How far will Digital go.

Originally Posted by Eero Makela

H3DII-50

Even the Brochures for the Nikon D3 were taken with a Blad. wonder why?

I know why but had no idea hasselblads were so high in MP's

Originally Posted by Take12

Here's a a 56mp back for you.

But you'll need a whole system.

Not to mention another 100 grand for a fully equipped studio, a lighting system, and a whopper of a workstation* to process those ginormous image files.

*e.g. a Mac Pro Octo core with 32GB RAM, 2x 30" Cinema Displays, and 3x 1TB harddisks...

Have fun, D.!

Eww my kind reading thanks guys.

[Ed Shapiro]

MATTHEW- Thank you for the precise information. It is funny that old gremlins of film photography such as excessive grain and moire patterns again rear their ugly heads in the digital world and originating in different parts or places in the system- I thought theses are all issues of the past. Grain is now NOISE and moire patterns and Newton rings were defects related to the use of glass negative carriers and cover glasses in slide mounts. The preventative method of choice in those days was to incorporate ANTI-newton glass which was sandblasted enough to create a light frosting or irregularity on the surface of the glass and some photographers used a light coating, on the glass, of talcum powder. I wonder if there is any correlation to the ANTI-aliasing filter.

JON- There is nothing wrong with obsolesce when a new and better methodology or technology arises out of engineering which make for better quality and more efficient performance in all sorts of photographic gear. This is a natural progression in progressive thinking where the old gives way to and is left behind by the new. The negative aspect that I was alluding to is when obsolesce is engineered into a product whereby an expensive piece of gear will unceremoniously drop dead beyond repair after a certain period or cycle of usage. We have cardboard cameras for that not DSLRs costing many thousands of dollars.

As professionals, we must keep up with the times and the technologies that enable us to remain competitive. This impitus oftentimes requires judicious investments in the latest gear. Photo enthusiasts and hobbyists can perhaps make do with their equipment for longer periods and can go in for improvements at their option. In all cases, there is nobody holding a gun to anyone's head that forces one to jump on the bandwagon and buy everything as it is introduced.

It is a known fact that professionals photographers are not the best customer for new equipment. In the old school of though, pros purchased equipment to last for an entire career and film camera and analog equipment has taken a lickin' and keeps on tickin'- just as anyone who owns a Hasselblad, Rollei, an F series Nikon, a Linhoff or a Lica can attest to. Just ask Kodak (as if they will tell you the truth) why over the years the have shifted their advertising ands their equipment and supply lines toward the consumer market. The consumer market buys more and responds better to their advertising.

Cameras are now marketed like home electronics were marketed in the past; "If you don't have the latest and greatest, you are behind the times". Many consumers believe that a more powerful sound systems will yield better reproduction because the don't fully understand that the acoustical and electrical set up of their speaker systems are more than half the battle for realistic and authentic reproduction. It's the same with cameras; so many photographers think that an upgrade in their camera system will automatically improve their photography and we all know that ain't necessarily so!

There is no doubt that progress in digital imaging will continue at a rapid pace regardless of anyone's reactionary or radical thoughts. Like everything else, good things and good technology cost money and people who insist on state of the art equipment have to bite the bullet and write the checks. If one can't afford that, they are certainly not left out of the craft of photography, they just have to make do with what the have, improvise, innovate and apply their talent. To me, talent and artistry count far more than mega pixels.

Ed

[Jon]

Originally Posted by Ed Shapiro

JON- There is nothing wrong with obsolesce when a new and better methodology or technology arises out of engineering which make for better quality and more efficient performance in all sorts of photographic gear. This is a natural progression in progressive thinking where the old gives way to and is left behind by the new.

This is not what I was talking about, and I agree with this.

Originally Posted by Ed Shapiro

The negative aspect that I was alluding to is when obsolesce is engineered into a product whereby an expensive piece of gear will unceremoniously drop dead beyond repair after a certain period or cycle of usage. We have cardboard cameras for that not DSLRs costing many thousands of dollars.

This is what I was talking about. "Designed to fail" is not an urban legend or an old wives tale. Many companies have discovered (some of them - like the DeLorean Motor Company and Tektronix - the hard way) that if they don't design failure into components that they'll go broke before long.

A friend of mine in Toronto works for Magna in the role of redesigning parts for an earlier failure. For those that don't know, Magna produces many many automobile subassemblies for most of the OEM brands and nearly all of the aftermarket brands (TRW, Moog, etc).

A non-failing product>no replacement sales revenue/no parts sales revenue/no repair revenue>company is in serious trouble.

This is a reality of business.

It is not different in the "DSLRs costing many thousands of dollars" market. If you have an S2 with a failed shutter today (I do), the cost of having the shutter assembly replaced will be cost prohibitive when comparing to replacing the camera. That is, if the parts are even available.

[matthewlrigdon]

Originally Posted by Jon

This is what I was talking about. "Designed to fail" is not an urban legend or an old wives tale. Many companies have discovered (some of them - like the DeLorean Motor Company - the hard way) that if they don't design failure into components that they'll go broke before long.

DeLorean Motor Company

A car company that only manufactured cars for two years doesn't help your case at all, unless you're arguing that DeLorean failed to build cars that wouldn't fall apart after three months, necessitating a new purchase at that point.

In fact (damn those pesky facts) it seems DeLorean's biggest problem was that they DID build cars that fell apart after a few months. You could try and say that's planned obsolescence, but most of us would just call it shoddy workmanship.

Nothing lasts forever. There's no manufacturing process that produces products that never fall apart or break down. If that were possible, we'd all be running our houses off Perpetual Motion Machines.

As Homer Simpson would say "In this house we obey THE LAWS OF THERMODYNAMICS!"

One shouldn't confuse entropy with planned obsolescence.

But who knows? Maybe the oil companies really did buy that water-burning engine technology and buried it. Maybe we didn't land on the moon. Let me go grab my tin foil hat and think about it for awhile...

[Loa]

Hello,

I'll try to bring this discussion back to the furure of digital phorography per se, instead of the future of technology in general.

The original question, as restated, was: has the image quality plateaued.

In my mind we could be really close to a quality plateau if your only 2 outputs are small size prints or high res screens. Heck; even a 1080i tv can only display 2megapixels at a time!

But if you want to think ahead a few years, imagine displays a lot bigger and a lot denser than 1080i and think of printing 24X36 and bigger. Those tv screens and monitors are coming, no doubt. And as cost will go down, those bigger prints will become more and more popular.

Only that wouldn't make much of a difference. But think ahead a bit more and you'll find a lot of other sources for improvement in image quality.

Adobe already has a working prototype of a 3D photographic camera. It takes only 1 image, from 1 point of view. But you can play around with the focus location and depth. This is going to transform how we think about image quality, and it's not going to come in the form of a firmware update...

Imagine a sensor on which the individual pixels can determine their own optimal ISO and "speed" to capture the most imformation. So no more noisy dark spots in otherwise well exposed images, no more blown highlights, no more cranking up the ISO at the cost of noise... This will force us to think of dynamic range in a very different way.

Imagine a lens made up of a series of flexible lenses that can automatically find the optimal configuration for best light transmission and transformation. Zoom lenses without a single compromise. Can we even label such a lens using the traditionnal symbols? How will that affect image quality?

Our imaginations are just not up to the task... There will be a LOT more revolutionary things coming to the world of photography, and only one thing's sure: image quality will never plateau...

Loa

P.S. We're really bad at predicting the future. I mean: really really bad. In 1984, Ray Kurzweil got laughed at (by Kasparov) by suggesting that in 1998 a computer would beat the world's best chess player. Again in 1993 he got laughed at by Kasparov again for the same exact suggestion. That grand champion got beaten in 1997. Now just about ANY computer will beat 99.99% of the world's population at chess. Chess was regarded, up until the turn of the century, as the game of Kings. Now, we just don't say that anymore...

P.P.S. If you want a funny (with a few insightful moments) account of what famous photographers thought would happen in 2150, go here:

Shutterbug: Future Of Photography

Some of the suggestions made about 10 years ago, for 2150, are already "old news"!

[Ed Shapiro]

I am not thinking or saying that cameras and other gear are designed to self distruct as per a policy of manufacturers. I am saying that in some cases when more longevity can be built in to shutters and other major parts simply is not.

A good automotive analogy- remember the Checker? This was a car built from the late 40 into the early 70s. It looked like a 55 Chevrolet on steroids and was put together like a brick outhouse. The body was made of aluminum and the general construction was military tough. The interiors were made of hard wood and leather and the motors and transmissions were selected from other manufacturers which were making the better products from year to year. A car am have had a GM engine and Chrysler transmission depending on the model year. The cars were commonly used for Taxicab service in large urban cities and were also issued in a consumer model with slightly more attractive trim. 7 Passengers could comfortably sit in the back seating area- there was the main seat and two folding jump seats- another passenger could sit next to the driver.

Theses cars went in excess of three hundred thousand miles and oftentimes faulty engines or transmissions were simply replaced. As the consumer demand leaned more toward style, the Checker soon became a thing of the past but thousands of them were on the road many years after the went out of production. The technology was there to build an almost indestructible car but this was not, in the long term, a profitable policy for the company. Ever tale a close look at a army 2 1/2 ton truck?...I will say no more.

We live in a very disposable world and most things are not made or meant for lifetime service- perhaps tombstones?

Ed

[matthewlrigdon]

Maybe you should clarify if this discussion is about digital that's available to amateurs to mid-end professionals, or digital available if you have more money than god

The liquid-cooled CCD cameras that astronomers use to make telescope images are great and all, but I don't think anyone on this board can afford one. I know none of the astronomers can; that's why they have to sign up for months or years in advance to use one.

We're starting to see some slowdown as digital matures. We may even be running into some disadvantages to upping sensor size. There are some cases where the new 50D is outperformed by the 40D (not many, but a few). As you keep shrinking pixel size, you lose some quality (it's why full frame cameras out perform APS-C, why medium-format outperforms full frame, etc). It requires more engineering (and much more cost) to try and overcome these limitations.

It's not just cameras; even CPUs are hitting real limits. The move to multi-core processors has partly been necessitated by the problem of running a CPU past 3 ghz. The chips are running too hot. It becomes exponentially more expensive to build each succeeding generation of fabrication facility. And even when you shrink die size, there's still the need to try and lower voltage in order to control heat, which introduces problems of noise.

This is certainly applicable to cameras. It's not just a light-sensitive CMOS sensor. There's a custom processor alongside that has to make an image from it. And digital camera manufacturers can't afford the latest fabs. I believe Intel is still the only company with 13um fabs up and running and they aren't doing work for anyone but Intel. Plus, the aforementioned factors afflict camera manufacturers. If you were willing to lug around a 20 pound dSLR with a nitrogen cooling system and multiple processors, you'd have incredible image quality. You'd also have to spend a small fortune. And your arms would look like Popeye's

You also have to figure in what is asked of the object in question. As for the Checker car, the first thing to note is that only the body was "indestructible". All the other parts wore out, like all mechanical parts. But we started asking for things from our cars that a Checker can't provide, like gas mileage. It doesn't do you any good to own a 6 ton, indestructible car if you can't afford to put gas in it. Plus, a lot of those cars stayed running because they were used for business. They received regular maintenance and were kept in service to pay off the investment. Ten years from now, there will probably be more 1Ds Mk3s around than XSis, even though more of the latter are built. Part of it also has to do with behavior; we became a culture that replaces its cars every three years. It doesn't make any economic sense to build a car that lasts forever, no would enjoy it.

Figure that it's easier to manufacture something new than repair something old, why shouldn't you buy new stuff? It's been wasteful in the past, but we have also gotten better and better at recycling old items and breaking them back down. Fifty years ago no one would have imagined an item that was 90% recyclable. Most people couldn't even define the word recyclable.

[Ed Shapiro]

A few points to ponder- thank you!

The Checker- In those days people did not really care about gas mileage- gas was cheap and people purchased cars the size of yachts and what with the popurlarity vans and SUVs many people, nowadays, still do. Just the other day a saw a little old lady behind the wheel of a Hummer. In the days of the checker, I drove a Peugeot 404- it looked like an undernourished 55 Chevrolet, was fuel efficient and that rack and pinion steering was to die for in New York City what with all the parking difficulties, that is, if you didn't die from the toxic air or get killed by the muggers. People laughed at my car- it was kinda funky. Back in the 60s air pollution was not a big issue with most people and even when catalytic converters came out, people took pride in removing them in the interest of gas mileage even though it was illegal. Go figure- Ralph is still running for president!

I do realize that all mechanical parts eventually do wear out but on the old Checker, the Chevy Red Seal engine would have set you back around $1200.- still less than a new car. A Benz diesel will go 300,000 miles but installing a replacement engine is far from cost effective and at that point the body might be gone.

The idea of the miniature format (35mm) was to reduce weight, increase mobility and enable speed of camera handling. In return for theses features, photographers put up with a smaller negative. Then came the Nikon F-tn with its’ metering head, motor drive and 100’ back it outweighed my Speedgraphic with its fabulous 4x5 negative. No problem for me, I used a 4x5 hand held camera for shooting wedding in the olden days so my forearms did resemble those of Popeye- you should have seen the calluses in my left knuckles- folks though I was into Karate- splitting cement blocks. Today, many DSLR bodies are made out of some sort of polymer mix that weight is no longer a problem and they will do a decent job unless you expose the to lots of sunlight and great or store them in a working oven.

Lower priced DSLRs- No problem. I am not a dyed in the wool gear-head although I do love my gadgets and cameras. Thig is, I have seen photographic masterpieces made with D-70 and Rebels and boring lifeless images come out of top of the line cameras. At the present time I am looking at a Hasselblad digital system because one of my specialties in food and beverage photography and I have to make images that will end up on large point of purchase posters, large bulk head transparencies for restaurant price lists, billboards and the sides of transport trailers. It would be ludicrous to carry a $50,000 rig to a wedding but 8x10 transparency film (for my food work) is becoming rare, Polaroid 8X10 for testing is gone and I have no choice but to go to a big sensor camera in order to keep my clients supplied with the kind of files the require. Necessity is the mother of spending too much money in cases like this. My wedding and event works requires prints in the 8x10 and 11x14 range and I do produce the odd 30X40 wall portrait from formal wedding images- any decent pro camera will accommodate theses requirements.

For the average shooter a high end consumer model with enable fine results as long as the photographer knows what he or she is doing. With low volume shooters, longevity is not a big issue.

Photography is always been an expensive hobby especially if one is into high end gear. I have been a professional photographer for 50 years and have accumulated a goodly inventory of equipment but I know rank amateurs who own more expensive state of the art gear than I do. Everyone has his or her own spending priorities and their own budgets. If some people have unlimited funds and are independently wealthy the can buy whatever they want and who says God has all that much money?

Ed

[matthewlrigdon]

Things change often in digital land, so something that's true one day...

IBM has announced that they're offering foundry services to everyone. I mentioned somewhere in here that Intel had the best fabs on the planet and weren't interested in the contract market. Well, IBM has the second best fabs on the planet and apparently they can't generate enough work on their own. So Nikon or Canon or Sony (or all of them) can get access to 45nm, SOI goodness in their next generation image processors.

D300 users tired of 14-bit slowdowns? Throw a few million more transistors at the problem. Digic IV isn't fast enough? Bring down temperatures with SOI and up the clock speed.

[E-Man]

The issue of the progress of electronic capability has been fairly well described as Moore's law - Wikipedia, the free encyclopedia

Moore's Law (named for Intel co-founder Gordon Moore) roughly states that the number of transistors that can be packed into a fixed space doubles approximately every 18 or so months. For you and I, this roughly translates into electronic equipment (computers, hard drives, cameras, etc) that doubles in storage capacity and processing speed every 18 months. This trend has held since the middle 1930's and is predicted to continue for approximately another 20 years before we hit physical limits of size, power consumption, and heat dissipation of current silicon manufacturing processes.

One of the issues that is buried in Matthew's response (particularly the link to the IBM article) is that silicon manufacturing for creating central processing units in PCs is running up against a physical wall: the width of the electrically-conductive traces that are etched into the silicon during the manufacturing process. Current state of the art is at or very close to 45 nanometers (10 ^ -9 meters). As you make the traces more and more narrow and pack them more closely together, the electrons racing along the traces begin to exhibit bad behavior such as randomly jumping from trace to trace. Needless to say, this causes absolute havoc with the processor because things no longer behave in a neat and orderly manner.

This physical limit is just part of the physical realities that are behind the push for multi-core processors as opposed to a single processor that just runs faster and faster (and hotter and hotter). The width and density of the traces on the silicon chips are also related to heat-dissipation issues.

In the larger context, we are exploring the issue of a plateau for electronic capabilities. I don't think we have yet reached this plateau. I don't think we will be there for another 20+ years. By the time we get there, there is likely to be a different paradigm that governs the R&D efforts related to our electronic gadgets, enabling the growth in performance to continue without much interruption.

Will it stop? Not in my lifetime - and I still have quite a number of years to go..

[matthewlrigdon]

We can only hope that the 20 year estimates are right. It's not like we've ever made mistakes before (subprime won't ever be a problem. 640K of memory is all we'll ever need. Disco will never die.)

But besides the technical issues, economics are often far more important than any technical developments. You might want to find your local Circuit City (if it's still open). There's liable to be a bunch of good deals there over the next couple of weeks.

[Songman45]

We talk about film sometime as if it existed in its latest incarnation from the begininning of time. Improvements were made to that media and its processing without end. I remember the hurdle of 400 speed fine grain color print film. You know that wide dynamic range of film we so praise when comparing it to digital? It took a long time to refine.

We finally have a DSLR that behaves like a "real" camera. It hasnt been so long ago that an affordable digital camera with a shutter reaction close to that of a 35mm didn't exist. I remember having to anticipate a moving subject and even then I'd miss fast action shots. Its come to image quality now. More pixels alone doesn't necessarily mean a better image. I still believe there will be higher count sensors and improved rear screens for a while.

I predict advances will be made to increase the dynamic range. The high ISO/Low Noise issue is becoming a thing of the past. Now what needs to happen is some sort of layering (just like HDR) where part of the pixels will handle shadow, part midtone, part highlight. Will this happen in a year or two? Who knows. Is there a masterplan for easing the features into offerings coupled with engineered obsolescence? Maybe. Will we ever buy a digital camera and expect it to serve us 20 or 30 years like the old faithful K100, F1 or A1? I do not know.

I know that American automobiles (driven here, not necessarily made here) used to be headed for the bone yard at 100,000 miles. The mid seventies was a disaster for autos. That hump seems to have been crossed. We've come a long way with the addition of technology in cars. They perform better (remember altitude changes playing havoc with carburators?) are more reliable and run better longer than 30 years ago. Back then we were coinsidering inexpensive cars as disposable. I see 10 year old cars with over 150,000 miles for sale quite often and the prices aren't exactly junk prices either.

So with digital cameras I think the same evolution will play out. We'll eventually get to the point where improvements will be introduced much less frequently and the longivity will increasse. And thankfully your 3 yr old camera will not totally devalued.

Optimistically yours
Steve

[E-Man]

Originally Posted by matthewlrigdon

It's not like we've ever made mistakes before (640K of memory is all we'll ever need.)

Ha! There are plenty more just like this one! "Nope - we've never been wrong before!"

[JimFX]

Peeked . . . no way I am waiting to be able to take an HDR image with one shot.

[generic]

I've been waiting for that too. I think the only reason we dont have it at the moment is processor speed because consecutive dual power readings withing nano seconds seems possible. Betcha a dual camera/lens/setup/processor behind a prisim or split light path lens in one box would only need some some controlled blending processing power and walla