[Aristophanes]

These examples more or less support the power of marketing, though. Most examples you listed supported my argument as well as yours. We've got people who buy hi-def TV's that won't even take advantage of the resolution advantage, but it's the 'next big thing' so they are all over it. I know people who have bought blu-rays to play on their standard-def TV. The blu-ray offers no advantage at all for their situation, but they buy it because it is "better" since the marketing man told them so. The same thing can and will happen with FF - people buy it whether they need it or not, because they know it is better. I bet we'll see people buying FF and then shooting only APS-C lenses, but a manufacturer would be stupid not to capitalize on that and get a few of those sales.

I'll grant you that the wafer process introduces inherently higher costs for FF, but those costs can still be made so small as to be negligible. When the cost difference is only $100 total or even less between FF and APS-C, you can be sure that some company will take advantage of that and sell a FF camera at APS-C prices.

Whether or not Sony has done well isn't really relevant. They haven't done that well with any of their cameras as far as I can tell. The argument isn't about right now, but that eventually, FF will be ubiquitous and inexpensive. Also, larger than full frame isn't part of the argument for an obvious reason - it would require an entirely new line of glass. With FF, though, all we're doing is taking advantage of potential in the lenses that is sitting unused. None of the mounts used by any of the big players were designed for APS-C, so we're basically using FF cameras right now, just with all the limitations of APS-C.

As well, I know that you think FF has drawbacks, but it really doesn't. Right in the middle of your FF sensor is a nice patch of sensor exactly the size of APS-C, so whether you have the body crop automatically or do it later you still have the exact same result, assuming constant performance per square mm of sensor. So, what would you rather have, a camera that can only do APS-C, or a camera that can do APS-C and FF? That is all we'll be choosing between.

FF can fit comfortably in the same form factor as APS-C. So all things can be equal save the APS-C could go into a smaller form factor, but then it would be a P&S and to many who need advanced controls, unusable. MF sensors are too big for convenient. portable carrying. FF is the sweet spot, not APS-C. APS-C is the economic compromise in an era where wafers were the main cost. Now, marketing and distribution are the major costs. For Canon and Nikon especially, consolidating both sensor and lens development around a single size sensor will make more economic sense in the long run.

When Canon (and it will be Canon) decides to put an FF sensor in APS-C body at an APS-C price point, APS-C is done for. This will happen as a matter of economics and competitive advantage. Canon will see that integrated development can transfer capital from APS-C to FF and still realize a profit, especially if in doing so they can take market share from brands stuck at APS-C or 4/3, and for whom such a capital transfer would be prohibitive to their profits and market share. I think Oly and Panny "get this" which is why they've pretty much thrown it all into a much smaller sensor, because APS-C is a dead zone once FF comes to play at lower price points. If you're going to compete on sensor size and price therein, you need to be a mile apart.

When will this come to be? I used to think 6-8 years, now I am sure it will be within 4 years. Canon is itchy and has been losing share. They have the most to gain by competing on price and using their industrial muscle and sensor development o push the agenda; Nikon the most to lose. Where Sony stands no one really knows, but Sony is riven by internal battles and terrible market execution across its lines (its TV sales are terrible compared to its 20 year average as Samsung and LG destroy the previous Sony hegemony). Heads are rolling at Sony and there's lots of chatter, obviously. IWhatever Sony does will not change the basic dynamic that FF will eventually replace APS-C.

We are surrounded by DSLR purchasers who never take the camera off Auto mode. People over-purchase and will continue to do so. By this habit, they facilitate better tech at relatively the same cost as the older tech. More advanced users,like those in this Forum typically, are the beneficiaries. Unless you are heavily and inflexibly invested in DA Ltds for 20 years, this is a good thing.

[nosnoop]

1. I have established this with actually some pretty decent logic. That being: digital era technology has, in every single case, become so cheap that today's top-of-the-line stuff will be in the bargain bin in only a couple of years.

No, you have NOT established anything! Your analogy and your "logic" was flawed.
Right off the bat, your 6MP sensor reasoning was wrong.

The main problem is that you think sensor size is the only feature left for the manufacturers to sell - and that is not true. With the rise of the mirrorless, there will be more and more features being introduced for the camera, or what would eventually become an media capturing device. And I don't think the extra cost of FF sensor would become "negligible" any time soon. I believe Canon and Nikon would keep FF at the top niche DSLR to command a healthier profit margin.

[jogiba]

I agree with alot of what you all have said.

My personal view is that Pentax will release an EVIL full-frame camera in the near future. They have already invested in some new full-frame capable lenses ie. DFA 100mm and have the legacy lens designs as well they could dust off and give a digital coating to. As has been said, the reality is that FF sensors will be cheap enough in the very near future to accomodate sticking one in a k-7 like body and taking the mirror out. FF is a better image than APS-C and there is one overiding thing that everyone wants, and that is a better image. And EVIL is the way to go. The mirror is redundant and causes issues that can be avoided by it's removal. No real reason to leave it if the electronic viewfinder can keep up.

FF is the future of Pentax if Pentax is to have a future. I've already bet a fair mount of money on this i the form of lenses, and hope they don't dissapoint.

Cheers,

Jake

A FF EVIL from Pentax would sell 10x better than the 645D IMHO. I have many FF K mount lenses and so does millions more. How many people have lenses for the 44mm x 33mm sensor in the 645D compared to 35mm/FF K mount lenses ?

[TaoMaas]

When Canon (and it will be Canon) decides to put an FF sensor in APS-C body at an APS-C price point, APS-C is done for.

That price point is going to have to be around $500 or else you can pretty much cut the number of people jumping from P&S to DSLRs in about half or less.

[mattdm]

That price point is going to have to be around $500 or else you can pretty much cut the number of people jumping from P&S to DSLRs in about half or less.

As I see it, in the medium-term future (5-10 years), p&s cameras are done for. They'll be wiped out at the low end by cell phone cameras, and at the high end by EVIL. The interesting question is whether there will be space between (and beside) the EVIL cameras for APS-C. If full-frame can get to the size of the K-7, there probably won't be. But if it can't, there's still a niche there.

All of the camera companies must be agonizing over how to best play this, because you don't want to miss the boat — or jump too soon.

In the longer term (10-20 years), electronic viewfinders will exceed optics in all meaningful ways, and dSLRs will be history too.

(In 20-50 years, lenses will be obsolete too. But that's another thread. )

[Winder]

Personally I want to see a Maymia 7 type camera. Medium format sensor in a range finder body with a couple of fast primes.

[Eigengrau]

No, you have NOT established anything! Your analogy and your "logic" was flawed.
Right off the bat, your 6MP sensor reasoning was wrong.

And I don't think the extra cost of FF sensor would become "negligible" any time soon. I believe Canon and Nikon would keep FF at the top niche DSLR to command a healthier profit margin.

You misunderstand my argument. First, I am not talking about right now, but about the most likely trend in the future.

Second, believe whatever you want about the cost of a FF sensor, but already the sensor has gone from being the single most expensive component in a camera to just another part to plug in. Honestly, look at the state of the SLR market even 5 years ago! We've got cameras now that put that generation to shame, for much cheaper. What makes you think that the future will be different? Today's pro model will be in tomorrow's bargain bin. All I'm expecting is for the future to be like the past, while you for some reason are expecting that APS-C sales will continue as they always have been and ignore the changing economics of the sensor business.

Where's the flawed logic?

[Groucho]

Mountain Vision - I'm a lifetime American too, so I tease with love about the whole "unwashed American" thing. However, the fact remains that Americans tend to forget that there's a whole world out there, a world that doesn't care about football and Nascar and big SUVs and some other things that many Americans care about. (OK, these are things that I don't care about, either! Give me WRC and small sports cars any day!)

Sony has not done well because of market timing. Just after the A700 and A900 were released Canon introduced the 5DII and HD video became the "must have feature" for DSLRs. Not even Canon knew how much of a game changer HD video would be. The A900 is a very good camera.

The A900 offers nothing compelling for most photographers other than too much resolution. The D700 offers better high ISO performance than anyone else's FF camera. The 5kmk2 has better video than anyone else. (Even the highest Nikon with video still only shoots 24fps with 11khz audio.) Those features make people buy FF. The Sony is FF without any good reason to buy FF. Hence, people didn't buy it.

I have been using FF and 4/3 for the past few years and to me APS-C is a bad compromise. Everything that APS-C does better than FF, my 4/3 does even better. Everything APS-C does better than E-3, my 5D does even better.

But APS is in the "sweet spot" for the majority of photographers, who don't want to carry two bodies, two sets of lenses, and two sets of credit card bills.

2. SUV are not "better" than sedans; they're just bigger And, since I live in Romania - Dacia is selling quite well, not because it's "better"; it isn't. But it's cheaper.

Totally off-topic, but I spent 9 days in Romania back in 2002 and am a car enthusiast - I had a great time but like you said, I certainly agree that people weren't buying Dacias because they were better.

The 10-20x claims are clearly false, written with a hidden agenda and making false assumptions, like APS-C be produced on 300mm wafers while FF be produced on 200mm wafers.

Some simple math should figure this out. The surface area of a FF sensor is 864mm and the surface area of an APS sensor is 369mm (slight variation between different sensors, especially with Canon's smaller ones.) So, that's a minimum of twice as many APS sensors out of the same size wafer. 2x is still nothing to sneeze at.

The Kx is a great camera at a great value. If Pentax continues to produce cameras like the Kx they will continue to grow their user base and demand for step-up bodies will grow as well. I do not think the body colors played as much of a role as IQ.

I disagree. The K-x's IQ is no better than several other DSLRs, but its high ISO performance is better - but many buyers won't even know what that means. The colors got them a lot of attention in the press and a lot of word-of-mouth, and it just so happened that those colors sold a lot of cameras. I bought a red one for my wife after making an unexpected killing on eBay selling a Cosina 55m F1.2 for over $1k - and it was the color as much as anything else that made it a "fun" upgrade for her K100D. People love the colors. Heck, I'd be perfectly happy buying a brightly-colored higher-end DSLR; why not have a little personality in the camera?

I also spend a lot of time on a Disney Parks photography forum and the K-x has been a big success there for people moving up from DSLRs, and the big question that always gets asked when another person buys one is, "what color?" Photography is supposed to be fun, last time I checked!

Pentax would LOVE to compete with Nikon and Canon, but realizes it is not in a position to challenge them in those markets at this time. I don't think it is an issue of them not wanting to compete. Nobody goes into business to be the "4th" best.

Oh really? That's the mindset that can lead to mass meltdown. Having a relatively small but loyal customer base can be safer. Subaru is a classic example of this - they make good money selling very good products to a small but devoted fanbase. Almost everything they sell in the US is built on two AWD-only chassis's - no trucks, no luxury cars, no two-door sports cars, etc... yet they are one of the only companies to grow sales in a horrendous economy. On the other end of spectrum, look at VW a few years ago. They wanted to move upmarket and began selling a $40k Passat and an $80k Phaeton, while not paying attention to their bread-and-butter. This hurt them and it took a new boss to refocus them on where they should be, and they've improved since then. (I still would never buy one, but they're moving in the right direction!)

[Kunzite]

1. I have established this with actually some pretty decent logic. That being: digital era technology has, in every single case, become so cheap that today's top-of-the-line stuff will be in the bargain bin in only a couple of years. This in and of itself should make it obvious that we can expect sensors to develop to insane levels and become really, really cheap simultaneously. This is the only fact I need to build the rest of my assertions off of, and as far as I can tell you haven't addressed it.

2. I never claimed that SUV's are better. I claimed that they were more than the average person needed, yet they sold like hotcakes. As an answer to your contention that people won't buy more than they need. (FF in this case)

3. FF is the feature. It has been relegated to the top of the top-end for quite a while, but as soon as a camera manufacturer finds a profitable way to offer a good one to the masses, we can expect to see them take off.

By the time FF is threatening APS-C it will be too late. I'm arguing that it will happen eventually. I have yet to see any good evidence of why it wouldn't happen.

1. Yes, there is a logic - but does it apply in this specific case?
Obsolete technology is obsolete technology, be it "APS-C" or "FF". The fault in your logic is that you try to estimate the future using only a single example of an obsolete, long ago abandoned product. The trick is, no one is using obsolete products; but current ones. They used the 6MP sensor when it was current; the same for the 10MP one, the 12MP one and the ones that follows. Following the car analogy, being able to buy a 10-year old car for under 1000$ doesn't mean new car prices will go below that price.
I'd like to see proof that APS-C is becoming obsolete; but we all know there is no such thing. As long as we'll see new high-end APS-C products that sells well, we can't say anything like this.
You also can't give any time/price estimations. What is "really cheap"? When is "eventually"?
2. Around here, SUVs are selling because they're big. People are either feeling secure, or want to impress - in very few cases, they need the limited off-road capabilities. You are buying them out of pure habit, I'd guess... (and very cheap gas ) I'm not sure how many would buy a FF camera for the same reasons
Yet most cars are "normal". No one is talking about SUVs replacing "normal" cars, btw - quite the contrary, the trend is toward smaller, cheaper and more frugal cars.
3. And how do you market "FF"? Remember, for now one must pay quite few extra $$ for that "feature". How do you explain to the consumer that this "FF" label is worth some $$$ extra?

And I have yet to see any good evidence why it would happen.

[Winder]

The A900 offers nothing compelling for most photographers other than too much resolution. The D700 offers better high ISO performance than anyone else's FF camera. The 5kmk2 has better video than anyone else. (Even the highest Nikon with video still only shoots 24fps with 11khz audio.) Those features make people buy FF. The Sony is FF without any good reason to buy FF. Hence, people didn't buy it.


Oh really? That's the mindset that can lead to mass meltdown. Having a relatively small but loyal customer base can be safer. Subaru is a classic example of this - they make good money selling very good products to a small but devoted fanbase. Almost everything they sell in the US is built on two AWD-only chassis's - no trucks, no luxury cars, no two-door sports cars, etc... yet they are one of the only companies to grow sales in a horrendous economy. On the other end of spectrum, look at VW a few years ago. They wanted to move upmarket and began selling a $40k Passat and an $80k Phaeton, while not paying attention to their bread-and-butter. This hurt them and it took a new boss to refocus them on where they should be, and they've improved since then. (I still would never buy one, but they're moving in the right direction!)

So the A900 has too much resolution but the Canon 5DII and Nikon D3s don't? As a long time Canon 5D user I will upgrade to an A900 before the 5DII. I have shot with an A900 many times and it is a very capable camera. Do you really expect the A900 and the D700 to compare? You are comparing a 12MP to a 25MP. They have 2 different target markets.

Your right. Pentax would hate to be in Canon's position and have all that market share, R&D money, brand name recognition, their own sensor fab & development. Pentax prefers to scrap by year after year wondering if they are going to be sold. Pentax is going in the right direction and hopefully we will see them continue to grow and gain market share. The Pentax 645D is an obvious attempt to attract pro users and with some new glass and improved support you will see D3x and 1Ds users move up as well as some current MF users switch over. Pentax found a soft spot in the professional market and they will take advantage of it.......... not because they dream of being a niche player, but because they can make a lot of money and take a lot of market share. Pentax can also boost the markets perception of the Pentax brand name by being associated with the higher end market.

[nosnoop]

Second, believe whatever you want about the cost of a FF sensor, but already the sensor has gone from being the single most expensive component in a camera to just another part to plug in. Honestly, look at the state of the SLR market even 5 years ago! We've got cameras now that put that generation to shame, for much cheaper. What makes you think that the future will be different? Today's pro model will be in tomorrow's bargain bin. All I'm expecting is for the future to be like the past, while you for some reason are expecting that APS-C sales will continue as they always have been and ignore the changing economics of the sensor business.
Where's the flawed logic?

The flawed logic is that none of your digital obsolescence examples support your claim that FF would take over APS-C; and you ignore the evolution of the camera as a whole with your narrow obsession on sensor size.

In 5 years time, the market may look very different from today. Mirrorless could become the dominant segment. And there are so many features that can capture the consumers' imagination or for the next must-have capability. DSLR would probably be relegated to higher end niche anyways.

[falconeye]

(In 20-50 years, lenses will be obsolete too. But that's another thread. )

Another thread? Agree with the rest of what you say but not everything follows Moore's "law". There is a reason why some century old lenses are still regarded "good" by today's standards.

[bertbert]

Another thread? Agree with the rest of what you say but not everything follows Moore's "law". There is a reason why some century old lenses are still regarded "good" by today's standards.

And by all means, if Moore's law applies to sensors in the way I think it does, the demands on lenses will be much higher than today, not lower.

[mattdm]

Another thread? Agree with the rest of what you say but not everything follows Moore's "law". There is a reason why some century old lenses are still regarded "good" by today's standards.

It's the same thing as with viewfinders. Moore's law can't improve the optics, so therefore things which it can improve will eventually be superior.

A lens is effectively a sort of analog computer. It takes as input the light hitting the front element, applies a series of transforms, and projects the output to be recorded by film or a digital sensor. Analog computers have some great properties, including instantaneous results. However, a sufficiently powerful digital computer can approximate any analog computer. The only question is getting the right inputs and applying that computing power in the right way.

I think in half a century, the majority of photography will computational photography; the image recorded by a system of hundreds or thousands of pinhole lenses, processed after-the-fact to give the effect of a traditional lens. Focus, focal length, and aperture will all be chosen in post-processing. Given enough computing power and storage, you just record everything and then, later, you choose a moment in the stream to be the shutter click and a window to be the exposure time.

It's hard to predict the timeframe for this. Twenty years may be too soon, but there is serious and promising research already. The computing power and algorithms will definitely be there, but there's materials-science too — if one could make it so the sensor directly records the direction individual photons come from as well as their energy, even the micro-pinhole-lenses will be obsolete.

Of course, people will still be shooting with then-two-century-old lenses, although probably mostly on film. Or silver plate. But that will be a niche market for a particular hobby. A few people may use lenses on the front of the new digital cameras, like people use Lensbaby single-glass optics on their full-frame dSLRs today.

And by all means, if Moore's law applies to sensors in the way I think it does, the demands on lenses will be much higher than today, not lower.

Yes, certainly. That's one reason they're gonna have to go.

[falconeye]

A lens is effectively a sort of analog computer. It takes as input the light hitting the front element, applies a series of transforms, and projects the output
[...]
I think in half a century, the majority of photography will computational photography; the image recorded by a system of hundreds or thousands of pinhole lenses, processed [...]

Sorry Matt,

but this is entirely wrong. I guess you skipped the nastier parts of your quantum mechanics courses and therefore, you are excused However, I ask you to revert your thinking by 180°.

A lens is not a sort of analog computer. Because there is no input data to work with. If you would try to obtain input data (by measuring the wavefront hitting the front lens) you would destroy any chance to obtain a result. And if you tried to insist, let me refresh your memory and say that you simply cannot measure both, location and phase, of a photon. Heisenberg and all the rest. As long as we talk about photons, lenses will remain large and their count remains small (1-12). Yes, a lens does a number of transforms to the light waves. But these waves exist in the complex-valued space of Schrödinger functions, not any kind of real-valued space a computer exists in. The only exception would be a quantum computer though its possibility to exist remains to be shown.

But never mind. You're not alone. People advocating light-field cameras have the same problem to properly understand optics (well, most -- some do and actually tell you that a light-field camera cannot compete with an in-focus image from a classic camera).

BTW, another one who missed his quantum mechanics course was God. He tried your idea first (many pinhole lenses hooked up to a neural network computer, aka an insect's compound eye). But after hundreds of millions of years in frustration (as far as we know, his beard turned white because of this) about the bad image quality, he gave up and eventually, gave green light to the development of the lens (aka lens-bearing eye or normal eye). [Smiley intentionally left blank]