I think you should have mentioned that many people have 2.35:1 constant height setups without using a scaler or lens, they just use the optical zoom on their projector to fill the 2.35:1 screen. And with a 1080p projector you’ll usually get better picture quality doing it this way. The only downside with most projectors is the manual adjustment of zoom and lens shift when changing aspect ratios.

I found this to be a well written, easy to follow, and very informative artical. I’ve read other articals on the subject and was left wondering if I was getting the whole picture.  Kudos.

Re: zooming the projector to attain 2.35… one of the noticable artifacts to this approach is the excess light spill one tends to get outside the boundaries of the screen’s frame.

John,
Killer story covering so many bases. Your benefits listed are good, and would add emphasis that the key benefit is we get closer if not beyond the best movie palace experience Hollywood can offer. And the second most important benefit, it gives high end specialists true differentiation from the ordinary.
Good work, I’m adding it to my tool kit for local training discussions.
Thanks also for the props to me and Runco on our work to date. All good and all fun.

I wrote this as a supplement to your article for my local dealers, it’s long but interested enthusiasts can pick and choose what they’d like to investigate further. Those who really want to read it will want to copy and paste it into a word doc for easier viewing. All the best.

What a great article for CEPro and the readership. There is so much buzz around the scope format, the screens, projection methodology and the system design and installation, that it’s nice to see a ‘primer’ cover the whole gamut. Thanks John also for the acknowledgement of Runco’s early work with CineWide and my co-effort with Sam’s blessing to define and promote it as a mainstream solution rather than the smaller niche it was thought to be when first shown 6 years ago. My main focus was system design and the sales opportunity a new format could bring to our high end dealers. To anchor my work I seized on the three areas I thought defined the ‘Cinematic Experience’ something we can now create with our new anamorphic tools that even Commercial Digital Cinema is still struggling to achieve. Those three areas are; 1. The Image Field of View, 2. The Screen reflected light levels, and 3. A Cinema reference for image Fidelity. As you said above, the field of view available for movie goers, is quite different from what we have been taught for home theater. Your comment about the back row being the ‘money seat’ in commercial cinema isn’t quite right though, and the correct answer bears clarification as it is a keystone to the cinematic experience. Since anyone taking the time to read your article shows an interest in the details, I thought I’d add some complementary notes to your fine work:

Cinema v/s Home Theater Experience: FOV is the difference – If our goal is excellent audio and excellent video image quality, then what we’ve always done is a fine approach. But if we add to our goal that we also need to meet or beat the finest Cinema Experience of a state of the art commercial theater, then we find our old approach lacking. And that last missing ingredient, as I’ve pointed out in my own series of articles in CEPro and WSR, is the field of view range in our home theater designs. Numbers like 3 screen heights or 1.5 screen widths for a 1st row viewing distance are used. And they are valid for TV, a 16x9 720p display for example which was common when those criteria were first promoted. These distances are what give the viewer the classically taught 36 or 40 degree FOV, which again is TV based. But cinema isn’t like that, and the last row defined by THX for commercial theaters is 36 degrees, and SMPTE uses 30 degrees as a last row minimum. These are not the money seats as you described them, but rather they are the “I want my money back” seats. It is the row that transitions away from a cinema experience and into a TV watching experience. Until the anamorphic process was brought to us, we needed to follow TV design criteria because we were using TV technology. A 16x9 image is a TV, even if its 10’ wide, and you need to be 15 to 20’ back to watch it without seeing pixel noise and other visible distractions of the technology. You can’t watch a movie if you’re always noticing technical junk on the screen. So simply stated, the anamorphic process allows the field of view to grow by using all the pixels of the chip, (black bar area pixels get put to work to help create the image). Hence better perceived resolution, 33% brighter, better color saturation and better dynamic range. So the tool that allows the viewing geometry to become cinematic also makes for a better image quality. The easy way to think of this is that if I design a room for an 8’ wide 1.78 screen (54x96) then my first row might be at 12’ back. If I convert that image to scope of the same height, my first row is still 12’, but my screen is now 127” wide (from 40 to 53 degrees FOV). It’s easy to assume this because you still have the same number of pixels per vertical inch, and that’s part of what defines how big a screen can be, or how far a viewer can be from it. Now factor in that our sources and display resolutions have been steadily advancing. A 1080p criteria should improve our ability to expand our screen size and field of view don’t you think? So now take a look again at the THX web site and the ‘Home Theater Viewing Geometry’. It shows a couch as the 1st row and says ‘viewing angle should be 40 degrees, and try to make it less’. Now go to the commercial THX side and they show a ‘real’ theater with a back row viewing angle of 36 degrees. We are being taught to design home theaters for ‘A Last Row Experience’. Now there is the 1.78 v/s 2.35 geometry factor in there, but the fact remains, we are being relegated to the last rows, and I suggest the anamorphic process changes all that, big time!

Resolution creates FOV capability – I am on the CEDIA committee developing an industry recommended practices document for home theater design (go figure) and recently in our kick off meeting in Indy one of the engineers on the committee said, ‘I just saw Dark Knight at an IMAX theater and I sat in the center of the theater (probably around 100 degree FOV) and I was blown away. The best cinema experience I’ve ever enjoyed’. He would never have thought to sit in that location much less suggest a home theater design of that geometry, but Joel Silver told him to try it, and he got the point. FOV and an immersive cinema experience is what it’s all about, from Cinerama to CinemaScope, FOV has always been the holy grail. IMAX is a 70mm process and more resolution equates to a wider field of view. Cinerama was 3x35mm film strips running side by side and blended on a screen that yielded 120 degree FOV’s for forward viewers. Today’s cinema’s have rows from less than 1 screen height to 5 screen heights to cover the viewer preference range and maintain that back row minimum (like the Academy Samuel Goldwyn theater at 24 rows and 1000 seats). The SMPTE recognized preferred seating position range is from 2H to 4H. That’s a range from a little over 60 degrees to 40 degrees. That is a geometry we can now duplicate, if we design carefully. Here’s how.

720p 3 chip DLP = 35mm film’s theatrical performance – The vast majority of movie houses are based on the world standard 35mm film projector. There are about 5000 digital cinemas, 4900 of which are DLP based. There are over 100,000 film theaters around the world. Our cinema experience is based on that 95% group. So an important study conducted a few years ago with SMPTE support explored the actual viewed resolution of 35mm film in SOTA theaters around the world: NY, LA, FL, Paris, London, and Milan. The test was simple; using the best cameras, film stock and processes, a film of test patterns was created that could be exhibited in these movie theaters. The resultant on-screen resolution was measured by industry experts for an analysis of the end to end resolution of the 35mm system. The results were stunning. The measured res was no better than 720p. The average across the screen (unlike digital, 35mm res varies by area) was only 685 vertical lines. Taking into account color space and the bit depth of 3 chip DLP and you have 720p 3 chips every bit the equal of 35mm film. (film stock res is much higher, but the degradation from projection weave, judder, focus shift, and the effects of high brightness and heat yield the net results noted). So we now have a reference, the viewing experience in theaters, and we have a reference point in today’s home theater projection technology, the 3 chip DLP.

DilA and DLP in small, medium, large, and extra large applications – The 35mm reference above allows a logic for FOV capability to be applied to our home theater projector technologies. The best analogy is film itself. Given 8mm is less res than 16mm is less res than 35mm is less res than 70mm we can suggest a DLP analogy. The resolution of single chip 720p DLP is the entry level equivalent of 8mm, using its 1 megapixel chip to create the image with an 8 bit depth per pixel. Next is 1080p single chipDLP with a 2 megapixel imaging device at 10 bit depth per pixel. Next is 720p 3 chip DLP with 3 megapixels in the imaging system and 30 bits of depth per pixel (this is one of several reasons 720p 3 chip looks better and sharper than 1080p single, a surprise to most people, until they see it). And finally we get to the 1080p 3 chip DLP with 6 megapixels and 36 bits of depth per pixel. Looking at a projector’s total end to end resolution which takes into account bit depth, optical MTF curve (I could tell you, but I’d have to kill you) and pixel resolution itself, you get support for the notion of a cascading resolution in our DLP engines. The other technologies are valid, but fall short in adjacent area dynamic range, color purity, and light output, so they only work on screens of 7’ or less by our definition. We need the horsepower to drive the 8’, 10’, 12’ and larger screens that our new FOV expanding tools allow. For DLP you can translate the progressive resolution of these machines to a maximum field of view capability, and express it as the number of screen heights to the closest viewer. The 720p single chip should be viewed no closer than 3H. 1080p single 2.7H, 720p 3 chip 2.45H, 1080p 3 chip 2H, and D-cinema 2k chip machines at 1.75H. This makes the concept pretty simple, pretty logical, and pretty easy to design and sell from as the example below illustrates.

16 foot Lamberts is boring, let’s take the light standard from TV and FOV from Cinema – The next issue after FOV is fL, the screen brightness. ‘No one ever complained because the TV you sold them was too bright’ (quoting Joel Silver from the Home Theater Cruise 2007). Fact is, no one has ever experienced plasma level brightness on 10, 12, or 16’ screens before now, but we have. A plasma is in the range of 35 to 50 fL, LCD a little brighter, and RP still brighter at levels that can exceed 100fL. Once you’ve seen a scope image at plasma light levels, you will never settle for less, if you’ve got the coin. This drives projection sales like Ferrari drives car sales. Horsepower and the rest of what goes with it make upward mobility logical. That’s good for dealers and end users, and it generates over the top home cinema that is truly dramatic, and a far cry from what Best Buy has ever heard of, and that’s our job, creating distance from the ordinary!

Cinema Fidelity Standards – Long ago Runco began publishing performance data based on calibrated projectors, setup to meet the NTSC, and ATSC color and video standards. Light output is given at D65 and calibration tools are included to allow calibration technicians to meet the standards through their adjustment. So in the key areas of image quality; Dynamic range, color fidelity and saturation, white field uniformity, and resolution, we now have a performance in home cinema projection that challenges commercial cinema. And with light output data that can be trusted as a high fidelity spec, we can now design theaters with FOV targets that challenge the cinematic experience of commercial’s best theaters, and we can add a light level that makes film and alternative content look better and more dramatic than Hollywood has yet imagined. We’re one up on ‘em now! We have now deployed more CineWide digital home cinemas than commercial has, and we use our theaters for movies, and concert video, and gaming and the rest using very cool tools to reformat HD content into a scope view experience. This is truly a secret weapon that few have recognized outside our high end circle of Personal Cinema Architects. Maybe that puts us ‘two up on ‘em’.

Put it all together and you can easily design a system of any size and budget – I’ve got this room; 24’ deep, 18’ wide, 13’ high, and we want 3 rows of seating. You need about 6’ between rows for reclining seats so; 24’, 18’, and 12’ rows are indicated. If the theater is to be state of the art, the first row can be 2H, middle 3H, and the last 4H. That’s a 60, 50, and 40 degree FOV in each row, an ideal design goal for a truly cinematic room that looks the part. So this tells us the screen is 6’ x 14’ for the scope image. By the way, it doesn’t matter if it’s a 1.78 or a 2.35 native screen, the 6’ height is defined by the 1080p 3 chip capability and it’s vertical resolution, among other things. So the 1.78 image of nearly 11’ wide is as big as it could be whether the system is scope format or not. There is no compromise in the 16x9 image size when you design systems using this cinematic method. The light level comes next, and for the 4 models I design with in the 1080p 3 chip and D-Cinema 2K chip category, we get fL results of 16fL, 32fL 45fL and 85fL. Price range is $60k, $95k, $115k, and $265K. In our New England market we’ve been selling scope this way for several years and have sold many of the ‘inexpensive’ $60k solution, and some of all the others including a couple of the biggest D-Cinema machines. And the success for dealers is because the logic is easy to convey, easy to understand and people with the means are happy to get the best. Now practically speaking, $60K is tall cotton for many, so if the budget is busted, the next step down is 720p 3 chip. Like 35mm film, it can drive a large screen in a room with a wide FOV, but not as wide as 1080p 3 chip. The first row criteria for our 3 Mpixel DLP device is 2.45H in my design model. Therefore the screen height goes down to about 5’ and you get a 12’ wide screen with projectors in the next lower price range. The light levels track with the smaller screen still yielding upper teens to low 20’s in foot Lamberts. Finally you can go to the single chips and get into CineWide projection systems for $12k to $20K. Single chip 1080p units can drive small to mid size screens and these 2 Mpixel devices can accommodate FOV’s based on 2.7H to the first row. Therefore a 10’ screen can be used for our design example, but light levels will need help from a screen with gain of around 1.3 to get the system into the 16fL class. So the logic is simple, given known seating positions, screen size can be determined categorically. In our example we have 14’ wide for the best class, 12’ in the mid class, 10’ in the entry level class and 8’ in the entry level class (we can do a 7’ with the very most entry level model, and it is usually done for around the house, over the bar type applications, cause it’s still cool). You need 450 post calibration Lumens to hit the entry level class as I define it because we don’t like light levels below 16fL, and we always want more. From a sales standpoint the logic is there, I can give you the widest brightest solution, but you’ve got to write the big check. I can give you the best solution for your budget, but you will have screens sizes that are less wide, and lower n brightness. This approach finally takes the pixel and contrast ratio lunacy out of the equation and gives the client something logical to think about. When our print media says a $6k projector is the best ever, and we give it a 96, it sort of damages the reality of real home cinema design, good for Best Buy, bad for us! It is far too common to see home theaters with 2 or 3 hundred thousand dollar budgets, and a $5k, $10k or $15k projector. I can assure you, Hollywood director private screening rooms are not using our CE toys to create their image reference screening rooms.

Postlude notes that Zoom and Focus in on 2.35 – John’s article is on the subject of anamorphic 2.35 systems, but there are more ways to skin that cat and they bear mentioning. A popular do it yourselfer approach is to zoom, shift, and refocus to get a 2.35 image on a 2.35 screen, and then zoom back down, shift and refocus to get the 16x9 image on the same screen at the same height. But this is a pseudo scope approach because the viewing geometry must be based on the height of the image if the screen were 1.78 at the same width as the scope image. (And the ZFS 1.78 image ends up really tiny to boot). What’s worse, you still have a big TV, you’re just letting the unused black bars spill over the top and bottom of a scope shaped screen. Light overspray is an aesthetic problem, but the real compromise is you only get 66% of the pixels, and you’ve zoomed up their size so the viewing geometry causes smaller screens or longer viewing distances. No getting around it. Two new approaches are called PMI 2.0 and Veriscope from Wolf Cinema. These do the same thing but use indexed motorized lenses to do the zoom, focus and shift automatically. PMI suggests a 2.0 aspect ratio screen to allow a max width for the scope image, and a different width but maximum height for the 1.78 image. It recommends a 4 way masking system to crop the images as they are re-composed. This approach is not constant height like the anamorphic approach, nor is it constant width like a native 1.78 system. To some it may be constant confusion, although the analytical quality is undeniable as every image can be a perfect pixel match to what comes off the disc. In a mastering lab environment that is the way of the world. D-Cinema is also doing zoom, shift, and refocus to yield fewer vertical lines for the scope image than the 1.85 image. But theater owners recognize the benefit of the CineWide approach and have written that method into their recommended practices document published by the NATO. The D-Cinema DCI documents also recognize the benefits of the anamorphic process, and allow for it in their initiative document. Veriscope is another name for the same approach from Wolf Cinema, a newcomer from Sumiko. Veriscope 2.35 takes it a step further by using indexed zoom, shift, and focus to eliminate the black bars that occur when Ben Hur is displayed on a 2.35 screen. Ben Hur is a film using a 2.76:1 aspect ratio which is preserved on the director’s cut edition. A new Blu Ray release of How the West Was Won is also coming out in its original 2.76 format re-mastered to eliminate the edge blending artifacts of the 3x35mm process. The Veriscope 2.35 allows a wider screen to be used with a 2.76:1 aspect ratio, but the anamorphic process is only used to the 2.35 width. After that zoom is used to overscan the screen with the same light overspray and enlarged pixel problems of all ZFS creations, whereby the screen width would need to be reduced anyway, eliminating the FOV benefit that Cinerama was all about. These new approaches have their benefits, but might be closer to the mastering lab experience than the movie palace. My approach is not to duplicate a post production engineer’s experience sitting in a pitch dark room and spilling his coffee, but rather the experience we all get when we sit in our favorite seat in the world’s best commercial theaters. That goal is tough enough, but with today’s anamorphic tools, we have an ability to get there in the privacy of our own personal home cinemas, where we might spill our scotch. The wider than 2.35 notion is a cool thing to consider but there are only an handful of films transferred to video that maintain the 2.55 AR of the first CinemaScope films, or the 2.76 of BH and HTWWW. But stay tuned for a really cool new anamorphic process, I call it Ultra CineWide which does allow a Personal Cinema Architectural design on a superwide 2.76:1 screen, and it does some alternate content tricks that you won’t believe until you have a chance to see it. I could tell you more, but again, I’d have to kill you!

Personal Cinema Architecture – I take an architect’s approach to the home cinema design challenge and have been teaching this process on behalf of Runco for several years now. I provide set of design utility programs called PCA, the Personal Cinema Architect and they help provide the light level and viewing geometry answers for all the projector models we design with in a quick and easy way. We factor in lens quality variations and the screen resolution and light loss to generate the best room and viewing experience for the money. By the way, no good anamorphic lens has a 10% light loss as mentioned in the article above, the best are a fraction of a percent, and the lower line prismatic types perhaps a couple of percent. I say this because an argument for the ZFS approach is sometimes made based on image quality degradation that comes from video processing and the added lens of the anamorphic process. Good video processing is the key, and it’s just as true for a 60’ wide D-Cinema screen where they use it, as it is for a 16’ home cinema screen where we use it. The lens quality is critical as well, and when you do both right, there is no perceived degradation, but rather quite the opposite! The image looks brighter, more saturated. has more dynamic range, and includes the big benefit of the expanded FOV, which ZFS cannot achieve. (We know there’s no degradation because most of the CineWide installs are ‘dual mode’. The lens moves out of the way and a pure 16x9 image is shown without the anamorphic process. And no one has ever said, the scope image looks poorer! In a well done system its dramatically the opposite, the scope image is the jaw dropper!) Screens are also a key component and they can have a wide variation in light loss and resolution loss. The ability to adjust for that reality is what we’ve learned to do. If the screen is acoustically transparent, then reduce the resolution and light factor by around 20% and you simply make the screen smaller or increase the viewing distances in your design. If the screen is of the highest resolution type such as the new Joe Kane surfaces from Da-Lite, then you can go bigger and closer. And don’t be locked into putting speakers behind the screen for audio benefits. Hollywood doesn’t do it because it’s better, they do it because they have no choice. Their screens are floor to ceiling and wall to wall. Localization of sound to the screen action is not helped by having speakers behind the screen, your ears, eyes and brain see the action on screen and place the sound there in a compelling way, you couldn’t stop that psycho-acoustical phenomenon if you tried. As long as horizontal rules of stereophony are followed, the image is stunning. Vertical localization is a red herring, as your ability to localize in that plane is nil, even without the visual cues that go with the cinema experience. Remember your ears are on either side, not the top and bottom of your head. I say this because resolution is a precious commodity, and no screen with holes or weave gaps can be used without that at least 20% loss of resolution and light. The image quality suffers dramatically for an imagined acoustical benefit. We can use them when we are faced with no place to put speakers, but recognize the deleterious affects and adjust for them in your design, and you’ll be fine. Just won’t be as wide an FOV or bright an image.

Hope you enjoyed further discussion on the most exciting topic in our custom business today! – If it weren’t fun we wouldn’t be doing it. And when we find ways to do it better than our competitors, jump on it. I jumped on scope the minute I saw it, on a 22’ wide screen at 35 fL in Sam’s original demo in Las Vegas 2003, and I’ve never looked back. My market is now 90 % CineWide at the high price points above $30K, and still over 50% at the lower under $20k levels. Hope you’re having success with it as well. It’s the most fun we’ve had in a long long time.
See you in Denver if you’re going.