Film-Tech Cinema Systems
Film-Tech Forum


  
my profile | my password | search | faq & rules | forum home
  next oldest topic   next newest topic
» Film-Tech Forum   » Operations   » Digital Cinema Forum   » Projector Imager Areas and DCI Compliance (Page 1)

 
This topic comprises 2 pages: 1  2 
 
Author Topic: Projector Imager Areas and DCI Compliance
Brian Guckian
Jedi Master Film Handler

Posts: 594
From: Dublin, Ireland
Registered: Apr 2003


 - posted 02-06-2009 09:45 AM      Profile for Brian Guckian   Email Brian Guckian   Send New Private Message       Edit/Delete Post 
I have a query relating to some d-cinema installations that are using imager areas that do not appear to be supported or endorsed by the current DCI Digital Cinema Specification (V 1.2, March 7, 2008).

I first became aware around two years ago of a practice whereby in certain installations, all ratios were being referenced to the 858 x 2048 "flat 'scope" imager area.

Narrower ratios were being fitted into this area, so that for example, 1.85 is displayed at 858 x 1587 and Academy at just 858 x 1180.

This problem seems to originate where, for budgetary reasons, non-motorised zoom lenses are installed for common height screen setups. The lens is set to one position, and because of this, everything is referenced to the 858 x 2048 'scope area.

The DCI Specification seems to be quite clear about how much of the imager should be employed for content display, viz.:

It is intended that the projector project the full horizontal pixel count or the full vertical pixel count of the image container (Section 8.2.2.7., Spatial Resolution Conversion, page 80)

My questions are: has anyone else encountered this? And, are such installations consequently non-DCI compliant?

And even if they are compliant, is there now a SMPTE Standard (given that DCI is a specification rather than a Standard), that clearly defines what the utilised imager areas should be for the different ratios?

 |  IP: Logged

David Zylstra
Master Film Handler

Posts: 432
From: Novi, MI, USA
Registered: Mar 2007


 - posted 02-06-2009 03:37 PM      Profile for David Zylstra   Email David Zylstra   Send New Private Message       Edit/Delete Post 
We have a Cinedigm solution and Christie has since come through and installed their WCL lens on all of our systems (except for RealD auditoriums due to some issue with the Z-Screen mount that was made for the lens changer).

The additional lenses were done at no charge to us since contractually we are owed a DCI compliant system and the studios clarified how they interpreted DCI in regard to lensing.

The practice was started because the original DCI mentioned a "single lens solution" and the language quoted above was not introduced until later (and there was not a good cost effective solution from one or more manufacturers, not to mention some liberal interpetation)

There is still ambiguity when interpreting "It is intended...." because it does not say "It is required....".

 |  IP: Logged

Brian Guckian
Jedi Master Film Handler

Posts: 594
From: Dublin, Ireland
Registered: Apr 2003


 - posted 02-07-2009 09:22 AM      Profile for Brian Guckian   Email Brian Guckian   Send New Private Message       Edit/Delete Post 
Thanks for the info David, accessories like the WCL are clearly a good, cost-effective solution to the problem.

A concern though, is what if an exhibitor or third-party provider doesn't want to pay even for a MALM-style mount, never mind a focal-length adapter?

DCI is consistent throughout in specifying the 2.39 ratio as 858 x 2048 and 1.85 as 1080 x 1998, for 2K d-cinema.

Although one could argue that sections of the DCI document, such as Section 3.2, deal only with the specification of, to quote, "uncompressed image structures and files" (rather than display of such structures and files), nevertheless the overall thust seems to be to ensure that:

"The number of active pixels...shall extend to the maximum in either the horizontal or vertical direction of the defined level of operation as defined in Table 1" [4096 horizontal and 2160 vertical for 4K; 2048 horizontal and 1080 vertical for 2K] (DCI V1.2, pages 26 & 27).

The other point is that the phrase "single-lens solution" never precluded the use of motorised zoom lenses so that, in a common height screen setup, ratios could be displayed using as much of the imager area as possible. In other words "single-lens solution" did not mean that say, only a prime lens could be used.

Also, if you read the rest of the text from DCI V1.2 page 80 from which the initial quote was taken ("It is intended that the projector project the full horizontal pixel count or the full vertical pixel count of the image container"), it is clear that it is a reinforcing, explanatory sentence, that doesn't dilute the specification.

A published Standard would seem to be the best way of preventing the use of compromised imager areas in the future.

 |  IP: Logged

Steve Guttag
We forgot the crackers Gromit!!!

Posts: 12814
From: Annapolis, MD
Registered: Dec 1999


 - posted 02-07-2009 10:35 AM      Profile for Steve Guttag   Email Steve Guttag   Send New Private Message       Edit/Delete Post 
I've now done it with Anamorphics, WCLs and just good old fashioned zooms. Any interpretation of the DCI that does not have the original pixel count of the source file is missing the point. That is, the image is to be displayed with as little manipulation as possible. Mere magnification is not manipulation.

Anamorphics do manipulate the image (vertical streching). However, that is a trade off. While the actual resolution does not increase, the anamorphic provides the ability to make more effective use of light. I've found that 1.26x anamorphic to be substantially less intrusive than the 1.5 and 1.9x anamorphics of the 1.3K machines. In fact, I just did a curve screen theatre with an anamorphic on "Scope" and it looked rather good with well defined pixels in either format from corner to corner.

One thing I wish DCI would allow is asymetric scaling to deal with keystones. Also, I hope that lens manufacturers bring about options for dealing with curved screens a bit better. Presently...you just wack off the pixels. I my recent installation, it is a digital beside an existing film system. The Film system does not throw away nearly as much image on the curved screen as the digital. This is because the film lenses were carefully chosen to pincushion the image out of th projector so it "fit" the curved screen. Why is the screen curved so much? Why to achieve even light on a gain screen...and that part works too.

Steve

 |  IP: Logged

David Zylstra
Master Film Handler

Posts: 432
From: Novi, MI, USA
Registered: Mar 2007


 - posted 02-07-2009 11:13 AM      Profile for David Zylstra   Email David Zylstra   Send New Private Message       Edit/Delete Post 
Page 26 & 27 deal mainly with the image structure in the DCDM so I don't think it can absolutely be applied to the projected picture.

Page 80 8.2.2.7 deals with 2 issues - scaling of content from 2k to 4k and projecting the image onscreen. Since the second paragraph starts out by saying "Should electronic image resizing or scaling be used to support constant height or constant width..." the door is open to scaling and the "intended" statement implies that projecting full pixel count is preferred but not implicitly required. The "intended" statement was not in the original 1.0 spec that the initial systems were installed under.

I think on some level the scaling solution was a stopgap while the WCL was developed (at least for Christie), but I'm sure some hoped it would be permanently allowed so they would not have to pay to retrofit the extra lenses on the 3000 Cinedigm screens. I only have experience with the Cinedigm installs so I don't know what TDC or Kodak were/are installing for lenses.

Don't get me wrong, I agree that we should project each format in its native pixel count - and as far as I know the studios are requiring it (as evidenced by the addition of the WCL).

Once exhibitors start buying their own equipment (and VPFs are gone) I'm sure there will be arguments about lens configurations - after all, the exhibitors currently have the financial choice now on how to configure 35mm (i.e. studios will still book 2:1 formats, poorly lit pictures, bad lenses, etc). If you read the NATO DC system requirement it talks about the end display being the exhibitors choice (i.e. scaling, anamorphics, WCL, etc) - I think the end result of no scaling may be more fixed width screens in new buildings.

It will be interesting to see where this goes once VPFs are gone (some people I've talked with think the studios are going to eventaully allow use of cheap off the shelf HD video projectors - I certainly hope this does not happen).

 |  IP: Logged

Mark Gulbrandsen
Resident Trollmaster

Posts: 16657
From: Music City
Registered: Jun 99


 - posted 02-07-2009 11:38 AM      Profile for Mark Gulbrandsen   Email Mark Gulbrandsen   Send New Private Message       Edit/Delete Post 
In theory the anamorphic mode on the DMD's would give more verticle resolution but less horizontal resolution. Going with a WCL or new Zoom lens would give more horizontal and less verticle rez. Percentage wise both scope versions of the active area provide just about the same number of operating pixels and hence just about the saem rez! An install I just did with one of the new Konica-Minolta lenses that has a bit more zoom ratio is clearly the best route to go as it eliminates the need for the WCL or anamorphic all together and there's alot less glass to go through noticably resulting in much higher contrast ratio AND sharpness. This location WAS actually set up anamorphic(ISCO). There are occasional artifacts caused by the rectangluar shape of the pixels themselves(which was the main reason DCI nixed anamorphics) as well as a bit of flare that was easily and for the first time observed by me while watching Gran Torino anamorphically... which has some very contrasty scenes. Gran Torino looked far better with the new zoom and running 2048 X 858.

Steve, If you need another anamorphic on the cheap let me know...

Mark

 |  IP: Logged

Brian Guckian
Jedi Master Film Handler

Posts: 594
From: Dublin, Ireland
Registered: Apr 2003


 - posted 02-07-2009 12:17 PM      Profile for Brian Guckian   Email Brian Guckian   Send New Private Message       Edit/Delete Post 
Well, no matter how you cut and dry it, there are images out there being displayed at 858 x 1587, and it looks like there's nothing to stop it.

Admittedly, it's easy to fall into the trap of associating image quality purely with pixel count, when perceived resolution is also a function of contrast and image stability.

So - on a small screen, with rock-steady stability and a very high contrast ratio, a 1.85 film displayed at 858 x 1587 may well "look" acceptable, in a subjective way.

But is it? If you are a top-line Cinematographer, would you accept your work projected this way, on a 2K digital system?

 |  IP: Logged

Tony Bandiera Jr
Film God

Posts: 3067
From: Moreland Idaho
Registered: Apr 2004


 - posted 02-07-2009 02:54 PM      Profile for Tony Bandiera Jr   Email Tony Bandiera Jr   Send New Private Message       Edit/Delete Post 
Oy, more proof of how fucked up D-Cinema is, they can't even agree on STANDARDS for image size!!!

For chissakes, what is it gonna take for the idiots who are pushing this to pull their collective heads out of their asses?

Film has had established STANDARDS for aspect ratios for many years!!

D-Cinema is supposed to be superior to film and home video, but like home video and the authoring of dvd's they can't even get the bloody image sizes to conform to a standard!!!

Why doesn't SMPTE step into this and FORCE a standard?

 |  IP: Logged

Steve Guttag
We forgot the crackers Gromit!!!

Posts: 12814
From: Annapolis, MD
Registered: Dec 1999


 - posted 02-07-2009 02:56 PM      Profile for Steve Guttag   Email Steve Guttag   Send New Private Message       Edit/Delete Post 
I use the anamorphics for the light, not the resoultion. If the source file is 2048x858...then that is all there is, period. Again, I've had good results...even on curved screens.

The WCL really just offers the equivalent of a zoom lens that otherwise doesn't have the range. One still looses light as pixels are splayed over the ceiling/floor. The anamorphic was not a stop gap until it was developed. It is nothing more than a 1.26x magnifier. Its simplicity allows it to be good at considerably less expense.

Schneider has a nifty concept where they put the "WCL" inside their prime lens.

Steve

 |  IP: Logged

Peter Castle
Expert Film Handler

Posts: 220
From: Wollongong University, NSW ,Australia
Registered: Oct 2003


 - posted 02-07-2009 08:48 PM      Profile for Peter Castle   Email Peter Castle   Send New Private Message       Edit/Delete Post 
All this discussion seems to me to herald major problems with DCI.

Here in Australia, there's a VERY slow introduction to dCinema, mostly with 3d systems. Unfortunately our exchange rate to the US dollar has dropped dramatically, so that from near parity last year, everything from the States now costs double. So we're looking at A$150,000 for a dCinema system. But there's still discussion about VPFs here, and things may improve - although the current global economic crisis doesn't bode well for an early recovery.

The use of non-DCI compliant projectors, at least in the independent cinema area, has developed a strong local eCinema market. Distributors not aligned to the US majors are using digital distribution to boost their boxoffice very effectively. "Slumdog Millionnaire" is one case in point. And the resolution of these digital copies are now often at full HD (1920 x 1080), although most of the cinemas are using 1.3K projectors to display.

If there's such discussion about projector imager areas, why don't DCI-supporting distributors consider allowing non-compliant projectors to be used?

I'm involved in a University cinema that shows once per week. I guess we're about as sub-run as you can get. Yet eCinema helps us get many quality films closer to release. And we'd love to be DCI but no distributor is going to support us with VPFs. But we do have fullHD projection just waiting to show major studio films.

 |  IP: Logged

Julio Roberto
Jedi Master Film Handler

Posts: 938
From: Madrid, Madrid, Spain
Registered: Oct 2008


 - posted 02-08-2009 10:07 AM      Profile for Julio Roberto     Send New Private Message       Edit/Delete Post 
DCI is messed up, which is what happens when you let an interested one-sided partner (Hollywood studios) decide what hardware and software manufacturer should do or not do. And they will even tell the owners of this equipment what they should or should not do. They are so jacked-up on security (for a change), they don't get the real matters (technical specs) right.

They, i.e., don't allow for technical improvements (i.e. TI DLPs being capable of better contrast currently), settling for fixed short-sighted standard.

All they care about is security and control of who is projecting what, at what time on what auditorium. If the could put a webcam inside each theater to count the attendance, I'm sure they would make it part of the requirement.

A POST phone line or, far worse yet, a broadband connection for EACH projector (server) requirement is bull-caca. In many countries, a phone line or ADSL is a $80 thing a month that it's TOTALLY uneeded as a single line can more than service an entire location.

And in many countries, a goverment regulated software ticketing system clearly reports every day to interested parties how much attendance each show had and how many times per day it was shown etc already, w/o the need for more keys or locks that can break one day and leave you with a missed show.

They keep delivering half-cooked requirements that are but compromises on their own quality goals. Now they don't seem to care about quality anymore but about making current equipment, whatever it is nowadays (i.e. triple-flash full res 3D, lens magnifiers, zooms, 4fl 3D screens, ... anything goes) even more "secure" and "good enough".

They made a mess, while neglecting important things.

How long until DCI 2.0?

They seriously need to let the technological market do their thing and just concentrate on the content delivery format standard. Let the hardware manufacturers make the most of it and hand out decryption keys to all the stuff that meets a minimun standard, period. It's none of their business to decide if an anamorphic should be used in my projector or not. If the result is right, the resolution is enough and the input format doesn't change, why should they care if I zoom-in, magnify, anamorph or use a 2-lens barrel?

I agree. SMPTE should take over the whole thing and require content to be opened (decryption keys given) to any manufacturer that can read and reproduce it to specs.

 |  IP: Logged

Mark Gulbrandsen
Resident Trollmaster

Posts: 16657
From: Music City
Registered: Jun 99


 - posted 02-08-2009 10:51 AM      Profile for Mark Gulbrandsen   Email Mark Gulbrandsen   Send New Private Message       Edit/Delete Post 
quote: Steve Guttag
I use the anamorphics for the light, not the resoultion. If the source file is 2048x858...
Figure both versions of active area for scope and you'll find they are pretty dang close to each other. 858 to 1080 is only a difference of 222 pixels or half that top and bottom. You'll probably loose 40 pixels off either if you're doing your screen masking using the Dolby screen masking file. You can now drop the $8,000.00 anamorphic or 2,000.00 WCL in favor of no anamorphic or WCL lens at all. Get more light and have a much sharper image or go to a smaller lamp in some cases. I too thought the anamorphic route was much sharper at first. The new lenses however proved me wrong. There is also a flare problem with the present incarnation of D-Cinema anamorphics as they have no internal light baffles(ditto for magnacoms) in them to stop internal relfections as Cinema anamorphics and primes do.

quote: Peter Castle
So we're looking at A$150,000 for a dCinema system.
About half that here now. You may be ahead to come over here and take delivery and dismantle the stuff down to parts and ship it back that way. Many times parts are much lower duty and far easier to export/import.

 |  IP: Logged

Brian Guckian
Jedi Master Film Handler

Posts: 594
From: Dublin, Ireland
Registered: Apr 2003


 - posted 02-08-2009 12:40 PM      Profile for Brian Guckian   Email Brian Guckian   Send New Private Message       Edit/Delete Post 
I think the general issue is also related to the fact that the DCI specification seems to have been designed around continuing use of Super-35mm for 'Scope films and common-width screens in theatres - a less than ideal format in the case of the former and incorrect practice in the case of the latter.

Does anybody know if filmmakers have experimented with mapping or scaling, at or prior to the DCDM stage, true anamorphic-originated images to the full 1080 x 2048 image container that is mentioned in DCI V1.2 Section 3.2.1.2. (Image Structure)?

Such images could then be unsqueezed using a 1.26X anamorphic lens, and one could then have the best of all worlds, as follows:

1.85 images displayed flat at 1080 x 1998

2.39 "flat 'scope" (Super-35mm) images scaled from 858 x 2048 to 1080 x 2048 and unsqueezed using the 1.26X anamorphic lens

2.39 "true 'scope" images displayed native at 1080 x 2048 and unsqueezed using the 1.26X anamorphic lens

This would use the full (or almost the full) area of the imager for projection of all formats, whilst also respecting the intent of the DCI specification. It might also be acceptable from the "stretched pixels" standpoint, since the expansion factor of 1.26X is relatively mild.

BTW I don't lay the problems I've described at DCI's door at all, and presumably SMPTE DC-28 are addressing the issue (of utilised imager area).

 |  IP: Logged

Julio Roberto
Jedi Master Film Handler

Posts: 938
From: Madrid, Madrid, Spain
Registered: Oct 2008


 - posted 02-08-2009 01:09 PM      Profile for Julio Roberto     Send New Private Message       Edit/Delete Post 
That would be too easy and too good.

I guess the line of thought here is that soon (virtually) no "true scope" is ever going to be used again.

And if (for practical purposes) 2K Super35-scope only has (about) 872 pixels of atteinable "resolution" in the DI, well, no point in upsizing.

I feel the industry moving to all digital 4K adquisition very soon, or 3perf Super35 those remaining on film. True scope is probably heading a slow but sure death.

But don't get me wrong. I'm all for a 2K 1.26x anamorph for TODAY. After all, they did use it in the previous generation of 1.3k digital projectors ...

I just wish they would just all move to 4K, where reduced vertical pixel count is not such a big issue and not using anamorphs allows sharp cheaper optics. The anamorph can give more light, but the lens elements themselves also take away some, so a bit of the advantage is loss there.

I would be happy with a square pixel 4096x1744 4K "scope" array even with the loss of 20% potential raw light and raw pixel count in widescreen films in exchange for the cheaper, sharper optics ...

Now we just need a 4K projector. Wait, we already have Sony. Now, seriously, we need ANOTHER 4K projector manufacturer.

Look at the current top box office offers: Taken (Digital Genesis 4K and 3perf super35), Coraline (digital cameras), Pink Panther (flat), Mall cop (Digital Genesis 4K), Push (super35), Slumdog (digital SI 2K and 3perf Super35), Torino (here is your true scope), Uninvited (flat), Hotel for Dogs (flat).

You get the point. Pbbly only 10%/20% of films are true scope anymore. Soon, digital will take over adquisition format too, as quality cameras with far (far) more true resolution than Imax (yes, you heard that right ... but let's leave that discussion for when we are past this 2K thing and the time arrives) come to market in just a year or two at prices unheard of.

Those that are sceptical remember you can have a 22 Megapixel "35mm" digital photo canon camera today in your house for $2500 or a 24 Megapixel Sony one for about the same price.

Or if you rather shoot 24fps CINEMA, you can BUY a 4.5K (12 megapixel, touted as 4.5K but truly equivalent to about 3.5K "DCI" specification, about 3400x1900 "true" full-color pixels, from the raw array of 4900x2580, 12bits, 4:4:4) for $17.500. Say another $3-$6k for a matching zoom lens. This is the PURCHASE price to shoot effectively, no compromises, 3.5K images into inexpensive flash memory media.

You would then need to downsize it to 2K to display them in a theater .... Welcome to brand new world of cutting-edge digital.

Movies like Bloody valentine or Jumper used them.

Panavision's Viper is a true 4K camera too, obtaining the image from a 6K subpixel array.

Those believing that Imax's huge frame equals an enormous quality ... are right, of course, but also haven't heard of the little hush experiment that Imax did. Better explained here by John Galt, Panavision Senior Vice President of Advanced Digital Imaging:

quote:
A number of years ago some IMAX engineers – and I don’t think IMAX ever let these guys out of their lab again -- did this wonderfully elegant experiment at the Large Film Format Seminar at Universal Studios Imax theatre. They showed this film they made that began with 2 rows of 2 squares: black white, white black, as if you had 4 pixels on the screen.

Then they started to double and double and double the squares. Before they got to 4K the screen was gray. Do you know what the means? There was no longer any difference between black and white, which is what allows you to see sharpness. It's the contrast that we see, not the actual information. Technically, the MTF (Modulation Transfer Function) was zero at 4K!

Let's just pretend for a moment that IMAX truly is 4K. You watch IMAX at between one and one and a half picture heights from the screen. But in order to get to appreciate 4K on a regular movie screen, you would have to sit much closer than normal. In other words, when you go to a movie theater, and most of the modern theaters with stadium seating are designed so that the middle of the theater is 2 ½ to 3 picture heights from the screen, for most of us who watch movies, that’s pretty where we want to be sitting. Maybe just a little bit closer from some of us who do this for a living, because we're maybe looking for artifacts or issues. If you sit much closer than 2 ½ picture heights, that's what you're seeing, artifacts, not movies!

I've run my own calculations and conclude that, generically speaking and in theory, for most auditoriums, 2K is indistinguishable or better than flat 1.85 projected film at 4 times the screen height (back rows of most auditoriums). Scope film is better than digital scope up to about 3 screens heights.

And 4K is indistinguishable or better than Imax at 2 times the screen height. Closer than 2 times the screen height, and Imax wins. So for all those not sitting in the first 3-6 rows or so, 4K *is* Imax (real 15perf 70mm Imax, I mean).

Imax huge frame size has reasons to be other than resolution, like support the light flux for such huge screens w/o melting or fading quickly.

4K at 2 times screen height is already at the about 36lp/degree of limit of the human eye of person with perfect vision ... so we just can't see more detail even if it was there. We would have to move closer to see it.

[ 02-08-2009, 02:29 PM: Message edited by: Julio Roberto ]

 |  IP: Logged

Steve Guttag
We forgot the crackers Gromit!!!

Posts: 12814
From: Annapolis, MD
Registered: Dec 1999


 - posted 02-08-2009 01:59 PM      Profile for Steve Guttag   Email Steve Guttag   Send New Private Message       Edit/Delete Post 
A popular misconception is that the SMPTE (or other engineering societies) have the power of law or some form of enforcement.

Just because the SMPTE codifies a standard does not require anyone to follow it. There is nearly 100-years of film projection to prove that.

I'm amazed at how much control the exhibitors have allowed the studios to have, thus far, in DCinema. They don't have this control in film, they shouldn't have it in DCinema. The DCinema conversion is at the studios' request, not the exhibitor's. As such, the studios should be forced to accommodate the exhibitors at every stage if they want it so much.

Steve

 |  IP: Logged



All times are Central (GMT -6:00)
This topic comprises 2 pages: 1  2 
 
   Close Topic    Move Topic    Delete Topic    next oldest topic   next newest topic
 - Printer-friendly view of this topic
Hop To:



Powered by Infopop Corporation
UBB.classicTM 6.3.1.2

The Film-Tech Forums are designed for various members related to the cinema industry to express their opinions, viewpoints and testimonials on various products, services and events based upon speculation, personal knowledge and factual information through use, therefore all views represented here allow no liability upon the publishers of this web site and the owners of said views assume no liability for any ill will resulting from these postings. The posts made here are for educational as well as entertainment purposes and as such anyone viewing this portion of the website must accept these views as statements of the author of that opinion and agrees to release the authors from any and all liability.

© 1999-2020 Film-Tech Cinema Systems, LLC. All rights reserved.