Tag Archives: silver screen

Black Screen Alert~! InterOp Losing Life Support

Long Live InterOp

It was the best of times, it was the worst of times. The engineers contributing to SMPTE, and the studios who contributed to DCI, came up with enough elements to create a secure and beautiful D-Cinema environment. The same studios financed the equipment qualification standards and partially financed equipment purchases for many exhibitors. These exhibitors agreed to buy this qualified equipment and use it in a way that somewhat assured that copyrights and quality-better-than-film would be typical on screens world-wide.

Fortunately, there were written and unwritten agreements which allowed the simple DCinema origins of MPEG and a fairly loose mechanism of security keys to transition to the full on (and just recently completed) versions of standards, specifications and practices known as SMPTE Compliant Digital Cinema, with SMPTE Compliant DCPs and Security and screen fulls of other ingredients. These transitional agreements are known as InterOp.

Unfortunately, InterOp worked well enough to be added to…and added to…and added to…

For example, the simplest multimedia tools use metadata to describe computer needed info and human interface info within the songs or movies that we get to and from iTunes and Hulu and Netflix. Workers who had to get equipment and people working together in the InterOp world had to come up with an interim…maybe one year or so to live…Naming Convention. It wasn’t useful for computers at all, and cumbersome for humans at best and kept getting added to without increasing the number of characters since some old equipment only had so many display characters…kinda like computers in the 60’s. There were (and are, since years later it is still in use) dozens of ways for it to go wrong, beginning with the fact that some studios chose to ignore it when it gets in the way (according to the logic at their end of the string) while projectionists might miss some nuance that is needed for logic at their end of the string.

What happened to adding metadata like modern sciences do, and which everyone knows eventually will be needed? There are other panics with higher priority. It sits partly formed, probably until it becomes a keystone item needed for some other important development.

There are other examples of InterOp and loose de facto ‘standards’ living beyond their time, the most garish being what is hopelessly called 3D.

Instead of using valuable engineering time to progress the computer to computer interface and give exhibitors a fighting chance at perfection, engineers have had to shoehorn one feature after another into the InterOp structure. It is done with the best intentions, of course. It begins with, “My customers were asking for this now, not at some point in the SMPTE-Compliant future.” It ends with, “I have to do this because my competitor is bragging about how they can do this at no extra cost even though it violates the spirit and the essence of every standard.”

There are too many examples to mention ranging from forensics and audio mapping. Specifics aren’t as important as the fact that the entire industry has floated out far enough from land that some see letters in the water, and some seem to think that they spell H – E – R – E    B – E    D – R – A – G – O – N – S

DCinema Dragons don’t breathe fire. They are light suckers. They cause Dark Screens. Coming to theaters and drive-ins near you.


Why?

Many reasons, partly centered around the effects of software upgrades. Because the upgrade from InterOp to SMPTE-Compliant software is not a simple ‘add a feature or two’ software upgrade. At the best of times, you just never know what you will be causing when you hit that ‘Upgrade’ button. Did the software writer anticipate every single parameter of combinations of hardware and software that is in your situation?

There just are some odds that you come out of the hospital feeling worse than how you went in (look up HAI). Anyone with a computer has had software upgrades that worked for thousands of others, but did not work for them (look up: damn, not again.) There is probably some inverse squared proportionality involved as well. Getting closer to a deadline quadruples the odds of failure.

So, don’t change~! Jeez. That is sooo obvious. Which is what many do. Don’t get the first generation of anything, including upgrades. Especially during summer when all the big movies are playing.

But a horizon event approaches. Some InterOp juggling just won’t work for some combinations of . There are an amalgam of changes coming though, prompted by the teams of Jackson and Cameron. It might be easy to ignore the 60 frames per second requirement of a Cameron release (famous for pushing deadlines forward as he is), but The Hobbit will probably not be delayed. 48 frames per second, stereoscopic 3D. Will it work in the InterOp world? And what other changes will be made

Why 48fps? Phil Oatley, the post group head of technology from Park Road Post (Mr. Jackson’s facility in New Zealand) who spoke at the SMPTE/NAB DCinema Days last April said that they choose 48 because they didn’t know if equipment and exhibitors could change to 60fps in time and in significant numbers. As it turns out, all server and projector manufacturers have announced 48 and 60 fps capability. Sony even put a price on it…$3,000…which they can more easily do for their 13,000 users as they have always used an internal media block in their system.

In this case, Sony has something like the Apple advantage: They control the server, the media block and the projector so the odds are higher of getting a smooth transition. And, they have gotten DCI Compliance (at one moment of software version time…does HFR cause enough of a technology disruption that they need to re-certify?)

A TI-based projector with an SD-HDI interface will be a lot more complicated. An IMB (internal media block) needs purchasing and inserting, which isn’t a cheap investment. It is dependent upon TI-code and code from the projector manufacturer as well as code from the server all working together. How different is the server, which will have had its graphics-serving guts ripped out? …will that need a new cert? Check the DCI site for Compliance passed equipment.

But we have gotten off point. Back a few years ago you could sign a VPF deal and promise that you would use DCI-Compliant equipment and run with the latest SMPTE specs and recommended practices. At the time there wasn’t one piece of gear through the compliance procedures. And since you know that there is no SMPTE Police checking your screen for the required 48 candela/square meter luminance standard, you didn’t feel bad breaking the luminance number when showing 3D, a number that approached moonlight-equivalence at the sides of the theater and barely reached 10cd/m2 in the center. (For info on the light fall off from silver screens, see: 23 degrees…half the light. 3D What?)

But the history of the studios has been to look the other way until there is a technology that fulfills the DCI requirement. When Doremi proved they could do JPEG as the standard required, MPEG suppliers were given notice. When laser light engines can provide 3D at 48 cd/m2 (14 foot-lamberts), will the studios insist that passive 3D systems with their horrid high gain silver screens are no longer allowed (as was done in France recently? See: The Death of Silver Screens~! Vive la France)

We’ll see, but this doesn’t have anything to do with HFR. HFR is outside the DCI specs. It falls into the ‘no less than’ zone, similar to the color primaries. Laser suppliers can pick primaries outside the capabilities of xenon if that is financially and politically worthwhile, just as long as they don’t chose primaries inside the DCI/SMPTE limits.

So what does HFR and SMPTE compliance have to do with each other? Only that they are two locomotives that are running on two separate but not parallel lines. There is no firm deadline for SMPTE compliant DCPs, and no one is saying that InterOp compliant DCPs have a limited life. In fact, the studios expect that DCI equipment will play future SMPTE-compliant DCPs as well as what will become ‘legacy’ InterOp DCPs.

But something, at some time, is going to bulge the balloon of InterOp to the point that going SMPTE-Compliant is the logical move. Engineers at the manufacturers are just going to say, “I can’t play this game anymore. We were promised SMPTE would be the container that fit everything, I did the work, I will InterOp no more.”

There is rumor that this will happen soon. There is a particular setup that is rubbing against the InterOp balloon. Exhibitors are saying, “We don’t want to change until the summer season is over.” Will everything play nice together if only one condition is changed in a system? Possibly. How can you increase your odds?

Go to the ISDCF site that lists all the latest software/firmware versions for the equipment in the field. See to it that you have the latest. That will increase the odds. ISDCF Current Versions

Another thing you can do is prepare a database listing all of your equipment at each projection position, all of the software and firmware versions and all the serial numbers, and leave a field where you can download your .pem file from each piece of gear. Save this and get ready for a note from your distribution center asking for this info.

 

It was the best of times, it was the worst of times,
it was the age of wisdom, it was the age of foolishness,
it was the epoch of belief, it was the epoch of incredulity,
it was the season of Light, it was the season of Darkness,
it was the spring of hope, it was the winter of despair,
we had everything before us, we had nothing before us,
we were all going direct to heaven, we were all going direct the other way
– in short, the period was so far like the present period,
that some of its noisiest authorities insisted on its being received, for good or for evil,
in the superlative degree of comparison only.

Charles Dickens – Tale of Two Cities

Black Screen Alert~! InterOp Losing Life Support

Long Live InterOp

It was the best of times, it was the worst of times. The engineers contributing to SMPTE, and the studios who contributed to DCI, came up with enough elements to create a secure and beautiful D-Cinema environment. The same studios financed the equipment qualification standards and partially financed equipment purchases for many exhibitors. These exhibitors agreed to buy this qualified equipment and use it in a way that somewhat assured that copyrights and quality-better-than-film would be typical on screens world-wide.

Fortunately, there were written and unwritten agreements which allowed the simple DCinema origins of MPEG and a fairly loose mechanism of security keys to transition to the full on (and just recently completed) versions of standards, specifications and practices known as SMPTE Compliant Digital Cinema, with SMPTE Compliant DCPs and Security and screen fulls of other ingredients. These transitional agreements are known as InterOp.

Unfortunately, InterOp worked well enough to be added to…and added to…and added to…

For example, the simplest multimedia tools use metadata to describe computer needed info and human interface info within the songs or movies that we get to and from iTunes and Hulu and Netflix. Workers who had to get equipment and people working together in the InterOp world had to come up with an interim…maybe one year or so to live…Naming Convention. It wasn’t useful for computers at all, and cumbersome for humans at best and kept getting added to without increasing the number of characters since some old equipment only had so many display characters…kinda like computers in the 60’s. There were (and are, since years later it is still in use) dozens of ways for it to go wrong, beginning with the fact that some studios chose to ignore it when it gets in the way (according to the logic at their end of the string) while projectionists might miss some nuance that is needed for logic at their end of the string.

What happened to adding metadata like modern sciences do, and which everyone knows eventually will be needed? There are other panics with higher priority. It sits partly formed, probably until it becomes a keystone item needed for some other important development.

There are other examples of InterOp and loose de facto ‘standards’ living beyond their time, the most garish being what is hopelessly called 3D.

Instead of using valuable engineering time to progress the computer to computer interface and give exhibitors a fighting chance at perfection, engineers have had to shoehorn one feature after another into the InterOp structure. It is done with the best intentions, of course. It begins with, “My customers were asking for this now, not at some point in the SMPTE-Compliant future.” It ends with, “I have to do this because my competitor is bragging about how they can do this at no extra cost even though it violates the spirit and the essence of every standard.”

There are too many examples to mention ranging from forensics and audio mapping. Specifics aren’t as important as the fact that the entire industry has floated out far enough from land that some see letters in the water, and some seem to think that they spell H – E – R – E    B – E    D – R – A – G – O – N – S

DCinema Dragons don’t breathe fire. They are light suckers. They cause Dark Screens. Coming to theaters and drive-ins near you.


Why?

Many reasons, partly centered around the effects of software upgrades. Because the upgrade from InterOp to SMPTE-Compliant software is not a simple ‘add a feature or two’ software upgrade. At the best of times, you just never know what you will be causing when you hit that ‘Upgrade’ button. Did the software writer anticipate every single parameter of combinations of hardware and software that is in your situation?

There just are some odds that you come out of the hospital feeling worse than how you went in (look up HAI). Anyone with a computer has had software upgrades that worked for thousands of others, but did not work for them (look up: damn, not again.) There is probably some inverse squared proportionality involved as well. Getting closer to a deadline quadruples the odds of failure.

So, don’t change~! Jeez. That is sooo obvious. Which is what many do. Don’t get the first generation of anything, including upgrades. Especially during summer when all the big movies are playing.

But a horizon event approaches. Some InterOp juggling just won’t work for some combinations of . There are an amalgam of changes coming though, prompted by the teams of Jackson and Cameron. It might be easy to ignore the 60 frames per second requirement of a Cameron release (famous for pushing deadlines forward as he is), but The Hobbit will probably not be delayed. 48 frames per second, stereoscopic 3D. Will it work in the InterOp world? And what other changes will be made

Why 48fps? Phil Oatley, the post group head of technology from Park Road Post (Mr. Jackson’s facility in New Zealand) who spoke at the SMPTE/NAB DCinema Days last April said that they choose 48 because they didn’t know if equipment and exhibitors could change to 60fps in time and in significant numbers. As it turns out, all server and projector manufacturers have announced 48 and 60 fps capability. Sony even put a price on it…$3,000…which they can more easily do for their 13,000 users as they have always used an internal media block in their system.

In this case, Sony has something like the Apple advantage: They control the server, the media block and the projector so the odds are higher of getting a smooth transition. And, they have gotten DCI Compliance (at one moment of software version time…does HFR cause enough of a technology disruption that they need to re-certify?)

A TI-based projector with an SD-HDI interface will be a lot more complicated. An IMB (internal media block) needs purchasing and inserting, which isn’t a cheap investment. It is dependent upon TI-code and code from the projector manufacturer as well as code from the server all working together. How different is the server, which will have had its graphics-serving guts ripped out? …will that need a new cert? Check the DCI site for Compliance passed equipment.

But we have gotten off point. Back a few years ago you could sign a VPF deal and promise that you would use DCI-Compliant equipment and run with the latest SMPTE specs and recommended practices. At the time there wasn’t one piece of gear through the compliance procedures. And since you know that there is no SMPTE Police checking your screen for the required 48 candela/square meter luminance standard, you didn’t feel bad breaking the luminance number when showing 3D, a number that approached moonlight-equivalence at the sides of the theater and barely reached 10cd/m2 in the center. (For info on the light fall off from silver screens, see: 23 degrees…half the light. 3D What?)

But the history of the studios has been to look the other way until there is a technology that fulfills the DCI requirement. When Doremi proved they could do JPEG as the standard required, MPEG suppliers were given notice. When laser light engines can provide 3D at 48 cd/m2 (14 foot-lamberts), will the studios insist that passive 3D systems with their horrid high gain silver screens are no longer allowed (as was done in France recently? See: The Death of Silver Screens~! Vive la France)

We’ll see, but this doesn’t have anything to do with HFR. HFR is outside the DCI specs. It falls into the ‘no less than’ zone, similar to the color primaries. Laser suppliers can pick primaries outside the capabilities of xenon if that is financially and politically worthwhile, just as long as they don’t chose primaries inside the DCI/SMPTE limits.

So what does HFR and SMPTE compliance have to do with each other? Only that they are two locomotives that are running on two separate but not parallel lines. There is no firm deadline for SMPTE compliant DCPs, and no one is saying that InterOp compliant DCPs have a limited life. In fact, the studios expect that DCI equipment will play future SMPTE-compliant DCPs as well as what will become ‘legacy’ InterOp DCPs.

But something, at some time, is going to bulge the balloon of InterOp to the point that going SMPTE-Compliant is the logical move. Engineers at the manufacturers are just going to say, “I can’t play this game anymore. We were promised SMPTE would be the container that fit everything, I did the work, I will InterOp no more.”

There is rumor that this will happen soon. There is a particular setup that is rubbing against the InterOp balloon. Exhibitors are saying, “We don’t want to change until the summer season is over.” Will everything play nice together if only one condition is changed in a system? Possibly. How can you increase your odds?

Go to the ISDCF site that lists all the latest software/firmware versions for the equipment in the field. See to it that you have the latest. That will increase the odds. ISDCF Current Versions

Another thing you can do is prepare a database listing all of your equipment at each projection position, all of the software and firmware versions and all the serial numbers, and leave a field where you can download your .pem file from each piece of gear. Save this and get ready for a note from your distribution center asking for this info.

 

It was the best of times, it was the worst of times,
it was the age of wisdom, it was the age of foolishness,
it was the epoch of belief, it was the epoch of incredulity,
it was the season of Light, it was the season of Darkness,
it was the spring of hope, it was the winter of despair,
we had everything before us, we had nothing before us,
we were all going direct to heaven, we were all going direct the other way
– in short, the period was so far like the present period,
that some of its noisiest authorities insisted on its being received, for good or for evil,
in the superlative degree of comparison only.

Charles Dickens – Tale of Two Cities

Silver Screens – French Quality Officially Declines?

The CNC is the group who sets the the “AFNOR” standards for French cinema (and TV). This is the group who wrote in 2007 that once the SMPTE specifications for digital cinema technology are finalized that they “would have the force of law” in France and on the international level. (See: Force of Law; France Issues DCinema Document)

And yet, this very group is about to allow ‘silver screens’ to be within the actual standard with the reasoning; that commercially, silver screens have become a de facto standard. This is the equivalent of allowing the old 625-line/50 field per second interlaced video as an acceptable HD TV standard because there are so many TVs of that nature in the field. The reality is that silver screens – actually, aluminum painted screens – for passive 3D systems (the ‘cheaper’ RealD and MasterImage systems) cannot provide the mandated levels of light except to perhaps 10 seats in the theater. The fall-off in quality from using silver screens is the number one worst aspect of 3D presentations. And when a 2D movie is played on them, the results are a disaster: dark and gray patches, diminishing light to every seat except the one “measured” position, screens that cannot be cleaned, and which get darker in fewer years than the normal screen.

You can see the reply by several French cinema organizations at the following link: Respecter la lumière dans les salles obscures

Roughly translated that is: Respect the light in the darkened theater. 

The players are no slouches. They are the technical organizations of every stage of the movie making process. It seems that they are respectful, but they are forceful. What they didn’t say was that was that the CNC should be embarrassed. What they didn’t say was that the artist and technology groups will create a grass roots education plan if they continue with the abdication of standards in the theater for financial reasons. 

We are looking for a better english translation, but until we have one, this will have to suffice:

Respect the Light in Darkened Theater 22 June 2011

The CNC is about to sign a decision to amend the technical specifications – Requirements for Registration – of cinema theaters. The proposed text refers to the French standard for digital cinema, but with the “amputation” of the article about the difference in luminance (point 5.1.2 of the standard 27100).

That article provides that the difference in luminance between the brightest spot and the lowest point of a bright image on a screen can not exceed 25%. This aspect of the standard is essential. It guarantees the respect of light and contrast of the image desired by the director and the cinematographer and allows that all the spectators in the same room see the same film.

The exception to this standard is clearly intended to allow the spread of “silver screens” designed primarily for 3D projections. At this time, three large French circuits equip themselves large amounts of these screens …

In the event that a classic 2D projection is performed on a silver screen, over 80% of the audience of a room would not see the film in the technical and with the ‘artistic intention’ that the directors, producers and technicians have designed.

This proposal is a dramatic decline in the quality and respect of the works: we sacrifice the quality of films, and the moral rights of authors, on behalf of the immediate profit potential of 3D, and seriously risk the experience in the auditorium.

We request of the CNC:

• to reverse its decision and enforce the entirety of the standard French NF S27100 particularly on point spreads luminance, taken by the international standard ISO digital cinema,

• to define together the means to monitor the implementation of this standard after a systematic inspection of the CST.

Signed

ACID Fabienne Hanclot 01 44 89 99 71 / 

AFC Carolina Champetier 01 42 64 41 41 /

ARP Florence Gastaud 01 53 42 40 00 / 

CST Laurent Hebert 01 53 04 44 00 /

the FICAM Hervé Chateauneuf 01 45 05 72 47 / 

SACD Agnes Mazet 01 40 23 45 11 /

SRF’s Cyril Seassau 01 44 89 99 65

 


 

For our earlier articles on the flaws of using silver screens see: 
Scotopic Issues with 3D, and Silver Screens
23 degrees…half the light. 3D What?  

23 degrees…half the light. 3D What?

Sillver Screen Light Failure Point3D Luminance Issues—Photopic, barely. Mesopic, often. Scotopic? Who knows…? 

We don’t mean to be picking on the good people at Stewart Film Screens by making an example of their Silver Screen light rolloff curve. They just happen to grace us with the most usable graphic description of what is happening to our light. Looking at Harkness Screens Data Sheet for Spectral 240 3D Screens is not better and may be worse. 

We know the problems of getting light to the eyes for any of the available 3D systems. The initial filter eats up to 50% of the light from the projector, plus the manner of each eye getting turned off 50% of the time, and the darkness of the glasses all steal a lot of light. If the projectors could produce enough light to overcome all these transmission problems…which they generally can’t…it would just mean more burnt expensive bulbs and higher electricity costs. 

But even if the exhibitor cranks it as best as possible, and tweaks the room to get the best RGB balance at the best seats of the house, if the auditorium is using a ‘silver’ screen to maintain the polarity of the RealD or MasterImage system, the patron who is 23 degrees off the center-axis will have half the light available. Put another way, as you can see from the full picture at the Stewart site, 3 seats away from center is a totally different picture…as is the 4th and 5th, etc., as the situation just gets worse. 

If the cinema had achieved 5 foot Lamberts (17 candela/m2) behind the glasses (most don’t get 3ftL – 10c/m2), then 3 seats off center will be 2.5ftL (8.5c/m2). At this point, bright reds have all turned to brick red or darker, and blues are becoming relatively dominant – it isn’t that there are fewer yellows or greens in the picture – it is that the eye becomes better able to discern the blue in the mix. (Another way to describe what is known as the Purkinje shift is that an object that appears greenish-yellow in brighter light will appear to be greenish blue as the intensity of the light descends lower than below 10 candelas/m2.) Combine that with stray light from a few EXIT signs, which not only mess with the contrast but puts non-symetrical data into the normally “practically-” symmetrical 3D mix, plus some reflections in the back of the eyeglasses and the patrons should not wonder why they don’t universally have an enjoyable experience. 

We won’t beat this into a pulp since most real-life scenarios just get worse.

What will make it better?

Consumer education to begin, which is the real excuse for this article. Patrons must know what to insist upon. 

Projectors can’t generate enough light to get 3D up to the 14ftL (48 candelas/m2) that 2D movies are shown at. But the new Series II projectors can do ‘more’ and industry tests show that ‘more’ is better, especially if the original was ‘mastered’ to be shown at ‘more’. James Cameron was prepared to ship theaters a ‘print’ of Avatar that was mastered at hotter levels for cinemas who asked for it…up to 10 ftL! Patrons must insist that if they are paying more for the experience, they should get better…perhaps 10ftL is not going to be the standard this year, but 7 or 8? Grass roots effort anyone? The studios set the intention in the DCI spec at 14, so one would think that they will come to the plate with ‘more’ if asked. [DCI Specification 1.2; page 48…and tell them that you want an order of Uniformity and some of that ±4 Delta E while they’re at it.] 

The future also holds at least two potential ways that will give a better picture. Brian Claypool at Christie points out that one of the features of the Series II projector is “more native support for faster frame rates.” For example, many people in the creative community believe that higher frame rates will do more for image quality than having more pixel resolution. Again, Brian Claypool, “Do you remember how rich every frame was in Avatar, that your eye just kept wanting to look around? Well, imagine having 2 times as many frames for your eyes to follow… it will feel like looking out a window on another reality”.

{youtube}WgrdKmgniGI{/youtube}

The other, also long-term, change is replacing bulbs with lasers in the projectors. Good news on that front was announced by one player, Laser Light Engines. We deconstructed their newest announcement and some of their potential at: Laser Light Engines gets IMAX funding—Putting Light on the Subject

Some mark this as digital cinema’s 11th year, but it wasn’t until 6 years ago that 2K was delivered, an example of the evolution of this industry. 

Links: Luminance Conversion Table

Scotopic Issues with 3D, and Silver Screens

Knoting Laser Light