Wondering how the whole digital cinema vs analog is doing.

[Caleb Williams]

Leo,
As long as Kodak and a couple of others manufacture film, then analog will exist. Smart theater owners kept at least one or two film projectors in a given complex. Especially with all the 70mm that's been out the last 5+ years.

When Interstellar was offered in 70mm, we had one single screen theater that still had its projector in place and could've played it that way. But the way it was told to me at the time, was the cost of the print was exorbitant, and the labor to bring back one of the original projectionists to run those shows was also very expensive. It was eventually decided that it would be better for business to just make sure the digital system was up to spec, and run the digital prints for that movie.

One of the biggest gripes we've had is how much control the studios want of our locations. We're the biggest chain in Wyoming, but we're still small apples compared to the big guys. That means that for bookings at that one-screen house, we would be forced (by the studio) to play a film for 5 weeks. There aren't enough people in this town to keep a film above water for 2 weeks in a single screen theater. We've had two locations like that, that because of diminishing returns, and absolutely asenine studio booking windows we've had to close.

The multiplex's generally fare better simply because there are more auditoriums. But on the big releases, when Disney wants 8 out of 10 houses, it becomes a real struggle for our scheduler to do things in a way that targets optimum showtimes for the best margins. It really is stupid that they don't allow the theaters to cater to their individual markets a bit more, and instead have an iron grip on what we're allowed to play and sometimes when.

[James Gardiner]

vinyl records are still in use today and have a following. blind tests have shown that vinyl does not compare to digital. it's the aesthetic value that people enjoy and people will pay to listen or to watch their favourite formats. There's nothing wrong with that. Some people make money from them. Technically vinyl is not as good as digital and cellulite film , in general, is not as good as DCI digital cinema.

There is still money in catering to those who still appreciate the aesthetic value of those older formats.
in some ways it's a different business to showing mainstream Hollywood films. it's a niche that some cinemas choose to take advantage of. and good for them for doing so and keeping celluloid film and its historic importance alive for us to see today.

in these days off contracting attendance levels and cinemas looking for alternative and niche areas to attract more people. if you can do it it's not a bad option.

[Leo Enticknap]

I agree that preference for vinyl vs. comparable (meaning uncompressed LPCM and at a decent sampling and bit rate) digital has to be subjective and emotional, as does 35mm vs. DCI digital, all other factors being equal.

Some time before the covid lockdowns, a friend proposed the theory that subjective preference for vinyl likely came from the RIAA EQ curve. During covid, when I had a bit of time on my hands, I put that theory to the test. I identified a recording that I owned the vinyl LP of, very much liked the content, but was disappointed by the technical quality of the LP. This was Simon Preston playing Reubke's 94th Psalm sonata on the Westminster Abbey Organ. The LP is thin and flimsy, with the entire 27-28 minutes squeezed onto one side: the dramatic climax of the piece takes place in the last 3-4 minutes, which is the worst part of an LP side for audio quality, with most of the effect of the fiendishly difficult pedal line with 32' stops open sounding murky and indistinct, and much of the detail in the right hand being lost in HF crackle and clipping. So I bought the CD version, ripped it to a 44KHz .wav file, applied the RIAA curve to it in Diamond Cut, and output the result to another CD. Listening to both through the same amplifier and speakers, I still found myself with a slight preference for the LP, even though the digital version unlocked detail that I was frustrated in not hearing on the vinyl playback. It was purely and totally subjective. There is no rational reason to explain why, to this day, when I want to hear that recording, I play the LP. I'm not even more comfortable with the "unmolested" CD, and I don't know why.

If there is enough of a subset of movie theater customers to feel the same way about 35mm, that's great. But like me and classical music recordings, I don't think that there are enough of them to keep the theater business alive by themselves.

[Marin Zorica]

So I bought the CD version, ripped it to a 44KHz .wav file, applied the RIAA curve to it in Diamond Cut, and output the result to another CD. Listening to both through the same amplifier and speakers, I still found myself with a slight preference for the LP, even though the digital version unlocked detail that I was frustrated in not hearing on the vinyl playback. It was purely and totally subjective. There is no rational reason to explain why, to this day, when I want to hear that recording, I play the LP. I'm not even more comfortable with the "unmolested" CD, and I don't know why.

Offtopic, but, how does it sound when you have applied RIAA to cd? RIAA on playback should be applied only to records which were recorded with applied RIAA eq! Otherwise it sounda wrong if you just applied RIAA on playback, whole low end would be boosted. Same with fm radio. On transmitter emphassis is applied and on tuner demphassis, used to improve signal to noise ratio.

Few times a year I to to do setup and service for some location still running 35mm, and visiting some 70mm festivals. None of them are better compare to restored version on dci. But, that feels of cracking in sound, scratches on picture, reel changes etc makes you feel you are in cinema

The best presentation I've have ever been was Hatefull eigth im Vienna Gartebaukino and also Lawrence of Arabia in BFI London (they have told it was New print). Presentation was superb, prints were new....but cost of it was such high, both prints and off course guy involved to set up and check after every show. But, i would wonder how that print would be after 150 screenings.....hopfully both were DTS, so no fading in sound.

[Steve Guttag]

Sticking to the OP's question (the feeding/care of Digital versus film). The vast majority of my "service" work is now done remotely and I can do it on 100s of screens...something that I could not do with film. I could service that many film projectors but but you can figure how much time one person can devote to each projector doing that within any calendar year. Both still require a degree of on-site maintenance. Both suffer from lack of or outright bad customer maintenance. I don't fully blame to customers. After all, they would sold the bill-of-goods about "maintenance free."

As for reliability...it isn't even close. A film projector will last 50-years plus, if reasonably well cared for. There are no illusions that the digital projectors will. Even if you care perfectly for them, their semiconductors are not designed around that...the mirrors will get stuck, an ICP or equivalent will fail (or its certificate will expire). The most expensive piece in a film projector is the intermittent movement. A complete rebuild of that (which can be done in a shop or, in a pinch, in the field) is on the order of what a typical part in a digital projector costs and nowhere near the price of the ICP or light engine/prism.

Comparing same-same. Most platters will outlast every server in use. Again, the servers are not designed to last that long and, as we've all learned, their certificates are fleeting. Their technology are also short lived.

In 100-years time, if someone pulls a reel of film from a vault, even without any documentation, they'll be able to figure out how to watch/listen to it (maybe not with digital sound, but perhaps). The encrypted digital files, depending on how they're stored and what the capabilities of the future people have, may not be able to get the information off the storage. Bump that to 500 or 1000-years, film is going to be all that much easier to figure out, providing that it survives. They think they have the storage techniques now to preserve film for 1000-years. None of us will know but I give it more likelihood than anything stored in present-day digital.

Sound wise. It depends on the criteria. The analog film sound processors (CP50, CP55, CP65, CP100, CP200, and some of their equivalent competitors) all can be running this very day...upwards of 50-years after they were made. There are cinemas out there right now using their film sound processors with a D/A converter like a DAX602. Those that are running are probably giving people no grief. Those that are giving people grief are likely not from the above list and probably have some form of digital processing in them that isn't working like it used to. When we either maintain or put in film sound systems, in this era, we use a film sound processor as a source to main sound system. There is no/little point to try and recreate that in DSP as the film sound processor has everything you need for those film projectors...including the noise reduction, changeover and optical preamps suited to what they are being connected to.

To be fair to modern digital processors, they can be refreshed with current models to handle digital cinema and most server companies now have some form of built-in sound processor as an option. However, the CP750 didn't exactly last as long as its predecessors and does anyone think they can run their CP850 as long as a CP65 or even a CP650? Like everything with digital...it is for the moment, not for long-term. As to the OP question, often the support/repair to something now is "Replace it." With film, it was "repair it." I guess, one of the things I like about Q-SYS is that you should be able to evolve it as technology changes. That is, if you need to change the Core, change the Core but not the rest of the system. Then again, sometimes (and more and more) they don't include some of the older I/O in the later software...which compels a greater depth in the refresh. Digital audio is definitely a different world. And if you get your sound processor via your server...I guess your audio refresh cycle is now tied to your server one. Traditionally, sound processors had a very long life compared to things like servers.

Now, when it comes to diagnostics, like with digital projectors and servers, if the processor is Q-SYS...I can diagnose the system, remotely from the point the audio enters all of the way to the output of the amplifiers and hear it remotely. I'm not dependent on the person working the theatre. As much as I love the Dolby CP200 sound processor, if something comes up, I can't remote into it (though I could put some GPIO on it, I suppose to see what status lines it has active...hmmmm). Then again, I have three of them in active use and the down time on them is essentially zero. Recently, I had one where, of all things, the CAT 137L (a link card) had to be reseated due to an intermittent connection. That card hadn't been touched in 22-years and probably won't be for another.

I'll conclude with...without a doubt, I'd rather make/update a screen file in a digital projector than cut a brass aperture plate. A screen file is faster, has a sharper edge and hurts my palm less. Then again, with the aperture plate, if you are good enough (with the right tools), you can shape it in a way a screen file just can't (with current technology).

[Frank Raia]

Thank you Steve. As always I was looking forward to reading a response from you. Very insightful. Now I can't wait to see if Brad agrees with you!

[Tony Bandiera Jr]

Sticking to the OP's question (the feeding/care of Digital versus film). The vast majority of my "service" work is now done remotely and I can do it on 100s of screens...something that I could not do with film. I could service that many film projectors but but you can figure how much time one person can devote to each projector doing that within any calendar year. Both still require a degree of on-site maintenance. Both suffer from lack of or outright bad customer maintenance. I don't fully blame to customers. After all, they would sold the bill-of-goods about "maintenance free."

<truncated to save space since I agree with all the points herein>

Now, when it comes to diagnostics, like with digital projectors and servers, if the processor is Q-SYS...I can diagnose the system, remotely from the point the audio enters all of the way to the output of the amplifiers and hear it remotely. I'm not dependent on the person working the theatre. As much as I love the Dolby CP200 sound processor, if something comes up, I can't remote into it (though I could put some GPIO on it, I suppose to see what status lines it has active...hmmmm).Then again, I have three of them in active use and the down time on them is essentially zero. Recently, I had one where, of all things, the CAT 137L (a link card) had to be reseated due to an intermittent connection. That card hadn't been touched in 22-years and probably won't be for another.

I'll conclude with...without a doubt, I'd rather make/update a screen file in a digital projector than cut a brass aperture plate. A screen file is faster, has a sharper edge and hurts my palm less. Then again, with the aperture plate, if you are good enough (with the right tools), you can shape it in a way a screen file just can't (with current technology).

To address the bolded points in order:

But the work is still billed remotely, at a standard rate and hourly minimum, right? Otherwise, it would not be a sustainable business model. My time and knowledge, even in the film days, was always billed out. To be fair, there were times when an existing customer, (and on super rare occasions, a new phone in customer) would get help from me with at no charge or a reduced rate. My point is, whether onsite OR offsite, a service call is a service call, and should be billed accordingly, at standard rates.

Absolutely agree. But take the onsite maintenance of film equipment, then compare that to BOTH the on and off site maintenance of digital. Put both of those on a spreadsheet, filter as number of calls per screen, per month/annually, and I'll bet that Digital requires MORE maintenance than film by a factor of 2 or more. And at what costs? Again, I am pretty sure current hourly service rates for digital are more than film, even accounting for inflation.

Sure, if you have Q-SYS installed. I am sure that there are far more locations that do not. What do you do then? You spend (and bill out) even more time making the phone calls to someone on site, who most likely doesn't have a clue, to try and get accurate feedback on what exactly is going on. (Obviously a moot point IF you have Q-SYS which will give you an accurate snapshot.) And sometimes that made you have to drive to the location anyways, so now you've spent even MORE time. (Same as in the film days... I hated doing phone calls to try to troubleshoot with a clueless popcorn kid, it would've taken less time for me to just drive to the location and see for myself.) Plus, when you "eyes on" and/or "hands on", you could ALWAYS catch something else before it became a problem. (Yellow or red taillights, server nags, etc.) Back in my service days in the 'plexes, EVERY service call I did was followed with a walk through of ALL the other equipment. I almost always found other issues, some major, (Century intermittents with "rusty" oil was the big one at one chain I serviced), amp channels locked in protect, bad bearings/belts, jammed pad rollers...you get the idea.) Not only did I often save a dark screen, I saved the customer money from not losing shows and doing refunds, but also MADE myself (and my company) money on top of it. It also made for a good customer rapport, many appreciated my proactive catches.

The "rainbows and Unicorns" that were promised and not delivered.

Your CP 200 situation proves the point. I know of locations with CP 65's still banging along just fine after all these years, and you can still find spares online and most experienced techs like Steve and myself can repair cards at the component level (if the part isn't NLA.)

Finally, I've had a few situations in the film days where that was 1,000% true. Yes, I also never really liked doing aperture plates... BUT when the results turned out spot on, that huge satisfaction was worth the time and sweat spent. I never got that from setting up digital lens files, since I KNEW that anyone with a pulse could eventually get there...try that with aperture plates and you'd spend hundreds, waste hours, and have a lot of brass dust and a shitty outcome.

[Steve Guttag]

Tony,

To you points.
Remote Digital Billing. I can only speak for our business model and not others. We have something we call Cardinal Care. It is a remote-only service (if we need to come out, it is not included but we can better make that decision with it). We bill a flat rate based on screen count and TMS (which counts as a screen unto itself). Think of it like insurance...though any one client may end up getting a very good deal time-wise, as a total group, we do okay with it. I might have a single that gets an issue that consumes some time but that 10-plex rarely, if ever consumes as much time as 10 singles. Also, we don't offer Cardinal Care to just anyone. They are going to be systems we install or systems we feel we can work with and have enough ability to provide said-care. We have remote access to our own device that is never owned by the client so it is the same as us standing there in your booth tapping on a laptop. That device will have whatever programs, bookmarks or whatnot that will allow the tech to diagnose whatever equipment is in that site. By and large we can power up equipment, update it, run it through diagnostics and, depending on the equipment's capabilities see what is going on. Yeah, if it is a CP950 you can't see as much as with Q-SYS but you can see a bit and have a feeling if the sound issue is the server, the CP950 or further downstream. When USL was in business, the CM-8E was a handy tool too since you could also hear/see the meters of both processor and amplifier outputs. You work with what you have...even a CP650 or JSD80 there are some tools to give the tech some ideas as to what is going on.

Depending on the site, we won't even offer Cardinal Care because there would be no way we could do our job or the system would be one call after another. One also has to be strict on EQUIPMENT issues versus Employee issues. You not knowing how to operate and perform basic maintenance (clean/replace the filters) is not our fault nor covered. We expect our clients to operate basic hand tools like a screwdriver (but not much else).

In the film days...we had a flat rate if we could "fix" the problem via a phone call. It was pretty low...something like $25 or $30. It was FAR cheaper than even the travel time to/from the theatre.

I make a point of taking pictures in theatres we service. It is a lot easier to tell someone where to put their hands when you can show them in a picture. And, push come to shove...take a picture of what you are seeing so I can see it too.

I don't know if our clients spend more with us on labor (adjusted for inflation) in the digital era. In the Film era, many of our current clients would have done a bit of the work themselves (not our screening rooms but our more rural customers, for sure. Now, those customers, via Cardinal Care pay regularly but at a very fair rate (in my opinion). Where they really pay more is in the equipment and its installation. We're changing out CP750s pretty regularly now. All of the original servers from just 10-15 years ago are now dropping (regardless of brand). Projectors even. You are one bad formatter/stuck pixel away from a VERY expensive day. In the Film days, we almost never replaced a projector or platter (or sound processor, for that matter). I had one client that had us put Kinoton in to replace working domestic projectors, but that is the exception to the rule. The only way new sound processors went in is if some new format came out...like in 1999 with Star Wars EP:1...that put quite a few sound processors in place of Mono and it put Surround-EX <blech> in too. But other than that...film equipment, once installed, pretty much just stayed where it was until the theatre closed or until DCinema happened.

[Harold Hallikainen]

Are digital products inherently less reliable or have shorter life than analog products? Why does it appear a CP65 lasts longer than a CP750? I can see something like a DMD having a shorter lifetime due to the extreme temperatures it has to deal with, and it's a mechanical device. Do digital systems just have inadequate cooling causing a short lifespan on components? Are we seeing data corruption in flash memory? So, what is going wrong in digital systems?

On DMDs, I recently read about a different application. They can be used instead of photomasks in the manufacture of chips and circuit boards. The whole photoplotting step can be skipped. Clever!

[Caleb Williams]

DMD's are electromechanical. They are, other than manufacture defects, technically the most robust part of the projection system besides the frame. They are rated for tens of millions of cycles per chip. (Field experience might disagree with this, but consider that to generate a greyscale pixel, you might have an oscillation of up to 5000x per second of a single mirror.)

Most of the problem with the expected lifespan of digital equipment comes down to the shear complexity of the hardware. Film projectors are very mechanical and use relays, motors, and simple switches. Projectors use microswitches, transistors, massive and tiny inductors, and probably the most prone to failure, electrolytic capacitors. If you were just apples to apples compare the number of parts between film and digital projection, it's not even close. There's a lot more to break. It is also that complexity that causes most issues to require the replacement of the larger assembly, like a full ICP board instead of just a component on it. That adds incredible cost compared to film projector parts. Also consider the expertise needed to replace some of those parts. A light engine, for example, might run you $17,000 USD used. Are you really going to pay an inexperienced manager $9/hr to install that part? Absolutely not.

[James Gardiner]

I would like to add.
from my perspective digital is a far better long term archive path.
sure with film, its common sense.. you shine a light through every picture etc. but to make a film projector when you have no historic understanding of how it worked is not trivial.

with digital. as long as we have computers. and we have to as we have it in digital storage. the open source code to make it all work is a part of the archive.
as long as all those ones and zeros are Kept valid in whatever formats are developed. we can recover exact quality as created by the director and not, as film does, fade and alter from producers intent over time.

in my book film is no match for digital. it's just the way digital is stored and it's longevity that is in question. and that improves all the time.

hopefully , like in Superman with the crystals.. yea that idea, at the time was based on expected future developments of etching the data into the structure of the crystal. they are still working on it but if perfected would last for an incredible amount of time (read only).

[Steve Guttag]

Harold, I think the fault tolerance with film equipment and analog is much higher. A film projector may (and often is never) perfect. However, before its sloppiness or the wear of a gear/sprocket...etc becomes an issue visible to the patron is MUCH longer. With digital, everything has to work, perfectly, all of the time. One noisy connection, the system stops. You can build in some error handling...RAID drives being a prime example but even those will start to stutter a picture if they can't quite keep up.

As for a CP65 versus a digital processor, like the CP750...the CP65 is built, primarily on low-tech that is inherently robust. There is almost nothing esoteric in it. What is there to fail? CMOS 4000 series chips? TL074 Op Amps? Resistors? Most everything in them can also be multi-sourced. Yeah, if a power supply has an issue you might get a hum or you could get a cold solder joint to make noise...but it WILL still work. Compare that to a CP750 or a JSD-60. A bad power supply can and has been a show stopper. Solder joints (particularly the Lead-free ones) can be a killer over time. And yes, I think the CP750's issues are partly due to heat issues. It has no fan in it though it has a place for a fan to plug in on the motherboard (near where the power supply plugs in). Then again, if it had a fan, how would the unit survive as it got slowly coated with dust and projection room soot from the popcorn popper? I do think that the CP750, in particular, has an unusually short life span. I wish Dolby had figured out just what kills its motherboard. It has an issue where the sound slowly diminishes (starting with the surrounds but it spreads) as well as boot up issues or lockups shortly after boot up. Then, people could "rebuild" them from the known/common issues. However, it could be that some of its problems are non-repairable due to semiconductors being unavailable (e.g. AKM no longer making the part and they being the sole-source).

However, even CP650s are getting flaky as are JSD80s. Many of them are over the 2-decade mark so they don't really owe anybody anything but the CP65 is still slugging along. However, with film, we really are in a world where we need to make this stuff last, indefinitely.

I do not agree with James on the longevity of digital storage as he cannot prove his case. The cost of storing movies born-digital, at present, is a high expense. Film-outs are done for archival purposes because we have test data on how long that copy should last. It will be interesting to see how people handle digital storage, long-term and how they deal with encryption methods as well as retrieval from, soon-to-be-obsolete storage mediums/technologies.

[Leo Enticknap]

I would like to add. from my perspective digital is a far better long term archive path. sure with film, its common sense.. you shine a light through every picture etc. but to make a film projector when you have no historic understanding of how it worked is not trivial.

If, in a century's time, I had a choice of being handed a reel of 4/35 camera negative and being told to go build something that could reproduce it, or being handed an LTO tape and given the same challenge, I know what I'd choose. Film transports are relatively simple mechanical devices. Someone who has no historic understanding of how they work could easily gain that understanding using a couple of thousand words of text and a couple of dozen diagrams. Film transport mechanisms were also in mass production all over the world and mainstream use for roughly 110 to 120 years (as examples for projectors, the Lumière Cinématographe began serial production in 1895, and Christie discontinued serial production of 35mm projector heads in 2009), which generated a lot of widespread primary evidence as to what they did and how they worked.

The media, hardware, and software used by electronic (both analog and digital) audio and video storage media and formats, were, as a general rule, produced in smaller overall quantities (with some exceptions for systems sold direct to consumers), and with a level of mechanical, electronic, and software complexity that makes recreating the ability to play some obsolete media practically impossible. About a decade ago, the BBC did an audit of their archival holdings that established that they have around 50,000 hours of 2" quadruplex videotapes that they will likely never be able to recover the contents from, even though most of them are in chemically good condition. The reason is that they simply don't have enough viable head hours left on their existing playback machines, and there is no viable way to extend their service lives.

Throughout the '90s and '00s, professional archivists warned that the proportion of "born digital" content involuntarily (as in, despite attempts to preserve it) lost would dwarf that of nitrate, thanks largely to the proliferation of media/formats, the fact that most of them were short lived, and the complexity of hardware and software involved. I have not worked professionally in the archive world since 2012 and so don't know what the current thinking on this is, but feel confident in speculating that few archivists would support the simple claim that "digital is a far better long term archive path." The economics may have changed since then. By the late '00s, the cost of the long term atmospherically controlled passive conservation (even though the combination of low temperature and low humidity is very energy intensive) for film was still a bargain compared to that of continuous format/media migration for even standard definition (PAL and NTSC) video.

But in the decade and a half since then, digital assets have become more file based, meaning that the migration process likely doesn't involve hard to maintain legacy hardware to the same extent as it once did, Moore's Law has driven down the cost of storage, and the data center (f-----g great, always online RAIDs, in other words) has emerged as a cornerstone of IT infrastructure. How much it costs to maintain a couple of hundred terabytes (a 4K feature in uncompressed DPX files) in a data center over ten years compared to 50,000 feet of 35mm (a typical feature in color separation negatives and audio elements) in a cold and dry vault would be an interesting question, and the math may be shifting in favor of those terabytes. But even if it is, I'd still feel cagey about asserting "far better" for digital, and that is even before the ethical questions of abandoning any attempt to preserve "born analog" media in their original form are addressed.

[Frank Cox]

I can see good points on both sides of this question.

However, one thing in favour of digital preservation is that it's instantly "portable" and can be backed up. If I have a can of film that's all I've got: one can of film. If I have a digital file I can copy it to as many other devices as I want to. DRM can interfere with this, but as Bruce Schneier famously said, "Trying to make bits uncopyable is like making water not wet."

It's also possible to create virtual machines and emulators that will play old digital media that are saved in the form of files. For example, I personally use dosemu, fs-uae and vice to view and run software for msdos, Amiga and Commodore64/128 respectively. I have a fair-sized library of software that I've collected over the years and it takes up no physical space. I store it on my current computers and when I get a new computer I can just copy it to the new computer along with all of my other files. So files that I've had for several decades are still here right in front of me, immediately available for my use should I want to use it.

If I had the same stuff as physical media of some sort I'd need a warehouse to store it in and it would take me hours to find something that I was looking for on a dusty shelf in the back.

[Carsten Kurz]

On DMDs, I recently read about a different application. They can be used instead of photomasks in the manufacture of chips and circuit boards. The whole photoplotting step can be skipped. Clever!

Similarly, they are also used in certain types of 3D printers.