A Reverse-Engineered History Of The Camera Business
By Ctein
Eamon's wonderful inside look at the camera industry inspired me to add my own musings to his historical perspective. What I have to say complements it. My connections to the business have been very different from his, so we are fondling different parts of a very large and complicated elephant. Let's wind back the clock about 40 years and run forward.
For the 1960s and most of the '70s there was a vast gulf between professional cameras and amateur cameras. The margins on the professional cameras were considerably better and they conferred prestige on the amateur lines (what Eamon nicely termed the "mindscape"). Their volume was insignificant, though, compared to the amateur market; the price differentials were huge.
In the mid-'70s, Canon got the notion that they could mass-produce a high-quality SLR. Meanwhile, Minolta in a desperate bid to save their failing business decided to invest in truly innovative manufacturing technology. The cost of quality SLRs started plummeting, and Olympus, Pentax, and Nikon followed Canon and Minolta's prescient lead.
In the 1980s, inexpensive, seriously good SLRs became the norm. A substantial fraction of the huge amateur market bought a "real" camera. Boom times for the camera makers! But, as will always happen, the market and the technology matured. That's a really bad thing for sales. When everybody owns a decent, well-made camera and you can't hand them radical improvements every year, your sales projections are gloomy.
Happily for the camera makers there was a new shiny trinket to seduce the consumers—the point-and-shoot camera. Innovative technology and clever design drove sales. In turn, the P&S's drove the technology. Modern compact high-performance optics, electronics, and sensors and control mechanisms owe much of their existence to oft-maligned P&S's. As a class, they were the most sophisticated cameras ever produced.
But that market was also maturing by the end of the last century. What was waiting in the wings to keep the industry rolling? Guess!
The slow roll-up to digital
Electronic/digital photography didn't surprise anybody who understood the technology. It had been developing slowly, steadily, and predictably from at least the early 1970s (I would bet the late 1960s, although I don't have facts to support that). Modern digital camera designs date from no later than 1971, non-chemical printing technology as well. (Side note: I'm talking about commercial/industrial technology here. If you're talking military/espionage, they were and are 20 years ahead of the curve. But that technology almost never trickles down to consumer products.)
The first digital camera prototypes (much cruder than the farsighted paper designs) appeared mid-'70s. By the early '80s, Kodak was hinting that photofinishing was going hybrid and by the mid-'80s, primitive consumer electronic photography products started appearing. Indeed, around 1994 a Fuji rep even asked me if I thought she should be giving up her film camera. I told her, correctly, that digital was indeed the wave of the future but that the wave was made of molasses and it was still January.
It took about three decades for digital photography to slouch all the way to Bethlehem. Then it got a big kick in the pants from an entirely predictable event that few planned for. The dot-bomb. Everybody remember The New Economy? Sorry—meet the new boss, same as the old boss. And we will get fooled again, because it's in our nature to want to believe that the streets are paved with gold and money will grow on trees.
The dot-bomb was more of a disaster for the consumer/electronics industry than outsiders know. We're still in recovery mode: the economic metrics don't reflect how badly those businesses were hurt. The scale of the collapse was so huge that (contrary to some popular assumption) terrorist attacks produced only barely a visible blip in the curves.
Paradoxically, that was just the break digital photography needed; it was not quite ready for prime-time, price/performance-wise. Still, it was close and manufacturers were now in desperate straits. They could not wait for the normal transition to the next wave of shiny consumer-lust-inducing baubles. Like Minolta, decades earlier, they needed to gamble and force the market or they were going to die of starvation for sure.
That's when (and why) camera prices started plummeting and the performance/price ratio started rising even faster than it did for home computers. Manufacturers drove down prices aggressively, almost suicidally. They rolled out new models long before the old ones had paid off their investment. This was consumer pump-priming on a desperate scale, and it worked to our benefit. That long, ponderous 30-year swell suddenly hit the uncharted reef of May, 2000 and slowly but irresistibly broke into a perfect wave. With a speed that couldn't have been sensibly predicted pre-dot-bomb, it drowned film photography (and several companies) and it floated whole rafts of new products and services.
We're still riding the wave today and it's a wild trip. I'm kind of wondering, though. What's in the cards when this market matures? It will, you know, and then the industry will have to move on to the Next Great Thing. I wonder what that will be?
______________________
Ctein
Featured Comment by Ken Tanaka: "It's fun to make predictions. But it's more fun to see past predictions. Rochester Institute of Technology has preserved a page of predictions made by several renowned photographers of the day and published in a 1944 issue of Popular Photography. With this being at the end of WWII there is a recurring 'post-war' perspective. Have fun!"
Mike adds: I'm impressed by Eliot Elisofon! Film never miniaturized as much as he predicted, and, in color negative film, dye clouds merely replaced the light-sensitive silver halide layers. But he correctly predicted the explosion of photographic education in the 1970s, and perhaps we can grant that his "built-in lens turret" is the zoom lens, in effect if not in execution. Pretty impressive predictions.
Ctein: "It will, you know, and then the industry will have to move on to the Next Great Thing. I wonder what that will be?"
Bankrupcy and corporate takeovers for the smaller players. For the larger beasts...reduce costs and try to survive until the next generation of consumers starts to get the photography bug (and has sufficient disposable income to satisfy that bug - in about 20-30 years).
I think you underestimated the role demographics has played in the entire consumer camera world. I don't think you would have been able to sell millions of $1000 cameras (sans lens) if there weren't so many people (sorry, being U.S.A.-centric here) at an age where they had sufficient disposable income and time to use the cameras.
I think it is no coincidence that the digital "explosion" happened right when the "boomer's" kids are leaving the "nest".
Posted by: Jeff Hartge | Tuesday, 01 April 2008 at 11:44 AM
"The dot-bomb was more of a disaster for the consumer/electronics industry than outsiders know. We're still in recovery mode: the economic metrics don't reflect how badly those businesses were hurt. The scale of the collapse was so huge that (contrary to some popular assumption) terrorist attacks produced only barely a visible blip in the curves."
Ctein, statements like this could really use some evidence to back them up. Otherwise it's hard to take you seriously.
Posted by: Chris | Tuesday, 01 April 2008 at 12:24 PM
My prediction for the next great thing: Computational Photography.
You can see some of it here already in the post-processing tools currently available. I'm talking HDR and super-resolution. The computer combines multiple exposures to produce data unavailable to a single exposure.
While we currently have to take several exposures over time in order to take advantage of these technologies, that may not always be the case.
Posted by: Ron H | Tuesday, 01 April 2008 at 12:48 PM
"We're still riding the wave today and it's a wild trip. I'm kind of wondering, though. What's in the cards when this market matures? It will, you know, and then the industry will have to move on to the Next Great Thing. I wonder what that will be?"
Improvements -- as film improved between ~1890 and ~1990.
Most of the radical changes in consumer technology since 1950 have been the result of a single invention, the transistor, played out and elaborated over various fields. The unbelievable home computers we have now are no theoretically different than the one I used in 1978 -- thirty years ago -- just much more elaborated. Digital cameras are simply the application of a wildly elaborated transistor technology to the field of light sensors. [Okay, this is somewhat of an over-simplification.] What you're seeing is not the beginning of a new technology, but the maturing of one that is fifty years old. Until there is another fundamental development in electronics, I think what we'll see with cameras is further elaboration of this older technology.
I also doubt that the military is twenty years ahead of the curve. The moon landings were in 1969, and I suspect the lander had as good technology as they could provide at the time. In 1977, the Apple II was introduced, and sold-over-the-counter. I'm told the Apple was more powerful than the computers in the lander. There were only eight years between the two events.
JC
Posted by: John Camp | Tuesday, 01 April 2008 at 01:02 PM
Dear Folks,
Oops! Just occurred to me to say that this is NOT an April Fools Day column like last year's was.
The coincidental similarity of topic and posting date didn't occur to me until just this moment.
pax / chronologically-challenged Ctein
Posted by: Ctein | Tuesday, 01 April 2008 at 01:18 PM
The next big break through?? How about good photography. Pixel peeping and the 100% crop are the new kings surfing the digital wave. Enough. Just show me some nice pictures please.
Posted by: Emm Jay | Tuesday, 01 April 2008 at 05:06 PM
Dear John,
No I don't think the military is far ahead of the curve in all (or even most) technologies. But in the areas of electro-optics and advanced photographic technology, they really are. Some of the things I know they can do are quite amazing, and I don't even have a clearance!
Mostly I stuck that clause in to make clear the limited scope of what I was talking about. For example, you have to wind the history of digital-electronic printing back by more than 20 years if you include certain military/espionage endeavors (I've seen some stuff printed in the early 60s that would just blow your mind). But these are people who operate on a "money is no object" basis and most of the technologies simply aren't ones that you can't commercialize at any price.
Speaking of military aerospace, as well as photography, are you old enough to remember when Skylab went up and the solar panel didn't unfold properly? Did you ever wonder how NASA knew within a few hours not just what was wrong but precisely what tools and jigs the astronauts would need to have to fix it? There is a story there. Maybe some other time [ secretive smile ].
~ pax \ Ctein
[ please excuse any word salad. MacSpeech in training! ]
======================================
-- Ctein's online Gallery http://ctein.com
-- Digital restorations http://photorepair.com
======================================
Posted by: Ctein | Tuesday, 01 April 2008 at 05:59 PM
The big question now is, as I see it: who will survive and who will drown as a result of the current economic crisis, which probably is bigger, and will last much longer than anything we`ve seen for the last twenty, or thirty years?
What about Leica? Pentax? Not to mention producers of tools related to cameras (soft- and hardware).
Any thoughts on that, anyone?
Posted by: Paul Norheim | Tuesday, 01 April 2008 at 06:12 PM
Dear Jeff,
I do think you're right. These things are always a confluence of circumstances, and disposable boomer income must be a part of it.
As the old joke goes, timing is
Everything.
[ahem]
Similarly, had the dot-bomb happened a half dozen years earlier, digital wouldn't have been close enough to prime time to be a life preserver for the manufacturers. And had it happened a half dozen years later, they wouldn't have had to scramble desperately to produce salable products. The timing was by happenstance quite perfect (for some value of the word "perfect").
~ pax \ Ctein
[ please excuse any word salad. MacSpeech in training! ]
======================================
-- Ctein's online Gallery http://ctein.com
-- Digital restorations http://photorepair.com
======================================
Posted by: Ctein | Tuesday, 01 April 2008 at 06:14 PM
Dear Chris,
Well, the thing about an 800 word column that summarizes a third of a century of manufacturing history is that there are going to be lots of unsupported statements. You could just as easily question my claims for digital cameras circa 1970 or hybrid printing systems circa 1980 or why Minolta did what it did. I didn't provide any support for those statements either.
If you can state with sufficient specificity why you don't accept my remark, I could attempt to address your objection directly and concisely. Otherwise, it's hard to take you seriously, too.
Let me put it another way: what specific data do you have to refute my statements? I'm always willing to learn something new.
~ pax \ Ctein
[ please excuse any word salad. MacSpeech in training! ]
======================================
-- Ctein's online Gallery http://ctein.com
-- Digital restorations http://photorepair.com
======================================
Posted by: Ctein | Tuesday, 01 April 2008 at 06:20 PM
Further expansion of digital video into the lower end of the consumer market will happen I think. As storage tech advances, both in capacity and minaturization, and bandwidth increases.
Looped video picture frames, and jewerly, phones. Video clips/loops will eventually replace the digital capture still snaps. Packaging itself may be moving images with sound.
Still imagery will itself become the province of the enthusiast and artist.
Posted by: Jay Moynihan | Tuesday, 01 April 2008 at 06:44 PM
Ctein
Yes, the Military is quite good at some aspects of electro-optics and advanced photographic technology and they are far more open about discussing it than they used to be. Consequently, many “Space Cadets” know the Skylab imaging story quite well. [I must confess to being one myself and was responsible for both organizing and executing imaging at Rotary Rocket some years ago.]
A quick goggle search brought-up the following:
KH-8 reconnaissance satellite imaged Skylab:
What was _not_ disclosed at the time was that the NRO used a KH-8 reconnaissance satellite to image the Skylab. This was done entirely on the initiative of the NRO, not at NASA request.
Supposedly the photos were extremely good and were also extremely helpful for the repair mission.
That story was revealed in 1995 at an Air Force conference
Source: http://www.friends-partners.org/pipermail/fpspace/2003-April/008337.html
For introductory reading, “Deep Black” by William E. Burroughs, was billed as an expose of the secret world of US Spy Satellites and was hyped as having upset many in the community!! Regardless, it is a very good introductory read on the subject. I certainly found it interesting. [Usually available second-hand as it is out of print, I believe.]
Some very nice work on your website. Congratulations.
Best,
Geoffrey
Posted by: Geoffrey V. Hughes | Tuesday, 01 April 2008 at 07:06 PM
Dear Geoffrey,
You were involved with Rotary Rocket? Boy, it's a small world! Gary Hudson's an old friend of mine, going back to the early 1970s. Way back when, I was involved in efforts to secure him military funding. Sadly, it never jelled (Gary wasn't as good at playing the Big Boys' games back then.)
Haven't talked to him in some years (maybe 10?) But that's just because our paths haven't crossed. Do you still run into him? If so, say hi for me, please.
You're right about those folks being not as secretive as they were. I'd be willing to scale back my claim of them having a twenty-year lead time to 10 years [ compromised smile ].
Thanks for the praise about my work. Always appreciated!
~ pax \ Ctein
[ please excuse any word salad. MacSpeech in training! ]
======================================
-- Ctein's online Gallery http://ctein.com
-- Digital restorations http://photorepair.com
======================================
Posted by: Ctein | Tuesday, 01 April 2008 at 08:31 PM
SLR-quality liquid lenses coupled with post-capture focusing. Yes, web-archaeologists of the future with your funny biotech tentacles, I am Nostradamus...
Posted by: Puplet | Tuesday, 01 April 2008 at 09:31 PM
The salient thing for me is which parts of the elephant specifically are ctein and Eamon fondling? Are Britney or Madonna or Keith Olberman anyway involved? I need to see video. Still pictures are too easily manipulated.
Posted by: mike b | Tuesday, 01 April 2008 at 10:57 PM
What's coming?
Bob: Camera
Camera: Yes, Bob
Bob: Shoot black and white, 24mm, where I'm looking
Camera: That's your shoe Bob
Bob: Not there.. here
Camera: click
Posted by: Bob Dales | Tuesday, 01 April 2008 at 11:00 PM
"Electronic/digital photography didn't surprise ... I would bet the late 1960s, although I don't have facts to support that..."
The CCD was invented in 1969 and there are claims of simple imaging devices in 1969, see http://en.wikipedia.org/wiki/Charge-coupled_device.
jfb
Posted by: Joel Bartlett | Wednesday, 02 April 2008 at 12:53 AM
great read, Ctein..man, i love you guys, you're so nerdy.........and so smart!
Posted by: dyathink | Wednesday, 02 April 2008 at 01:12 AM
I would guess that Ctein's statement about the military being 20 years ahead is actually correct. The key is what aspect you are measuring.
My guess is the optics, sensors and general physical hardware in the spy programs are not necessarily 20 years ahead (I think there is a point where more resolution doesn't necessarily give you more valuable intelligence).
However, I would bet my money that the various intelligence agencies take millions of very high resolution images during the course of a year. I don't think they have a human staff big enough to look at all of those images. My suspicion is that they are 20+ years ahead in image processing software. Specifically object recognition, frame to frame object tracking, etc.
I would not be surprised if the average satellite intelligence team member has never seen a satellite image. Instead these members interact with a database of "known" objects and object locations that the software compiles automagically (possibly with a little human intervention).
Something along these lines would definitely be 20+ years ahead.
(For the record...I have NO evidence to support anything I have said. No winks. No smiles.)
Posted by: Jeff Hartge | Wednesday, 02 April 2008 at 10:56 AM
Re the RIT article, I found this amusing and familiar:
"It is well known that our present day printing papers have nowhere like the tonal range (latitude) and brilliance which the older gold chloride and albumin papers had. And what good is a photograph as a serious expression, unless one can count on the print's surviving at least a generation or two? Photography must realize its destiny as the "language" of the 20th century. "
http://people.rit.edu/andpph/giants/POP-PHOTO-future-1944.html
Posted by: Roger S | Wednesday, 02 April 2008 at 04:02 PM
Roger S.,
Same as it ever was, eh? I've been reading much the same comments my entire life....
Mike J.
Posted by: Mike Johnston | Wednesday, 02 April 2008 at 06:24 PM
Dear Jeff,
I suspect there is a lot of truth in your guesses. I do know that over 30 years ago (in the early 1970s, I think) the government contracted with Kodak to develop the capability to obtain photographs with better than 1 inch ground resolution from Earth orbit. This is physically possible (just barely) but phenomenally difficult. Kodak succeeded!
They were mightily ticked off when the government never issued a full-blown production contract after the proof-of-principle. The folks who asked for this realized that they had no idea of how to make use of such a huge amount of information with the technology of the time.
15 years ago I used to regularly see articles in my journals about massively parallel optical processing of images for the purposes of pattern and object recognition. The bandwidths were truly impressive. I haven't been seeing any such articles of late. I think that means one of two things, either the technology came into real use, so they don't want to talk about it anymore or else conventional electronic data processing surpassed it. Either way, the number of teraflops involved is awesome.
~ pax \ Ctein
[ please excuse any word salad. MacSpeech in training! ]
======================================
-- Ctein's online Gallery http://ctein.com
-- Digital restorations http://photorepair.com
======================================
Posted by: Ctein | Thursday, 03 April 2008 at 03:20 AM
interesting analysis, but completely disagree with you on the dot com bomb being the impetus behind the growth of digital photography. While that event may seem like a tipping point, realistically its just another example of Moore's law in effect
Posted by: Gary R Boodhoo | Thursday, 03 April 2008 at 10:04 PM
Dear Gary,
I'm not asserting the dot-bomb caused the growth of DP--half my column points out that the growth was ongoing for decades. But there was a substantial discontinuity in the growth rates and the price/performance ratios--a most pronounced kink in the curve.
I won't claim it was entirely due to the dot-bomb; alternate hypotheses are welcome.Remember--it's a big elephant. I will swear it wasn't 'business as usual.' T'weren't just ML.
Moore's Law is good for analyzing price/performance on a decade scale. It's not good for telling you what's going on from year to year. RAM street prices are a good example-- from 1990 to 2000, they do about what you'd expect. But from 1991/2 to 1995/6 (2 full generations in ML terms) they hardly change at all.
ML is boring from an historical perspective because it's so predictable. The stuff which doesn't obey ML is interesting. Mechanical storage (hard drives), for example, have improved their p/p ratio much faster than ML over a span of several decades. No one expected that semi-mechanical manufacturing could outpace silicon fabrication over the long term. Messed up a lot of planning.
But I digress (frequently). Main thing is there is a distinct, substantial kink in DP starting early 2000's. Explanations for why are solicited.
Posted by: Ctein | Friday, 04 April 2008 at 12:37 PM