future of film cameras

[Deniz Zagra]

Apparently, no new film motion picture camera has been manufactured since 2009 which means that rental houses rent out existing cameras. Is it possible that, after a while, the on-the-market cameras will start disappearing due to age, requiring Panavision or Arri to make new ones? I guess film cameras last longer than digital cinema cameras since most of the parts are mechanical, but they would still need maintenance and repair.

Were cameras back in film days mass-produced or did they require intensive human labor with little automation? Do big companies like ARRI and Panavision still have the necessary equipment, personel, and expertise to manufacture existing models (it's unlikely that new models will be designed), and if so, is this a feasible for them? I heard that Panavision made 2 new XL2's for The Force Awakens.

The only niche camera manufacturers that I know of is Magellan and Kodak. I think Magellan made a super 8 and a 65mm camera. I don't know if they get any use on the field though. Kodak came up with the new Super8 camera which seems like a good thing. Has anyone used it before or heard of it being used on sets?

[Giray Izcan]

The current modern cameras will be used until they last or until Kodak stops making film unfortunately. 

[Aapo Lettinen]

The camera mechanisms of higher end cameras tend to be pretty durable and with the lower use they see nowadays they probably last pretty much forever with minor repairs and occasionally changing a part or two. The electronics of those cameras are old but new electronics can be made if enough budget is provided to design and build them. There is still very skillful mechanics around who can make new mechanical parts for film cameras, it would be expensive because it is very labor intensive but it is worth it if it is a higher end camera and original parts are not available.

Custom optics tend to be difficult to arrange if someone would break optical parts of the camera and original parts or close-enough replacements are not available. Surely someone could do that but it is very challenging and needs tons of experience (like 20 or 30 years minimum) and lots of time to happen. So I would be careful to not break any optics as they can be absolute nightmare to reproduce if original parts are not available.

A good quality film camera can easily last 100+ years in use if it is taken care of even a little bit and not stored in someone's damp basement to collect mold and rust. The electronics age at some point and stop working or at least become very unreliable, usually the time limit seems to be something around 50 years after which camera electronics seem to be aged enough that they often start to have more issues than you'd like and need to be completely replaced with new ones if wanting to have some reliable shooting with the camera still

[Aapo Lettinen]

on the question why new film cameras of higher end type are not made anymore:  the cameras typically cost, converted to today's currency, something like from 30k to 60k usd a piece and were mostly manually assembled and most parts made manually too if needed to be precise.

On some other threads we tried to calculate how much it would cost to make similar quality high end camera nowadays with all the cnc and 3d printing technology and advanced electronics etc...  the surprising result was that it would still need to be mostly manually made and cost about the same per camera than back in the 70's or 80's, from 30k to 60k usd a piece for Eclair ACL or ArriSR or AatonLTR type of camera.

Back then 50 years ago when the cameras were used for professional work daily it was a no-brainer to purchase a, for example, 50k camera, no problem at all, the higher end professional video cameras today cost about the same and one will get the cost back quickly if having real use for it.

But what if wanting to make a 50k film camera today? Does anyone really use it daily for high paid professional work and would get the purchase price back in a year or two? Absolutely not. It would sit most of the time on the shelf or used for film tests and low paid indie stuff. then once or twice a year something which one could charge premium for to purchase some noodles, the camera still not paid for. One would probably never get the purchase price back, ever. If looking for alternatives the used 50 year old cameras would be +/- 10k , well over 20k in all cases. they would be not as good on all fields than completely new camera but mechanics quality lose enough and one would save like 40k compared to new camera with little to no difference in the quality of the end result.

This is why new cameras are not made, they would be so expensive that it would not make sense compared to purchasing a similar quality used camera which is still working fine but costs tens of k's less

Edited May 29, 2024 by Aapo Lettinen

[Robin Phillips]

6 hours ago, Deniz Zagra said:

Apparently, no new film motion picture camera has been manufactured since 2009 which means that rental houses rent out existing cameras. Is it possible that, after a while, the on-the-market cameras will start disappearing due to age, requiring Panavision or Arri to make new ones? I guess film cameras last longer than digital cinema cameras since most of the parts are mechanical, but they would still need maintenance and repair.

Were cameras back in film days mass-produced or did they require intensive human labor with little automation? Do big companies like ARRI and Panavision still have the necessary equipment, personel, and expertise to manufacture existing models (it's unlikely that new models will be designed), and if so, is this a feasible for them? I heard that Panavision made 2 new XL2's for The Force Awakens.

The only niche camera manufacturers that I know of is Magellan and Kodak. I think Magellan made a super 8 and a 65mm camera. I don't know if they get any use on the field though. Kodak came up with the new Super8 camera which seems like a good thing. Has anyone used it before or heard of it being used on sets?

the magelin 65mm is made by logmar. they made the best super 8 camera ever made, and they also designed the kodak one.

the good news is film cameras mechanically can last more than 50 years, in some cases more. Their nature and build quality makes them easier and cheaper to repair/modify than to build a new one from scratch. Building an SR3 Advanced replacement would probably have to retail somewhere in the 50k USD range. Logmar actually looked into this and abandoned it. They were open about cost being the issue. even their 100ft load 16mm camera was going to have to cost around 25k I think to the end user. That assumed I think a minimum of 50 orders, might have been more. fewer cameras = each more expensive.

You can get an SR3 Advanced package with a few mags and optical accessories for around 20k USD right now. 

There is certainly a potential if not likely world where in 10 or 20 years, between film costs rising and older cameras failing or becoming less reliable, that the only way to shoot film will be to rent cameras from Panavision. 

But for now, Arris are repairable. Aatons are reparable. Panavision rents film cameras, as does Arri Rental and a few other big camera houses. Kodak is still making film. And while if I wont the lotto tomorrow I'd quite wrecklessly commission and build a new fleet of film cameras, the reality is that the way to keep film going in the future is to keep shooting it now. If the demand remains, someone will figure out how to keep these machines going.

[Aapo Lettinen]

8 minutes ago, Robin Phillips said:

There is certainly a potential if not likely world where in 10 or 20 years, between film costs rising and older cameras failing or becoming less reliable, that the only way to shoot film will be to rent cameras from Panavision. 

But for now, Arris are repairable. Aatons are reparable.

I think the issue with the cameras is not that it would not be possible to repair them, it is that people give up on them easily because they don't use them often enough to justify spending lots of money to making them work again and to ensure that they work reliably.

I have the same problem too, despite I am making new crystal sync electronics for various camera models I have too many cameras collecting dust (need different models for reference to be able to make the motors) and I can't afford to shoot with all of them or keep them in shooting condition better than absolutely needed to run motor tests. This is the issue with other people as well, rarely anyone has possibility to shoot as much film as one would want to and one needs to very carefully evaluate if broken gear is worth repairing or not , if spending money on it you would shoot even less than otherwise 😕 

I could make new electronics for Arri SR or Aaton LTR/XTR/Prod no problem. The issue is that it takes long and creates lots of costs. I estimated at least a year of designing and about 20k minimum order paid up front would be needed with about 5k cost per camera (so 4 cameras to convert to start the process might do and one free spare for me which I could ruin in testing phase if needed or at least keep as reference to be able to make more of these later). I could arrange single camera needing new electronics but the minimum order amount is the issue, where to get multiple people ordering stuff at the same time? 

People start to hesitate if it is worth it to invest on the camera to make it run reliably for 15 to 20 years straight and they choose to save some money and hope that the original electronics could just last a little bit, even tiny bit longer so that they could make one more small project with it ,then hopefully one more, and so on

[Tyler Purcell]

9 hours ago, Robin Phillips said:

the magelin 65mm is made by logmar.

But nobody is really using it tho, they really made it to prove they could, rather than filling a need. 

9 hours ago, Robin Phillips said:

they made the best super 8 camera ever made, and they also designed the kodak one.

Arguably... it delivers the best image, but is not the best to use. Ask them if they designed the Kodak camera and they will deny., 

9 hours ago, Robin Phillips said:

the good news is film cameras mechanically can last more than 50 years, in some cases more.

I know Arri 35mm model B's from 1950 still running perfectly. Those cameras are bulletproof. Only some of the more modern cameras would struggle to survive. Things like the Bolex are pretty indestructible as well and will run another 100 years, without much of an issue if properly lubricated well. 

9 hours ago, Robin Phillips said:

Their nature and build quality makes them easier and cheaper to repair/modify than to build a new one from scratch. Building an SR3 Advanced replacement would probably have to retail somewhere in the 50k USD range.

To match the quality of the SR3 would probably cost 5 million to setup the tooling and another 2 or 3 million to reverse engineer and manufacture enough to start paying bills. Retail price would be in the 150k area for a camera of that quality with magazines. The mirror shutter and viewfinder optics alone, would probably COST in the range of tens of thousands and would need to be done by a 3rd party. It's frankly, not possible to re-create ANY optical viewfinder camera in the modern era, there just aren't enough small optical houses around to take on such a project. 

9 hours ago, Robin Phillips said:

Logmar actually looked into this and abandoned it.

Logmar are a few guys having fun on the side, they are not a serious camera company. They lucked out with the super 8 camera and have been living off it's laurels for years. I don't think they made any money off it and any company who would exist to make cameras, would need to be in the business of making profits.  

9 hours ago, Robin Phillips said:

There is certainly a potential if not likely world where in 10 or 20 years, between film costs rising and older cameras failing or becoming less reliable, that the only way to shoot film will be to rent cameras from Panavision. 

Panavision isn't magical, they don't have an endless supply of Arri parts for their 16mm and 35mm cameras. So once those parts are gone from Arri, that's the end of that. Panavision does have a few in-house options, but they're pretty bulky and expensive. Professionals will be able to rent them no problem, but sadly that does nothing to help perpetuate the use of film with young people, who are the future. 

9 hours ago, Robin Phillips said:

But for now, Arris are repairable. Aatons are reparable. Panavision rents film cameras, as does Arri Rental and a few other big camera houses. Kodak is still making film. And while if I wont the lotto tomorrow I'd quite wrecklessly commission and build a new fleet of film cameras, the reality is that the way to keep film going in the future is to keep shooting it now. If the demand remains, someone will figure out how to keep these machines going.

Yes, I think demand is the problem. We have a design I've been working on for a hot minute, which is intriguing, using a mix of easy to manufacture plastic parts and an easy to build movement with a fancy beam splitter to deal with the video aspects rather than a mirrored shutter. It's all in the white paper phase, but I reckon we could probably sell bodies for the $5k figure, or at least that's my goal, with a decent investor. We will be making a prototype next year and be focused on the 16mm format, which is the bread and butter for most filmmakers who shoot on film. To me, any new camera needs to be open source, with lots of easy to program boards and off the shelf components as much as possible. This way you're building a long-term ecosystem and not just a box which can't be serviced. We even contemplated making it a kit system where you buy the movement, video and boards from us, but you build your own housing to hold it all. Personally, I think there is quite a bit of business in that market. 

[Tyler Purcell]

9 hours ago, Aapo Lettinen said:

I could make new electronics for Arri SR or Aaton LTR/XTR/Prod no problem.

It's more about a dead quiet motor that can do high frame rates as well. The motor control is easy, the motors are hard. We are working on one as well, we will send you notes as we get closer. 

[Tyler Purcell]

16 hours ago, Deniz Zagra said:

Is it possible that, after a while, the on-the-market cameras will start disappearing due to age, requiring Panavision or Arri to make new ones?

Over 50 years, we will for sure see many cameras simply not be on the market anymore due to various failures which can't be easily fixed. I don't foresee this happening anytime soon, but yea I can see Arri's and Aaton's especially going belly up from simply being worn out and board/motor failures. We put a LOT of wear and tear on our Aaton's, I can't imagine how many feet I've shot alone, let alone the prior owners. Parts are one thing, in 50 years the knowledge will also be long gone. None of the guys who worked on them when they were new, will be around. The knowledge will be 2nd or 3rd generation, which means it's basically going to be lost. I don't know anyone my age servicing film cameras, nearly everyone is in their 60's or older. 

No major firm will make new film cameras, in their eyes it's a long dead market. Arri can no longer make film cameras, they got rid of their entire line and Aaton is out of business. Bolex could theoretically continue making cameras based on their parts inventory, but I don't know if they will ever bother again. They did up to the 2010's. Panavision does have machines to work on cameras, but they aren't building from scratch. Most of their cameras were hobbled together from Mitchell movements. 

16 hours ago, Deniz Zagra said:

I guess film cameras last longer than digital cinema cameras since most of the parts are mechanical, but they would still need maintenance and repair.

Film cameras need way more maintenance than digital cameras. 

16 hours ago, Deniz Zagra said:

Were cameras back in film days mass-produced or did they require intensive human labor with little automation?

Almost entirely manual labor and some assembly lines. Bolex had full on assembly lines, but Moviecam didn't. So it varies. 

16 hours ago, Deniz Zagra said:

Do big companies like ARRI and Panavision still have the necessary equipment, personel, and expertise to manufacture existing models (it's unlikely that new models will be designed), and if so, is this a feasible for them? I heard that Panavision made 2 new XL2's for The Force Awakens.

No and I don't believe they made two brand new XL2's. I bet they took pre-existing movements and shoved them into new housings. If Panavision could make new cameras, they would have a fleet of 5 perf and vistavision cameras. 

16 hours ago, Deniz Zagra said:

The only niche camera manufacturers that I know of is Magellan and Kodak. I think Magellan made a super 8 and a 65mm camera. I don't know if they get any use on the field though. Kodak came up with the new Super8 camera which seems like a good thing. Has anyone used it before or heard of it being used on sets?

Magellan is logmar. They aren't a camera manufacturing company. They basically are a bunch of part time guys having fun designing stuff. Great team, good guys, have some talent for sure, but they aren't in a place they could really make new cameras for the industry. If the Kodak camera or 65mm camera are any indication, they're far away from making anything worthwhile anymore. They did make a little super 8 camera that created a good image, but it was short lived. 

[Simon Wyss]

2 hours ago, Tyler Purcell said:

Bolex could theoretically continue making cameras based on their parts inventory, but I don't know if they will ever bother again.

I’ve heard a rumour that at least part of the inventory is sold. They sure don’t bother.

[Aapo Lettinen]

2 hours ago, Tyler Purcell said:

It's more about a dead quiet motor that can do high frame rates as well. The motor control is easy, the motors are hard. We are working on one as well, we will send you notes as we get closer. 

motor control is the hardest part, the customers ordering electronics update for such a high end camera would ask for features which cannot be done with factory made bldc boards and one would need to make everything completely out of scratch withtout possibility to use factory made modules, the bridge drivers and commutation logic with emulated sine wave commutation, the crystal control and user interface and everything. easily a 1 to 2 year job if done well, no matter how experienced the designer is.

Getting suitable motor drives is easy, just throw some Maxon drive in or get something custom made. I don't know how much more quiet the drive would need to be than a good quality stock bldc motor already is, it is all about the drive mode you run it on which hugely affects noise levels (simpler logic = more noise,  less noise = very complex and expensive and slow to design commutation)

[Karim D. Ghantous]

16 hours ago, Robin Phillips said:

And while if I wont the lotto tomorrow I'd quite wrecklessly commission and build a new fleet of film cameras, the reality is that the way to keep film going in the future is to keep shooting it now. If the demand remains, someone will figure out how to keep these machines going.

Same! Of course you'd need a few million, so let's say $8M to allow for cost over-runs.

6 hours ago, Tyler Purcell said:

It's more about a dead quiet motor that can do high frame rates as well. The motor control is easy, the motors are hard. We are working on one as well, we will send you notes as we get closer. 

Motors have gotten very silent over the years. I was very impressed when Leica released a dead-quiet focal plane shutter that was introduced in the M10-P.

[Victoria Sagady]

I think it’s worth noting that one of the biggest threats to using film is not the age of the cameras or cost of repair, but the institutional knowledge that is slowly fading away as less and less people are apprenticed/trained in on using these cameras and maintaining them.  Beyond manuals and instructional videos many of these cameras have personalities and quirks that can be very difficult to work with if you don’t know them or the best ways to work with them.  

As each lab closes we have less and less people who know the ins and outs of the development process and how to maintain that equipment.  We’re going to hit a point where all that experience retires or passes away without being passed on to enough people to keep working with film practical.  We’ve seen this with many industries as technology advances and while there has been a good effort to keep film alive I beleive it’s sadly a matter of time. 

[Uli Meyer]

10 hours ago, Tyler Purcell said:

I don't know anyone my age servicing film cameras, nearly everyone is in their 60's or older. 

Daniil Nevskiy is half that age and will keep some of that knowledge alive for a few decades longer. All by himself 😉

[Tyler Purcell]

3 hours ago, Uli Meyer said:

Daniil Nevskiy is half that age and will keep some of that knowledge alive for a few decades longer. All by himself 😉

Yep, Daniil and myself are very much outliers and unfortunately, we need to duplicate ourselves 5 times and spread ourselves all over the world to keep these cameras working. In the US, I don't know anyone doing what I'm doing. Most other techs are focused on cameras like Bolex's and such, totally different market. 

[Tyler Purcell]

4 hours ago, Victoria Sagady said:

I think it’s worth noting that one of the biggest threats to using film is not the age of the cameras or cost of repair, but the institutional knowledge that is slowly fading away as less and less people are apprenticed/trained in on using these cameras and maintaining them.  Beyond manuals and instructional videos many of these cameras have personalities and quirks that can be very difficult to work with if you don’t know them or the best ways to work with them.  

Correct, the knowledge is no longer transferring. The current techs who are close to or already retired, don't have much interest in training new people and those who have, are only training one or two people. They aren't doing multi-person courses on how things work. There are very few understandable instruction manuals, most cameras do not have anything a consumer would understand OR have the proper tooling. Even with the proper tooling, you have to know what the tools are and what they do. I have a dozen tools I purchased that nobody I've met has any idea what they do. So yea, the knowledge IS the problem and it's going to disappear. 

4 hours ago, Victoria Sagady said:

As each lab closes we have less and less people who know the ins and outs of the development process and how to maintain that equipment.  We’re going to hit a point where all that experience retires or passes away without being passed on to enough people to keep working with film practical.  We’ve seen this with many industries as technology advances and while there has been a good effort to keep film alive I beleive it’s sadly a matter of time. 

Labs aren't closing anymore tho, those days are behind us. People are building new labs and sure, there are some issues with building labs, but it's pretty straight forward, it's a lot of trial and error for the uninitiated. Many labs hire young people to work there as well, so unlike cameras, lab tech knowledge is being transferred to young people. Could those people build a lab? Eh... maybe? 

[Geffen Avraham]

Tyler, I'd love to know more about this camera you're making! There was a company back in the day that made a transflective mirror, that could electronically switch between full transmittance and full reflectance. This could be a perfect replacement for a spinning or oscillating mirror - no moving parts, completely quiet.

Honestly I don't think optical viewfinders are necessary these days, we have great EVF systems, if you can have a sensor the same size and position from the mirror as the film plane, phase-detect autofocus may be possible too, making run-and-gun 16mm convenient. The holy grail would be a BMPCC OG type sensor that delivers beautiful digital dailies that are almost as nice as the film itself - and is calibrated with LUTs for different filmstocks, allowing the video tap to accurately simulate your exposure and color.

I'm also curious to try the new DJI Lidar system on a film camera...

[Karim D. Ghantous]

3 hours ago, Geffen Avraham said:

Tyler, I'd love to know more about this camera you're making! There was a company back in the day that made a transflective mirror, that could electronically switch between full transmittance and full reflectance. This could be a perfect replacement for a spinning or oscillating mirror - no moving parts, completely quiet.

The closest thing I have heard about is Red's Motion Mount. I'm sure there are similar products. For cinema cameras, I'd prefer an EVF to an OVF. That way, you could have a fixed pellicle mirror that transmits 1/10 of the light upwards towards the sensor.

[Tyler Purcell]

5 hours ago, Geffen Avraham said:

Tyler, I'd love to know more about this camera you're making!

It's a pretty cool leap in tech. 

It'll take a long time for us to build the prototype, but stay tuned! 

Some of the things you mentioned are absolutely being thought of. It's all down to the cheapest way to make something good. 

[Tyler Purcell]

2 hours ago, Karim D. Ghantous said:

That way, you could have a fixed pellicle mirror that transmits 1/10 of the light upwards towards the sensor.

A beam splitter, but yes that would be the idea. Similar to the Canon RS series still cameras. That beam splitter is made of a very special and proprietary coating, but it works very well. 

[Mark Dunn]

9 hours ago, Tyler Purcell said:

A beam splitter, but yes that would be the idea. Similar to the Canon RS series still cameras. That beam splitter is made of a very special and proprietary coating, but it works very well. 

Sony had another go at the pellicle for their later Alpha stills line, calling it SLT for "translucent", before it jumped on the mirrorless bandwagon. It works very well. I've had an A55/A58 since 2016. Sensor dust is almost a thing of the past.

Only drawback- it doesn't work in pinhole mode because the super-high DoF extends right back to the pellicle and sees every bit of the surface like a microscope. It's like photographing through measles.

Edited May 31, 2024 by Mark Dunn

[Karim D. Ghantous]

13 hours ago, Mark Dunn said:

Sony had another go at the pellicle for their later Alpha stills line, calling it SLT for "translucent", before it jumped on the mirrorless bandwagon. It works very well. I've had an A55/A58 since 2016. Sensor dust is almost a thing of the past.

Only drawback- it doesn't work in pinhole mode because the super-high DoF extends right back to the pellicle and sees every bit of the surface like a microscope. It's like photographing through measles.

Sony could make a film-digital hybrid camera using SLT tech. Film in the back, sensor above the mirror.

[Dom Jaeger]

On 5/31/2024 at 7:42 AM, Tyler Purcell said:

Yep, Daniil and myself are very much outliers and unfortunately, we need to duplicate ourselves 5 times and spread ourselves all over the world to keep these cameras working. In the US, I don't know anyone doing what I'm doing. Most other techs are focused on cameras like Bolex's and such, totally different market. 

Lol, anyone would think from reading this that Tyler is single handedly keeping film cameras working world-wide. I think he’s been doing it for about 4 years, compared to plenty of techs like myself who have decades of experience. 

It’s worth remembering that there are still quite a few movies and other projects being shot on film every year world-wide, plenty of rental houses that still rent film cameras and therefore techs that are maintaining them. I’m in my 50s, and not planning to retire for a good 20 years, and there are at least three other film cameras techs I know in Australia alone. Tyler often talks as if he has personally interviewed every tech in the world, but thankfully the world is a lot bigger than LA.

In the US there are techs at Panavision, Arri, AM Camera, Visual Products, Du-All, TCS and no doubt other rental houses that have film camera inventories that need upkeep. I don’t know what expertise they all have. In the UK there are techs at Panavision and Arri still working on film cameras, as well as a few other places. There’s a few in Europe as well, and I know at least two young guys there who are learning to service Arriflexes and other things.

The problem is that you do need expensive test equipment, tools and jigs, as well as the knowledge. That tends to mean rental houses are often the best places for servicing these cameras, as they have service departments with collimators, gauges, jigs and machine shops. That’s always been the way most techs learned the trade. They also often have inventories of parts left over from the film days, and manuals. You also often need one person for mechanical and optical issues and another to deal with electrical problems, as it’s beyond most people to master both. So a team of technicians rather than one guy.

Expecting older techs to simply pass on their knowledge is difficult. You can’t teach camera service in a week long workshop. You can’t even teach it properly over a year working with someone in a service department. It takes minimum three to five years of mentoring, especially if you want to know how to work on more than one or two camera models. After over two decades I’m still learning things. What I have learned well though is how to avoid causing damage, and being careful not to upset certain settings that were factory set. And that’s the main issue I have with people rushing into fixing cameras. Some of the worst damage I’ve seen was caused by inexperienced, sloppy or over-confident service techs.  

My take is that filmmakers should be supporting rental houses, who are best placed to train a new generation of techs. But they need enough people renting their cameras and lenses to afford a service team. 

But hey, if there are ten people willing to pay for my time I’d happily do a workshop on film camera service! 

[Geffen Avraham]

 

2 hours ago, Karim D. Ghantous said:

Sony could make a film-digital hybrid camera using SLT tech. Film in the back, sensor above the mirror.

I suppose even if you rate 500t at 3200 ISO, and give only 1/16th the light to the sensor, digital sensors today can handle 51200 ISO. Not well, but you get a usable video tap image.

[Aapo Lettinen]

1 hour ago, Geffen Avraham said:

 

I suppose even if you rate 500t at 3200 ISO, and give only 1/16th the light to the sensor, digital sensors today can handle 51200 ISO. Not well, but you get a usable video tap image.

I regularly shoot at 12800 iso on S5 + atomos and that is almost 4 years old mediocre camera, nothing high end at all. And have used the end results for cinema release on 4k dcp (local run + festival stuff), works fine.

I'm sure a newer sensor could handle basic high iso video tap use just fine when the modern sensors are so good that you could probably use the image for cinema release too even at very high iso 😄  the optical system before the sensor itself may degrade the image of course, mirror shutters have slightly uneven surface and may be just tiny bit out of balance or otherwise oscillating. but that might actually make the digital image more interesting by introducing small organic imperfections to it?