Has 14fl 3D become easy in a primary Laser world?

[James Gardiner]

It has become more important in recent times to offer cinema experiences that cannot be achieved at home.
3D has largely done poorly as 3.5fl sucks.
Domestic 3D never really worked as TV in the home is not suited to wearing 3D glasses unless you are in a fully immersed environment (like a cinema)

In Sydney there is a Cinema with a new RGB primary laser projector. Runs about 33% on a 12m screen. Adding a polarising lens beam splitter system with triple flash, turning the laser power up to 90% (from 30%), 14fl is easy to obtain.

However, is this satisfactory? I have not installed a RealD system for a very long time, but from my understanding, due to polarised lens never archiving total block out of the other eye, Polarised systems were subject to the worst crosstalk of all 3D systems. Ghostbusting, in that one image has a contrast alteration to make it easier for the brain to ignore the crosstalk.

One notable issue was that the more you turned up the light, the more crosstalk became a problem.

I was surprised to hear that this solution as mentioned above was one of the only 14fl presentations of Avatar2 in Sydney. (Being a generic Cinema, not IMAX) I was under the impression, utilising polarised 3D, 14fl was not an acceptable experience.

So. the question is, has polarised 3D gotten better? Do the primary lasers allow for better isolation of the left/right eye, making this method of cinema-grade 3D at 14fl work exceptionally well?

The fact is, I am hoping so, as it gives cinemas a differential from the home cinema system. And if we can do 3D well, 14fl and shot right as JamesCameron demonstrates. It is much more compelling, and a point of difference we should nurture as a way to lift the cinema experience in this more competitive environment.

Can anyone shed light on if they have seen other cinemas do this for Avatar2, and how they have obtained an acceptable 3D experience considering the issues mentioned here, and how they were overcome?

[Steve Guttag]

14fL (or any light) on a silver screen polarized 3D remains problematic, at best. The physics of light just don't work well for it (by the way the 2D light on that silver screen suck as well and will cause them to look dark too).

In a nutshell, here is silver screen 3D's problem (or any high gain screen...once you go about 1.3 in gain, you are hot spotting outside of SMPTE tolerances).

SilverScreen.jpg

This is a "typical" site with Real-D XL on a flat silver screen (it is a xenon but the physics of light remain). Okay...so you set your light to 14fL in the center...what do you think it will be on the sides? It is going to be around 3-4. The "average" light, if you were to divide up a silver screen into 45 separate sections, with the center section measuring 14fL will be? Anybody? Just under 9fL (about 8.7fL, I measured). That is how that image will be perceived to people. It will "feel" dark even if the very middle is at spec.

There is too much emphasis placed on the center brightness and not enough placed on overall (except for Dolby Cinema because they do not require gain screens).

It is true, you can get yourself a laser projector that will get you the 14fL in the center now (and have the range to back down for 2D)...and that is better but it will still appear dark (remember, they change the dynamic range of the movie to suit the peak white level). One can curve the screen to get a MUCH more even light distribution, however.


CurvedSilver.jpg

However, nothing is free. You've introduced geometric distortions to the image as the prism outputs as a flat plane. Warping or purpose built lenses would be needed to really deal with the curve. Sure, you can "mask" the image but you are not fixing the underlying distortion. And, with Real-D XL, you can really forget about a precise alignment. Something has to give because you have multiple sources that are trying to overlap on a curved screen. I tend to get the middle and bottom as best as possible and let the "uglies" be in the upper corners (there is less light up there too). So, in the case of Real-D XL, you, essentially, take a resolution hit on the curved screen because you are no longer precisely matching the two images. Remember, the XL box is not putting out left and right eyes at the same time, it is overlapping whatever the projector is putting out and reclaiming light that would have otherwise been lost.

Dual projector 3D can certainly hit 14fL in the center on any screen surface. Dual projector also doesn't have to worry about as precise an alignment (each projector is a separate eye) So, there is no resolution loss by virtue of the double-image...just eye-strain and possible z-axis movement, depending on the misalignment. Then again, 3D is a farce to begin with as it is not true-to-life. The focal plane will always be the real screen. Even if an object is moved in z-space, your eyes will still need to focus, not on the virtual z movement but on the real screen...that isn't natural.

So, James, it isn't really lasers that allow higher than 4fL 3D, it is how one does 3D. What lasers do for the likes of Dolby is to negate the need of the filter wheel by using primaries on the laser that already encode each eye. And, if you use dual projectors, you've picked up another 50% of your light right there (same as xenon...because, if you think about it, in a single projector system, 50% of the time, minus the dark time, the "other eye" is projecting. You can get a rather efficient Dolby 3D laser system. That said, I believe Barco had dropped their "Flagship Laser" system that had 6-primaries to allow a single projector 3D, leaving just Christie's dual-projector 6-primary system. Barco may have a dual-projector 6P system but I'm not aware of it.

The bottom line is, if you really want 14fL 3D, you are talking 2-projectors. If you are going with polarized (Real-D), you want to lower your gain (and they now have white-screen, lower-gain that retain polarization...they are not cheap) and size your projectors and/or curve your screen as appropriate to the gain.

[James Gardiner]

One of the bigger problems back when I was installing 3D systems on xenon based projectors. Due to the limited range of a xenon lamp 70-100% it was impossible to do 2D and 3D well at the same time. You simply did not have the extra headroom to turn up the lamp enough to get much more than 3.5fl, and that is at optimum performance, new lamp etc.
Laser allows you the ability to get a projector where you CAN turn up from 30% to 100% easily allowing a 14fl at 2D and 3D with all (single projector) 3D type systems.

To be honest, I am bullish on active glasses on smaller screens. The glasses are a lot so the smaller audience makes it easier to police the return of the glasses.. You are on a low-gain non-silver screen and all the benefits that brings you. No hotspot, less compromise in image etc.
Looks great on my (small regional) screens. I have 2 utilising this technology and I can tell you they look much better than a typical big chain old-school 3D system.

Doing 3D right, to me, is a key to its possible success in the future and a solid point of difference cinema can offer to the public.

So I am bringing this topic up as to estimate the validity of this path for cinema in the future. I am not so much talking Dolby Vision and the ultra-expensive (typically dual projector) PLF solutions. But a solution all cinemas exhibitors can access.
Having a few large PLF cinemas, like we have a few large concert halls for major live acts, is not the trend I want to see form for cinema.


But back to the topic at hand.
This cinema doing 3D 14fl in Sydney. I would judge from your response Steve, is not advisable. (Or likely to be able to display consistent brightness across the screen based on the SMPTE standards.)
To tell the truth, still much better than 3.5fl 3D. I was more worried about the crosstalk and the poor effect it would have on customers. Headaches etc.

[Steve Guttag]

Remember, if it is at 14fL, then the "print" will be timed for 14fL. That is, the darker parts will be notably darker than on the lower fL ratings. So, your crosstalk will diminish with that. But, again, 14fL in the center is great for that small spot of light. It won't be 14fL anywhere else. Again, one can curve a screen, like a reflector (none of this 1:20 BS that has no basis on anything...and yes, I've read the paper...long ago...its conclusions were not based on their own data). You have to do ray tracing based on the screen's reflective properties, where the projector is located, where the audience is seated. THEN you can get a higher gain screen to be reasonably even (as seen in my post above...those are actual pictures of screens as they were being projected white light...that is what they look like to the patron).

I agree that with laser one can have an enormous range, as compared to xenon. Furthermore, you don't hurt the lasers as much when you turn them up like Xenon. Xenon doesn't like being turned down once it has been burning hard...they start to flicker/hard-strike.

I also agree with respect to Active Glasses. They don't force one to compromise their 2D shows for their odd 3D show that really benefits from 3D. They just cost a lot (to buy, to collect, to sanitize...etc.)

[James Gardiner]

Steve, just a comment. Yes they are using the 14fl graded version of Avatar2, 3D.
I must admit, I did try turning up the A2 trailers which are obviously targeting 4fl, and it does suffer noticeably. For me anyway. I don;t think many general punters would notice the difference. The way the darks fall off does not look right and it can look like the brightness is turned up too much like on a TV. But then again, still looks much better than running it dark. It's really frustrating.

[Marco Giustini]

Slightly off topic, I remember commissioning a system which had a good amount of light on a nice, large screen, by doing a 9-point average (it was a silver screen on a flat frame). I believe I ended up around 21fL in the centre to get 14fL average, the 2D picture was pretty good (all considered) in my view.

Too bad I was told off for doing that as apparently if I had just shot 14fL in the centre, the cinema would have been able to downgrade the lamp.

[James Gardiner]

Marco, Yes, I find it most frustrating, especially when in technical meetings like ISDCF and others, sometimes, when technical specs are sprouted out by attendees without any understanding of the commercial imperative that cinema owners and vendors operate under. They find it hard to understand why compromises occur so often. Or why some of the good work these committees do in regard to improving the industry.. are not freely adopted.

Integrators and vendors would expose themselves to competition etc if they did implement some of these standards. For example an Open TDL and universal FLM message sharing. There is no reason this shouldn't occur apart from commercial ones.

[Harold Hallikainen]

I participated in some of the ISDCF Picture Level tests. I was there mostly to compare the luminance readings from an LSS-200 with other instruments being used. We found that it was difficult to compare luminance since a change of a few seats or rows made a fairly large change in measured luminance. In the tests, there were qualitative tests of images on a variety of screens. The image on some screens with substantial side and top/bottom luminance fall-off was judged to be pretty good. This made me think that PERHAPS the luminance specifications in the SMPTE standard are tighter than they need to be to provide a good image. I don't know where the numbers in the original SMPTE specification for luminance uniformity came from. There are some ISDCF notes on all this at https://isdcf.com/MeetingNotes/20200...s-at-ISDCF.pdf .

I was also on a SMPTE committee dealing with luminance measurement in general. We were looking at luminance at various locations on the screen AND luminance measurements at various points in the auditorium, including the ends of the front row where I assume luminance and uniformity would be worst.

Harold

[Marco Giustini]

Indeed, James. As usual it boils down to ignorance, incompetence and business.

If I try and do the right thing and specify the correct lamp and screen to get an average of 14fL, I get beaten by competition who use the same equipment but only aim at the hotspot and quote for a smaller lamp/projector. For the ignorant customer, "it's the same brightness but it's cheaper".

[Steve Guttag]

Marco is correct. If you want a silver screen, on 2D to have a non-dark "feel," you should be up around 21.8fL. At that point, a typical silver screen with a 2.2-2.4 gain will have about the same average distribution as a 1.3 gain white screen. The 1.3-gain white screen being the highest gain that will achieve (or be very close to) SMPTE specifications for light distribution.

[Stefan Scholz]

From my 1980s text book on film projection, starting with the visual mechanism, the authors start pointing out, 50 candles is the border to night vision, where you loose full color impression and everything turns to be grey. That's about the imperial 15 fl that are target standard, for a reason.
In 3D going lower into full night vision not a lot of color rendition is left, one reason, beside poorly adjusted 3 D systems and the fact, that 3D only works for an adult only works, if the images are specifically created in a "wrong" manner for eye/brain. It can never work truly. E.g., as Steve pointed out correctly, as the focal plane is still the screen surface, and not the virtual image infront.
The late 1930s research on the subject clearly showing, an adult benefits most from grain free, large, bright, vision filling and lifelike dimensions on screen imaging. Eventually combined with a higher frame rate, that the minimum compatible standard at the time, 24 fps for sound on film.
In the 1950s these results ended in 70 mm ToddAO, which offered a superior experience. Beginning with standardized theatre designs, higher frame rates, and dual brightness specificication at 32 fl, if reading in the literature of the time, it offered a sensational experience for some movies.
Why do I explain this here? I'm a little tired on a discussion "how to make 3D better looking", there are too many compromises and no real benefits. If our industry would be focusing on correctly designed halls with good screen sizes, even and bright images using the full 12/ 36 bit dynamic range possible, good blaxck levels (you may need more semiconductor light modules, improved light pipe design, and Ultra High Contras Premium Plus lenses at top additional cost), I feel this is worth the efforts to to combine it with sound system designs with the slider in recommended configuration, to offer a truly unique experience. The home can do it, there's Residential projectors from Barco and others outperforminmg the Cinema line in regard of image quality. Combined with full blown sound systems. Limited to those with the space and money.
§D, as it was done in the past ruined more, than it gave benefits. Moviegoers complaining "not wanting to see it in 3D", "if it's again in headache glasses presentation, son, it's last time we go for the movies", heard too often. Let's go for real improvements.

[James Gardiner]

With A3 and more films now likely, I feel a real discussion on doing 14fl / 48nit 3D needs to happen. Dim 3D sucks, and I and I feel the general public, in the long run, would prefer just a 2D presentation.
Still, if laser projectors make it easier and more cost-effective to achieve higher brightness 3D, maybe 3D as a cinema-only technology, a promotional edge on streaming, does have a place after Avatar sequels are over.

With the effect of streaming after 45 days becoming the expected by the consumer, we are definitely seeing a contraction in average visits per year per person. So putting some effort into a quality Cinema 3D experience becomes more compelling for the industry. i,e, enforces a tiered release. 3D only in cinemas, then watch again in other release tiers. More likely to get more shots at transactions from consumers. It's been well documented that the demise of the plethora or release windows eroding down tio... Cinema, if lucky, then Streaming, DONE. Has not been a commercially beneficial path.

So a good question here is. What's the best and most cost-effective recommendation installers reading this group would give to a cinema?
Projector, screen size, and 3D technology used?

[Steve Guttag]

14fL as the arrived standard had nothing to do with perceived color. It, and its 16fL film counterpart had everything to do with the technology of presentation. With film and a typical dual-wing shutter, as one goes above 16fL, the perceived flicker, and its annoyance, goes up geometrically. You also have the limitations of the film density after it goes through the entire OCN -> release print chain.

One of the things that ShowScan brought to the table, with 70mm and 60fps frame rates...was also higher white levels...something often not mentioned in its benefits. Even Todd-Ao had 30fps to cut down on the flicker and also, with the 70mm aperture, get more light to the screen.

In the digital world, particularly at the start of DCinema...the target for center brightness was just 12fL as that was deemed "equivalent to film's open-gate 16fL (remember, film was not measured with film in the gate...so one has to factor in processed film's lowest density and what light that is). The realities were...12fL was all the "Mark-VII" projectors (e.g Barco DP40, DP50 and Christie's equivalent slanted projectors) could muster, light wise and they had NO black levels...they were grey with a 12fL full-white. 14fL came about with the Series-1 projectors that had a nominal sequential contrast ratio of 2000:1 (if everything is going their way). That first generation Series 1 projectors were not as efficient with light as the series 2 projectors (all else being equal). We then regressed a bit with the 4K and S2K (.69") chips as their contrast ratios were between 1650-1850, if you are being honest about them. So, you really can't raise the white level because it reveals just how bad their black levels are (and some of those S2K machines are hideous on how poor they are with black levels).

With laser light (RGB more so than Laser-phosphor), we can get contrast levels up on 4K (and the S2K) so seeing, out-of-the-box 4K projectors with contrast ratios at 2300:1 and above are achievable. With just a lens swap, Barco can often bump up to 3000:1 and will a full light path improvement, closer to 5000:1. Christie, with just lens swaps is claiming up to 6000:1. Mind you, there are light hits (substantial) by using these higher-contrast lenses/light paths. However, as 3000:1 and above seem achievable with RGB laser light sources, it could be possible to bump up the white point to something closer to 19fL (65 cd/m²). However, before something like that could happen, you'd need a significant amount of systems ready to do it AND you'd need different versions of the movie (like Avatar 2 where there were something like 84953476 versions) as the vast majority of theatres will still need the 14fL version (and, seriously, the lamp based S2K projectors should get more like 12fL versions).

Very high contrast projectors, at the moment, are limited to Dolby Vision and are incredibly expensive.

What I see may be a compromise is a light-steering projector. Barco demoed one in the pre-C19 CinemaCon. The content used made it look a little gimmicky (like forcing HDR on your TV for SDR content)...but it did show that they could get a much higher white level while keeping deep blacks. For something like that to work...it will need to be priced well (and not at the crazy "Flagship Laser" levels) and content will need to be mastered for that technology. I think the average consumer will instantly recognize the benefits and won't have to "learn" to like it like the recent HFR and 3D HFR "experiments" are demonstrating.

[Carsten Kurz]

3P Lasers on silver screens create high levels of speckle. 6P systems are expensive - so expensive that manufacturers stop making them. The best bright 3D I have seen so far was with a Barco DP4K-60L on a matt white screen using color separation with the large glasses. Unfortunately, not manufactured anymore.

[James Gardiner]

Carsten, interesting.., I thought colour separation was the way to go with premium dual projection systems. I am not a fan of silver screens or the crosstalk of polarised lenses. (Especially when running bright, ghostbusting exists because of this)
From my experience, its active glasses or colour separation is the best path to 14Fl 3D experiences. (No silver screen required)

Personally I feel the success of Avatar: The way of water, is going to push us back to active 3D for smaller independent cinemas as it's easier to deal with them when under 100 people are leaving a cinema. AND, small independents cannot afford to drop the money they would have gotten if they DID have 3D, as most do not. (Or if they do, it's a dim bearly 3.5FL system)
I expect many are feeling a little worse for wear after this Christmas and Avatar: The way of water, experiencing much lower attendance levels as in my region, over 52% BO gross went to 3D screens, likely pulling substantial numbers away from those who did not have 3D.

We now have Avatar ep 3 and 4 on the way.