Participation list

Andrea (AP)
author of rumors sites, such as www.43rumors.com

Mitch Aunger (MA)
author and owner of www.Planet5D.com

Thom Hogan (THH)
writer/photographer, www.bythom.com

Vitaliy Kiselev (VK)
UX designer, author of GH1 hack, https://www.personal-view.com/faqs/gh2-hack/ptool-software

It seems that almost no one that is testing new revolutionary and top secret cameras from Canon or Nikon will tell you any details.
Very little escaped early about even radical departures, like the Panasonic GH1 or the Sony NEX-5.
Why? Does this indicate:
(1) that the camera companies are really good at security,
(2) that things are happening in a frenzy at the last minute,
(3) that outside testing isn't really happening, or
(4) what?

VK: (1) plus (2) and very limited testing, so part of (3). Plus bosses do not trust their own marketing departments. :-)

AP: I remember to have leaked anything of the GH1 and NEX :)
Easy answer. I think Nikon and Canon are very good in keeping secrets. Too bad for Canon I already know the 5DmarkX has 145 Megapixel (let's keep it secret!).

What do you think about future of system cameras? It'll be mirrorless only landscape in low and medium level or we'll see DSLRs as primary force?

TH: The future of cameras can be summed up like this right now: looking for growth in all the wrong places. Meanwhile, Smartphones are generating more and more images and videos than compact cameras and getting them where the user wants them faster and more efficiently. It's almost as if we went back 20 years and the camera companies didn't understand what film was being used for. Oh wait, they didn't.

В

AP: I think we will see mirrorless growing and less entry-level DSLRs. In a very long term 90% of the market will be mirrorless. Full frame DSLR cameras will be the last to die.
I do agree with Thom when he says smartphones will be THE business. That's why even Olympus recently bought a phone company!

TH: The phone company Olympus bought is more like a place that sells phones, not a company that makes phones. As I'm already on record on this, I'll just repeat: Smartphone cameras will obsolete and cannibalize compact cameras, mirrorless cameras will eat the high end of compact cameras and the low-end of DSLRs (actually already have on the latter part). This leaves falling unit volume in both compact and DSLR sales on top of falling dollars-per-unit. Follow the money. That's what the camera companies will do (a little too slowly).

VK: I think that no all is so easy for mirrorless. Significant progress in EVF is still required. And entry DSLRs - they still have some space to cut prices.

TH: Ah, but cutting prices while also having declining unit volume is not a good thing. That equals declining profit. Better to just join the crowd, especially since the mirrorless cameras are enjoying higher margins than low-end DSLRs at the moment.

VK: As for smartphones. My view is opposite. All so called "smartphones market" will be seized by Chinese manufacturers, except for top segment (with 3-5% of sales). Their chip manufacturers already play catch game with Apple, and not without success. iTunes store won't save Apple. Whole empire will be disintegrated in next few years. And cameras in phones will remain good fixed AF with tiny sensors. Compacts will still be competitive in their niche.

TH: Wait. All smartphones are already made by Chinese manufacturers ;~). I really hate to sound like I'm racial profiling here or generalizing too much, but the problem has traditionally been that the tech future of the recent past has had a tendency to be invented by the West and produced by the East. The problem with compact and DSLR cameras is that we have the East inventing and producing. I'd argue that they're lagging at the imagination on the inventing side.

MA: The issues for me have always been responsiveness after pressing the shutter button. P&S and smartphone cameras have always had a delay when pressing the shutter button. DSLRs do not. To most pros and serious amateurs, that's a big deal. Maybe down the line if that gets fixed, then people will be ok with smaller cameras.

TH: That's really just an engineering problem. It's similar to the autofocus issue we discuss below. The "old" way was to just run the CCD in its slow frame grab mode to get data for focus and metering input, but that also meant that when you wanted to take a picture, you had to shut that down, clear the electron wells, and then put the sensor in a different mode to grab the still data. In other words, the "lag" was designed in. Every year we get faster frame rates and more clever methods of dealing with the sub-issues. So, to put it simply, lag will slowly go away as more engineering resources are thrown at the problem.

Nikon and Canon seem to be late to mirrorless banquet. Do you think that they'll present really revolutionary cameras?

TH: No.

AP: No.

VK: Probably. At least I expect few very interesting features.

TH: But isn't that the problem? Trickling out a few features at a time is getting less and less interesting. If you bought a D700 or 5DII two years ago, exactly what "interesting feature" is going to compel you to drop your current camera and buy a new one? The camera companies are into the auto industry mindset: "all we have to do is convince people they need to replace perfectly good products every X years." Meanwhile we have companies like Apple taking a more radical look at products and producing hit after hit and inventing (or at least invigorating) new markets.

It is now clear that economic situation in USA and Europe will be worse and worse with each month. How decreasing income will affect system camera market?

TH: I don't fully agree with the question's premise. Right now the economy is a teeter totter: it could go either way. Here in the US it would take a significant (and unexpected) event to dislodge the expected growth in 2011. We'll have 3-5% GDP growth and an increase in consumer spending during the year. The problem is the declining dollar value. So if you were expecting to sell X US$2000 widgets in 2010 the problem is that it's difficult to sell 1.1X US$2200 widgets in 2011. Not impossible, but difficult. The expectation is that tech products get more features and slowly go down in price. The problem is that DSLRs in particular are likely to go up in price with the same feature/performance improvement as before. Something has to give, and it will likely be unit sales. As I've written many times before, the market is ripe for true innovation. We need communicating, modular, programmable cameras. We'd pay more for them. We'd buy them to replace what we've got because they would give us something we currently don't have and aren't close to having (amongst other things: improved workflow). I'm reminded of the same problem in mp3 players, high-end cell phones, and tablets. The market was small, splintered, and not really growing usefully in all three until Apple came along and showed people what a modern, well-considered product looks like. Now everyone is rushing to copy those. The condemnation of the Japanese camera industry is that we have no Apple-like company leading the way. We instead have a dozen companies all looking at each other and mimicking one another, thinking that they'll discover some superpower that'll allow them to steal market share while just copying.

AP: What a question! If you ask me about the current economic situation the last thing that worries me is the digital camera market.
I really don't know. Maybe the economic crisis will lead people to buy less cameras and take less photos. That will make us better photographers :)
And I also do believe that the crisis will lead manufacturers to become more innovative! As you see I always tend to see the bright side of the story!

VK: If we'll look at Africa, we could see that prices on food jumped considerably and we could observe results of this price hike. Situation with many commodities also mirror this behavior. It is clear that we are now looking at inflation (and it'll rise exponentially) for really necessary goods and deflation for many high tech gadgets or luxury items. Deflation is also extremely complicated as extreme amount of money flow to top 5-10% of population instead of helping to recover situation for other 90%.
This is why I think that we'll see drop in the middle of price range. After about a year or two we'll see big consolidations amongst camera manufacturers. And various methods to cut all expenses.

TH: The Japanese don't tend to consolidate the way American companies do, though lately there's been more of that than previously. Ries and Trout would say that the correct approach is to simply create new niche markets where you can be #1. That may be what Pentax is thinking with their small sensor mirrorless entry. But I think it's too small a niche for them to get any traction from it.

Is it necessary to constantly update product lineup in low and middle end each year?

TH: It's where the money is. And "new" sells.

AP: For the mass YES. In reality NO!

VK: I think it is more of marketing books influence. Even bike companies each year introduce products with different frame paint colors and new rim design :-)

TH: Yeah, but that's the sign of a stale industry with no growth. You're relying upon fashion to sell the product, not functional changes. I have three bikes, all many years old, but I don't need a new one because mostly what's changed is fashion and I frankly don't care what color my downstay is.

System cameras are much better in this regard, but compacts lineups are complete mess.

TH: Compacts have gotten on the "once a year" wagon. That means that the complete life cycle for them is less than two years now. But that means that the 2011 models aren't all that much different from the 2010 models, as you don't have much time to change big things. And so the camera makers just keep putting themselves into a deeper hole. Nikon would be better off with a single Coolpix model that was programmable and communicating and done right than it is with 10 that just iterate the hell out of the same narrow list of items (lens specs, sensor specs, menu additions). You don't see Apple doing 10 random iterations of the iPhone. They don't seem to have any problem selling more of them than Nikon does Coolpix.

AP: I am with Thom!

We can see dramatic price drop on ultrazooms. Panasonic and Canon no longer rule this field.
How it can affect system cameras (as both companies no longer get such large profits)?

TH: Well, one thing the camera companies don't get is that in the modern era, just building iron (hardware) isn't the way to big profit. It's the connection that the iron makes to something else that's important (e.g. computers/phones/TVs/etc. connect to the Internet). Done right, you build an eco-system that's above and beyond the basic iron. iPod/iPhone/iPad/iTunes is a good example of an ecosytem. Apple won because they made the whole user experience correct, not just the hardware.

AP: Can I disagree with Thom? No I can't because he is right again :)

VK: I agree completely :-)

TH: I like being right, but it doesn't get me anything ;~) I'd rather be wrong and have the camera companies delivering the cameras I want...

Maybe patents play negative role in whole system cameras progress? Too many restrictions?

TH: Good question and on the mark. Camera companies trade patents amongst themselves, but this also keeps newcomers out as the existing players build a patent wall against newcomers. Of course, they don't seem to be patenting radical new ideas, just iterative ones, so the next big thing will disrupt that. Indeed, to some degree the cellphone cameras already have. The camera companies are going to have a hard time patenting software extensions because apps on the iPhone are already getting there.

AP: Not anymore. Everyone is patenting everything. If one starts the war everybody would lose. Could be the beginning of the third world war. In that case I will fight with Canon :)

TH: There are examples of patent pools that work (MPEG-LA, for example) and patent pools that are leveraged for money (Kodak's library these days). But the whole patent system is pretty much broken. It's going to take a lot of court decisions to make it untenable enough that change has to happen, though. Meanwhile, you're best off if you operate like Apple: patent what you can, especially the far forward stuff, build the product that customers will swarm over, then deal with the legal residue later. That so NOT describes the Japanese companies it's pitiful.

VK: From my POV all patent and copyright system needs to rebuild from scratch. Right now it mostly prevents proper economic development.

TH: You have Disney to thank for that on the Copyright end. Not sure who to blame on the patent side. The original Copyright and patent notions by our founding fathers seem more and more "correct" to me, even though if they were followed I would have lost rights to many of my IP by now. The purpose of IP laws is to give the creator of something some time to take advantage of it, not for them to perpetually lock it up. Locking it up forever is a lazy way to be competitive.

Feedback in large corporations seems to be broken. It is extremely hard to reach someone who has any influence on actual product design and implementation. Even fixing small and simple things can take months.

TH: That's the way they're designed. It's a very paternal system ("we know what's best for you"), very insular, and very isolated from its customer. Less than 20% of the sales for cameras occur in their home market. Most of the companies rely upon a Japanese trained head in their subsidiaries to relay customer preferences, but those same executives aren't measured on that, they're measured pretty much solely on brute force sales.

AP: Having studied philosophy that is the sort of question where I can try to appear smart :)
The very general answer is that the described issue partially depends from the Japanese culture and history. They are a vertically oriented society. The leaders do decide, not the slaves. To be honest many times leaders do know more than the slaves, so this isn't necessary a bad thing. But you have to work on both directions. From up to down and vice versa.

VK: Without repairing feedback system Japanese corporations will be slow and sometimes awkward. Repairing this information channel must be their top priority now. One who'll do it first will get big competitive advantage.

TH: The world is global now, especially for something like cameras, where 80% or more are sold outside the home market of the developers. Truly global companies act differently. Even the Japanese figured this out at one point in the auto industry. With the US consuming 35% of the DSLRs made, there's absolutely no reason why you wouldn't have a design center in the US and try to come up with variants of your product that cater to the local market.

MA: Agree that feedback is broken - especially since Canon was so shocked to see how the 5D2 impacted the world of moviemaking. Canon told me they were building an R&D center in USA to try to get closer to the customers - but they'll need to do much more than to move to a new location to fix the issues of listening!

May be actual problem is that US and Europe consume most of system cameras produced? Looking at our local market I see complete nightmare and havoc.

TH: A rough rule of thumb is that North America is 33% of the camera market, Europe (with all extensions) is 33% of the camera market, and Japan is 20%. That leaves 14% for the rest of Asia, Australia/New Zealand, Africa, South America, the Middle East, and let's not forget Greenland and Antarctica ;~). So you really have four places you need to pay close attention to: North America, Europe, Asia, and future markets that are still emerging. You can sell your old stuff to the latter, but the first three all have different tastes and desires, and certainly have different sales patterns and mechanisms.

Do you think that the Japanese companies are not global enough? Do they not see their customers clearly? Do their engineering departments operate in too much isolation?

VK: Yes, yes and yes. Engineering departments seems to be on other planet, may be even outside Solar system. This can be main reason of camera products that we have now. And most branches of Japanese corps are just marketing stuff.

Last breed of cameras seems to be made in a hurry:
GH2 with supply problem and known banding problem.
Pentax K5 with some dirt under AA filter, plus AF problem in some light conditions.
Nikon D7000 with stuck and hot pixels.

TH: All cameras are made in a hurry. The design cycle is short, the whole life cycle short. A product like the D7000 has a complete life cycle of maybe three years from conception to end-of-sale. Pro cameras are a little better off, as the big changes occur on about four year cycles and they actually spend that time investigating, developing, and improving technologies. But in the consumer realm where most of the money is the name of the game is iterate, iterate, iterate. Back when I did the Quickcam, from the initial decision to do a product until first customer ship was six months. Six months. At the time we made the decision, all we had was a prototype sensor generating a data stream. It's really a miracle that complex products get done at all in the time frames we're talking about, but it also means that there are a lot of things that don't get dealt with fast enough to avoid a problem in shipment. Moreover, the modern product methodology says it's better to fix a problem when revenue is coming in from it than delay the product. In case you think cameras are the only product in which that happens, most modern OS's and complex applications ship with more than a thousand known bugs in them these days.

AP: I am with Thom on that. I can see that in every aspect of our life. We are always running and running. That's why I love to live here in the middle of the mountains. We should start to slow down things. I know this is not going to happen without a general crisis.

VK: It seems that cameras become more and more complicated and in 3-4 year time frame we'll see cameras initially released as beta versions and getting better towards SP2 release.

Latest canon 600D seems to be another evolution step towards video, with articulating screen, implemented ETC like zoom. Do you expect Canon to continue moving by small steps?

TH: Yes. They're programmed to move that way ;~).

VK: I do not agree. I think that we'll see big things from Canon.

TH: As much as some think that Canon's been a pioneer in digital, I'd argue the opposite. It was Nikon's move into digital (E2 followed by D1) that pushed Canon to respond. They got their sensors right more quickly than Nikon, but they were late with lithium ion batteries and a whole host of other technologies we now take for granted. I see them as being behind. But that's okay for a market leader, you can afford to be slightly behind your competitors and let them steer you to the correct technologies to deploy.

MA: I may be disappointed, but I believe at least this time, Canon has understood what a monster they have with the 5D2 and the next release will be more than just an incremental step. There are rumblings of this... but yet, I'm torn by the monstrosity that is the modern corporation like Canon and Nikon etc... having worked in a huge corporation for 30 years, I learned the hard way time and time again how slowly they move.

Pentax 1/2.33" camera. Is this a joke?

TH: From what I've seen of the specs, I don't know what Pentax was thinking. One problem is that the Japanese companies tend to visit the spaces they've been to before rather than actually try to figure out the next "right" product. The Pentax entry fells a lot like the old System 10 (110 film camera) introduced in 1978. As I recall, it wasn't exactly a winner (110 film didn't stick around all that long, nor did it become the mainstay Kodak had hoped it would). But what Pentax essentially has created is a compact camera with two interchangeable lenses that are less capable than those of most modern compact cameras (47mm f/1.9 and 28-80mm f/2.8-4.5). Makes an LX5 with its 24-90mm f/2-3.3 look pretty darned good. Kenko, at least, picked the C-mount for their compact sensor interchangeable, which makes more sense to me. But overall, it's as if a whole country of engineers didn't get the memo: Facebook, Twitter, Blogs, Email, and more is the future of imaging. How does just downsizing a DSLR fit into that future? It doesn't. It ignores the future. Completely. So I call Dead End on Pentax.

AP: No joke. I have seen the camera! And honestly, don't know why Pentax is doing that. Maybe Hoya is so incredible smart that I do not understand that.
I have a suggestion for Pentax-Hoya. Next time contact the customers and ask them what we think about your ideas before you realize them!

TH: Sometimes just putting it on the market is more efficient. When it doesn't sell, you just move on ;~).

VK: According to pentaxforums.com opinions Pentax could just make some kind of toy camera. Not serious one. But looking like old rangefinder and with cool looking lenses. For customers who actually need compacts or ultazooms, but want to look cool. Something like K-x in crazy colors.

TH: As I've written before, the minute that style becomes the overwhelming change, the market is near dead in growth and subject to instant and nearly unpredictable whim. To think that the camera companies can outdo the Pradas and Guccis of the world is unfathomable. Indeed, there's a future prediction for you: you'll see style-oriented companies licensing core products from camera companies and adding the style to them.

Modular cameras. What do you think about them? May be future of system cameras lie in this field? And "body only" in future will require to buy lens mount module, screen, VF, IO module?

TH: As someone who's been a proponent of modular from the very beginning, I'd like to point out that we already have modular cameras: they're called DSLRs. The bodies are a module, the lenses are modules, and the flash units are modules. The problem isn't that we don't have modular cameras, the problem is that the points at which the modularity are made are no longer 100% valid. If you take the D100, D200, and D300 as an example, you see that you've got basically the same body design where three major components have been updated (sensor, digital support, and autofocus system) and a lot of finicky button/control iteration has gone on to support iterative feature creep. Yet internally, not a lot has changed in SLRs (and now DSLRs) since the 80's. Indeed, Nikon uses a modular construction internally, putting all the digital support on one board. That's why the D3s and D3x can be essentially the same camera but different. The sensor and digital board modules internally have been swapped, new firmware loaded, and voila, you've got two camera models. Modularity is being done for the convenience of manufacturing, not to create useful functions for photographers. As I've noted in my articles on the subject, if Nikon had done modularity right I'd still have bought two pro D3 base bodies, but I'd have actually bought four or five sensor/digital modules had Nikon chosen to produce them (color low light [D3s], color high rez [D3x], black and white high rez, infrared, etc.). In other words, they left money on the table by choosing to put their modularity where they did. Instead of getting US$13,000 out of me (D3s/D3x), they could have extracted US$20,000 or more. Bad decision.

AP: I know for certain that Olympus had working prototypes of such modular systems. Don't know why they didn't introduce it. In my opinion all Japanese companies tend to be very conservative. Such a major change will happen when some mysterious forces will force them to do that. Usually big innovations do come from small companies. So, come on, Olympus, do it!

VK: My understanding of modular is as follows: camera parts will be connected via two power lines and all else will be using high speed optical connections similar to Intel Lightspeed. You'll be able to keep screen, flash, VF, IO module, lenses and lens mount module while changing camera, even if it is from different manufacturer. Most interesting here is that IO module will be similar to current smartphones in a way that most software will be portable and only low level functions will be unloaded to LSI in main module.

TH: If you're going to do cellular communication, you have to do that modularly or externally. You can't be replacing US$5000 camera bodies because a provider is now doing 4G, 5G, 6G, or whatever change comes next. Moreover, though GSM is prevalent in much of the world, not all providers are GSM (and there are frequency variations, too), so to be able to travel internationally with a communicating camera, you have to either support everything or use modularity.

MA: When I first saw the RED ONE, I thought "modular is brilliant" - but the issue at this point in time (and of course that changes as time goes forward) is the size of those modules. The HDSLR body shape is small and compact and until modular units can be similar size, modular won't convert many. Tho as I say, 5 years down the line that may be different.

VK: I suggest you to look at modern cameras motherboards. They could be placed in very small modules. Very small.

MA: this is true - but the question comes down to how does the consumer replace those modules? At this point, there isn't that capability and I tend to believe most manufacturers don't believe the consumer can/needs to replace modules - especially when their trained to buy entire new units - which brings higher profit/sales.

VK: But if someone could show consumers that they gain savings by using this route it'll work. Not on entire market, but it'll work. People also like persistence, like to use similar menu, similar handling, etc. So, new manufacturers could play on this field.

TH: I think the size issue is a red herring. 25 years in Silicon Valley makes me wary of any "you can't do it" claims. You always can, it's just a matter of ingenuity and actually tackling the right problem. But in terms of "savings" I don't think that's the way to market things to the customers we're talking about. Modularity makes no sense at the compact camera market, but makes a lot of sense at the serious shooter and pro market. Those folks want flexibility and extension. They don't really want to be locked into buying everything new every two or three years just to get a particular feature. Those of us in the far tail, who experiment with B&W and infrared and more really don't want to have to own five cameras just to do the five things we want our film, excuse me, sensor, to do.

Why do you think that camera needs GSM module? May be wireless connection will be much better solution?

TH: Personally, I'm a fan of being technology agnostic. I don't care which technology wins, only that it's available to me in some form that doesn't require jumping through hoops. For example, we have Wi-Fi capability for cameras right now, but it tends to make us jump through hoops to configure it and use it well in a workflow. So I want Bluetooth, WiFi, GSM, LTE, USB 3.0, and every other form of communication possible. Again, if you do it modularly and open the spec, you can just let the user decide what the right technology is. If you allow programmability, again the user (or his surrogate, an independent developer) solves the problem for you. But you have to build "communications enabling" features into the camera.

VK: It's all cost a lot. May be 2-4 core ARM based integrated LSI with build in functions will be real center of each IO module? They can be really powerful.

TH: We've got plenty of evidence that the actual communications pieces (the chips controlling the communications) can be done inexpensively. I see no reason why we couldn't have US$300 cellular and Wi-Fi modules.