relatively easy to implement as the projection booths already had two projectors for changeover. The dual-projector approach also had the advantage of bright images, but proved to be problematic in almost every other respect. If the projectionist managed to get the two reels threaded and framed properly, the synchronization between projectors was loose, and each had its own characteristics, so the eye-images rarely matched with the precision for extended comfortable viewing. Whether it was the headaches or bad movies, by 1955 3D had become a rarely used cinematic novelty. Instead, audiences went for widescreen 2D, in the form of Cinemascope and Cinerama, which ironically was marketed as “like 3D, but better.”
From the 1960s through the 1980s, 3D came and went several times, but most of these systems were “single-strip” single-projector formats, which eliminated many of the earlier problems. They were more economical and could run full-length features without an intermission, and with both eye-images on one reel, synchronization was less of a problem.
Two systems have particular relevance to the current generation of film-based 3D systems. “Space-vision,” introduced around 1966, was an “over-and-under” approach with two widescreen 2.35:1 images, one above the other, printed in a single academy frame. The projectors were fitted with a special dual lens with polarizing filters and the audiences wore the familiar glasses. (Technicolor also had a 1960s “over-and-under” frame configuration known as “Techniscope,” but it was a camera format and not used for 3D.)
In the early 1970s, a competing 3D format, “Stereovision,” put two squeezed images “side-by-side” with anamorphic lenses to fill the screen. Stereovision later evolved into an over-and-under process and was used on the highest-grossing 3D movie of the time, The Stewardesses.
Forty years later, we have two modern-day versions back with us. Both of today’s film 3D counterparts, Technicolor 3D and Oculus3D, have significantly improved on their predecessors’ techniques, making use of recent advancements in film stocks, digital intermediates, and advances in optics.
Technicolor 3D, like Space-vision, puts the two eye images in 2.4:1 sub-frames, each half the height of the four-perforation academy frame. The top of each image is oriented in a normal fashion toward the top of the full frame. A new dual-lens designed and manufactured by Schneider Optics with circular polarizing filters is used to project and converge the images on a silver screen. The circular polarization used is the same as their digital counterparts, and so is compatible with current silver screens and 3D glasses.
With the Oculus3D system, like Stereovision, the images are side-by-side, but with a significant difference—the images are rotated 90 degrees and positioned so the top of each sub-image is “head-to-head” along the frame’s centerline. Each image is normally 1.85:1, but squeezed horizontally as projected (vertically on the print) by 17%, to use the maximum available area. An external adapter with an arrangement of mirrors and linear polarizing filters is used to twist images back to normal, expand horizontal axis by 17%, and re-converge the images on the silver screen. Both Technicolor 3D and Oculus3D point out that should the theatre eventually change to digital, the investment in a silver screen has already been made.
Both digital and film 3D systems suffer from a common constraint—poor light efficiency—and getting brighter pictures on the screens is a goal for all 3D vendors. Representatives from Oculus3D claim to have the brightest picture while Technicolor claims theirs is equal. While both Technicolor and Oculus3D illuminate approximately the same amount of the frame area for images, the Oculus3D images are 1.85:1 while Technicolor’s are native 2.4:1 with the sides cropped for lower aspect ratios. Oculus3D also believes that their approach of using the single 35mm lens with external adapter is inherently more light-efficient than splitting the available lens barrel area in two with a dual lens.
An important difference between the Technicolor 3D and Oculus3D formats is how each deals with a projection characteristic known as vignetting. All projectors, particularly film projectors, suffer from uneven light across the picture. The image is brighter in the center and falls off toward the sides. To complicate matters, every projector is different and the degree of vignetting depends on its design, its lamp, mirror setup, and its maintenance.
In Technicolor 3D, vignetting causes the bottom of the top sub-image to be brighter than its top, and the top of the bottom image to be brighter than its bottom. The brightness across the two sub-frames is therefore not balanced or asymmetrical. Technicolor takes the asymmetric vignetting into account by dodging each sub-image accordingly when creating the digital intermediate (DI) used for making the release prints. In fact, much of Technicolor 3D process depends on proprietary pre-processing done during the DI phase to get the best possible images on the screen.
In the Oculus3D format, with both sub-frames aligned head-to-head along the centerline, the projector’s vignetting falls symmetrically on the sub-frames, and therefore the two eye-images match on the screen. But, since each sub-frame is brightest at the center-top, there is potential for an amplifying effect when the two images are combined. In a projector with a center hot-spot, the center-top of the projected image will likely be disproportionately brighter than the bottom or sides.
Technicolor 3D uses circular polarization while Oculus3D uses linear polarization. Each has its own pros and cons. While the physics get quite complicated, linear polarization offers higher extinction ratios (the ability to reject the opposite eye-image) when the glasses are properly aligned on vertical axis with the screen. Circular polarization offers lower extinction ratios, but is more tolerant of rotational head movement.
Technicolor 3D and Oculus3D also differ in how each deals with film’s instability in the projector. Film needs to be stopped when the frame is being projected, resulting in a low-level jitter is known as “jump” when moving top-to-bottom and “weave” when moving side-to-side. Much like vignetting, each projector’s jump and weave is different and depends on design, mechanical wear and maintenance.
In the Technicolor system, each sub-frame is half-height, but must be enlarged to fill the screen’s full height, so any physical film “jump” will be magnified correspondingly. Since the Technicolor 3D sub-frames are full-width, the horizontal “weave” seen on the screen would appear the same. And, to Technicolor’s benefit, both eye-images move in the same direction at the same time, so there should be minimal 3D-related artifacts created by the movement.
In the Oculus3D system, however, due to their 90-degree image rotation, the projector’s “jump” and “weave” are reversed. Physical vertical movement of the film will be seen as side-to-side on the screen, and vice-versa. Since the full height sub-images are twisted and are half-frame width, weave on the film will result in increased vertical movement on the screen. But because the Oculus3D images are positioned “head-to-head,” mechanical jitter causes both eye-images to move in opposite directions when projected on the screen. And these monetary misalignments, at least in theory, could produce 3D-related artifacts.
The Technicolor 3D system is symmetrical in compensating for the film frame’s jitter but is asymmetrical in dealing with uneven illumination, while Oculus3D is the opposite and the jump and weave are reversed. Both Technicolor and Oculus3D have demonstrated their systems in controlled situations with relatively new prints and carefully aligned projectors. In an ideal situation, these effects may not be visible, but it remains to be seen how the systems perform in the field with typical projectors over extended periods.
The critics of film-based 3D look upon the announcement from Technicolor and Oculus3D with skepticism. In their view, the development of film-based 3D will create an industry-wide diversion that potentially delays the long awaited full-scale digital deployment. They point out that while the start-up costs for a digital system are high, in the long run digital produces savings for both distributors and exhibitors. They are also quick to point out that digital opens up a whole new world of streaming alternative content such as broadcast of live sports and events, not possible with film.
Another looming issue is whether distributors would be willing to support multiple film 3D formats. In addition to having to provide standard 2D film prints to non-digital screens, they are being asked to provide two additional versions of 3D prints, creating a triple-inventory situation. Worst case, different distributors may choose to support different film 3D formats, causing exhibitors to equip with multiple systems.
It remains to be seen how theatres will market film 3D, but early indications are that simply calling it 3D will be enough—at least for a while. Certainly, most patrons will pay more for 3D, but it is unlikely that the high box-office multiples will continue as more and more screens are equipped and more titles become available. One likely outcome will be that the digital 3D advocates will start promoting the “digital” aspects of their 3D, instead on relying on “3D” itself to draw patrons.
New presentation technologies have always been a driving factor in moving the motion picture experience forward. In many cases, they are disruptive, in that they shake up the existing way of doing things. Much as the recent emergence of 4K digital may have delayed the wide-scale deployment of 2K digital systems, film-based 3D has now given some exhibitors new reasons to wait—and some choices to make. Regardless of what vendors passionately believe and say, the market over time will establish which technologies are viable. While it is unlikely that the rebirth of film 3D will alter the direction of movement toward digital, it will certainly change the timeline.
Posts
