Why is the rum always gone? Some thoughts on digital camera design.

Jock Elliott

Hall of Famer
Location
Troy, NY
In one of the Pirates of the Caribbean movies, Captain Jack Sparrow asks the plaintive question: “Why is the rum always gone?” (The answer: he drank it all.)

Well, I have a plaintive question of my own: Why are there no “classic” digital cameras?

The answer: they don’t live long enough to become classics. By “live long enough,” I mean that in five years or less, the technology within the camera is totally eclipsed so that it is (A) non-repairable, (B) the essential electronic components are no longer manufactured (the imaging and processing chips) and (C) it is not worth keeping for decades.

Further, I guarantee that within six months of buying your new, state-of-the-art piece of imaging gear, the same company will announce the next new and greatly improved version.

I think there may be an answer, if any company is brave enough to embrace it, but first let you, me, and Mr. Peabody step into the way-back machine for a moment. Back in the 1960s, Nikon manufactured a film-based modular camera system. On the same built-like-a-tank camera body you could change backs, prisms, view and focus screens, winders, adapters and the like so that the same camera could be morphed into whatever application you had at hand, whether it was high-speed photojournalism or making images while peering through a microscope.

Why not start with a great camera body (which you have bought with the idea of keeping it) to which you attach your favorite lenses (which you also get to keep). When imaging technology changes, you undo a couple of screws, slide out the imager, associated processing electronics and the back screen which are built as a unit, and replace it with the latest gee-whiz imager, processing engines and high-zoot back screen. You keep all the rest (including your EVF, which may be swapped out from time to time); the cost is lower both to you and the manufacturer, and I bet there would be less resistance to upgrading.

The Ricoh folks kind of approached this idea with one of their cameras (the EXR, maybe?), but they made the tactical error of marrying the lenses to the sensor modules.

I think we need a better brand of rum, one that doesn’t become obsolete so easily. Then maybe it won't be always gone.

Cheers, Jock
 
It's an interesting idea, Jock. But I have to wonder whether the consumer would end up spending more money with that approach versus simply buying a new camera every 3-5 years. I'm certain it could be done but I have to wonder about pricing. The camera makers are going to make sure they make their money on the deal.

And then there's the question of human behavior. Professionals might prefer the modular approach (note that digital SLRs intended for pro use usually have much longer product lives) but many enthusiasts simply want something new periodically. The general public will still shoot with their ever-improving smartphone cameras.

But it seems the current trend of every camera maker introducing a new camera every 12-18 months can't be sustainable either. Fuji has tried to slow that cycle (perhaps by 50%), opting instead to improve their cameras via firmware updates. So perhap that's the first step in the right direction. Maybe Fuji's X line will be classics in more than just style.
 
I think the GXR was pretty much what you suggest. They married the sensor to the lens so that each module could be optimized. That way you could have a superzoom with a small sensor and prime lens with an APS-C sensor. The modules sensor never really becomes "obsolete" since I still see wonderful shots taken with them still. Maybe they will try again and keep the glass and the sensors separate.
 
In simple terms, the sensor was updatable for film cameras. If we could update that, it might be enough to let us keep the camera. Of course, AF and EVFs are still moving targets so we'd have to concentrate on a manual body like an old Nikon F or Spotmatic.
 
I think the GXR was pretty much what you suggest. They married the sensor to the lens so that each module could be optimized. That way you could have a superzoom with a small sensor and prime lens with an APS-C sensor. The modules sensor never really becomes "obsolete" since I still see wonderful shots taken with them still. Maybe they will try again and keep the glass and the sensors separate.

I could never understand why Ricoh didn't introduce a K-mount module for the GXR. Pentaxians were begging for one. The M-mount module seemed to work out well. On the other hand, perhaps we have our answer in that the GXR was discontinued. The broader market simply didn't embrace and support it.
 
In simple terms, the sensor was updatable for film cameras. If we could update that, it might be enough to let us keep the camera.

Exactly!


Of course, AF and EVFs are still m ovine targets so we'd have to concentrate on a manual body like an old Nikon F or Spotmatic.

EVFs already are modular in some cameras. I am assuming (perhaps wrongly) that you could swap the AF electronics with the rest of the package, but my ignorance is profound.

Cheers, Jock
 
The more I think about it, perhaps "classic" in the digital age is defined by camera line rather than individual camera. For example, Pentax has used the same body and control/user interface in the K-7, K-5, K-5 II/IIs and K-3. The improvements have been essentially to the sensor, autofocus module and image-processing engine.

The same might be said for Canon's Rebel line but I am not as familiar with that. And, as I posted above, Fuji has tended to keep individual models around longer while improving them via firmware update. But the X-E2 is almost identical physically to the X-E1. The same goes for the X100 and X100s. We'll see about the XPro 2 versus the XPro 1 and the rumored X200. But we might be onto something here.
 
anybody think the epson rd1 or the leica m8 or the leica digilux 2 or the 5d are classics? i recently bought an rd1 and love it. i also think the original x100 either is or will become a classic.

further thinking about this, with film the quality result depended on the lens and film used, much less if at all on the camera, which was more a feature set for user experience. with digital the camera plays a pivotal role in IQ, and did not like film start from pretty much the peak IQ one can get from the medium. there was an image quality ramp up time for digital cameras that really didnt exist with film. i think we're about at the point where IQ improvements will be so minimal that we will begin to see more classic digitals.

and this is probably a bad time to talk to me about how sturdy film cameras were compared to digital as my beloved xpan just died. unless its fully mechanical--and even then as any russian medium format camera owner knows--electric parts die in film cameras too. and mechanical parts wear out. pretty much everything wears out, no?
 
I think we're about at the point where IQ improvements will be so minimal that we will begin to see more classic digitals.

+1. And we're only just arriving at that point now. Sure, the camera makers will bring out cameras with higher and higher resolution but, in the real world, the benefits will be incremental while there will be fair amount of complications resulting from ever-increasing file size.
 
Dare I say that the camera manufacturers, whether or not they come up with something completely new and awesome, will continue to market products as if they are, and those of us with incurable GAS will continue to buy. Incremental improvements aren't worth paying for, IMO - which is why I still have my X100, not an S model, a K-5, not a K5II/S, Nikon 1 J1 and V1 rather than V2, v3 etc. Its not because of anything other than financial restriction but I'm glad to be restricted because I really dont see the value in the "upgrades" of those systems. I'm sure many of you can, and thats fine, too...

The words are profit, shareholders, economic gain and any other word you can think of that represents what they are doing. Seriously...
 
As much as I find the basic idea sympathetic, I don't think it'll work, at least not if you want a small camera. Modern digital cameras are as much about computer power as they are about sensors and lenses. Much of the progress in sensors - higher resolution and higer fps in particular - also requires more computing power. The same goes for ever more advanced jpeg algorithms. This part of the camera is usually packaged in such a way that it takes up as little space as possible, which is generally speaking what consumers - and us SC'ers in particular - want.

A modular camera with replaceable sensor would also require a modular computing parts, which would inevitably require a larger body than the shrink-wrapped packaging that modern manufacturing can offer. Think of it as a desktop PC versus a tablet or even a modern laptop - the desktop is fully customizable but much, much larger.

Also, new technology - think curved sensors, or new stabilization systems - might not fit in so easily (for instance, Olympus had to put the gyroscope for their OM-D's stabilizer in the viewfinder hump for lack of suitable space elsewhere). To take such changes into account, the design would have to be extra spacious. Wi-fi / NFC / whatever is on the horizon in terms of connectivity requires antennae or similar technology, adding yet more to the list of things to make space for in your modular body. You'd end up with a giant box of a camera, not unlike a medium format body, which is something I'm not remotely interested in.

A fixed interface coupled to modular packages of lens, sensor, computing power and other features would work better, but offer little benefit over just using a constant interface on a range of complete cameras, as sensible manufacturers already do nowadays.
 
I pray that I remain content with my 2012 model X-Pro1 for a while longer yet. 2 years it's been now. That's about as classic as I have managed.. Hmmm Perhaps I had my Nikon D90 longer before I sold it.. Or perhaps my E-PL2 is older.
 
Why not start with a great camera body (which you have bought with the idea of keeping it) to which you attach your favorite lenses (which you also get to keep). When imaging technology changes, you undo a couple of screws, slide out the imager, associated processing electronics and the back screen which are built as a unit, and replace it with the latest gee-whiz imager, processing engines and high-zoot back screen. You keep all the rest (including your EVF, which may be swapped out from time to time); the cost is lower both to you and the manufacturer, and I bet there would be less resistance to upgrading

We live in a society that, sadly, is unable to perform even the most basic tasks–I bet that 8 out of 10 car owners can't even locate the dipstick in their car. And no, it's not in the mirror. So I'd hazard a guess that this kind of approach will be nigh impossible for most, and I'd hate to be the QC guy at the camera company having to test the latest gee-whiz imager for compatibility with all the high soot screen options, EVF options, image processor options, etc…

Come to think of it, it sounds like asking a consumer to build their own computer from component parts, which is possible but very few people do.

You're posts are always thought provoking though, so keep 'em coming!

The more I think about it, perhaps "classic" in the digital age is defined by camera line rather than individual camera. For example, Pentax has used the same body and control/user interface in the K-7, K-5, K-5 II/IIs and K-3. The improvements have been essentially to the sensor, autofocus module and image-processing engine.

Yup. Pentax found some secret sauce with that body and good on them for not changing things for the sake of change. And the fact that they do make slight tweaks (i.e., the built-in grip on the K-5 is excellent but they somehow made the grip on the K-3 more excellenter, and the bottom of the K-3 battery grip is raised ever so slightly so that you can swap batteries while on a tripod) demonstrates an attention to detail that I appreciate.
 
Back
Top