Rounding out my recent discussions of where mirrorless cameras stand, we pivot from the past and present sales figures to the near future. Between discussions with engineers and executives, patent filing, and a host of other sources, it's possible to get a broad sense of the near-term future of mirrorless.
I'm going to go about this two ways: general, and mount by mount.
Broadly speaking, we're seeing technology start to become more important to mirrorless's future:
- Autofocus: both phase detect on the imaging sensor and higher frame rates on the sensor are right around the corner. Only Nikon seems to have mastered the phase-detect-on-sensor bit so far, but it won't take long now for others to figure it out. Higher frame rates, though, require more internal bandwidth in the camera itself. Even dealing with a sub-sampled 240 fps off the sensor will be a big step forward in data moving around inside the camera. Nikon's EXPEED3, which is about as big a bandwidth pipe as we've got so far, is being pressed to the limits with the V2. Enter EXPEED4, DIGIC whatever-the-number-is-now, and imaging ASICs from everyone that are giant fire hoses for data.
- Sensors: as we move to smaller process fabs we get additional benefits, as in either more transistors at the photosite or larger light gathering area at the photosite (or both). The former is starting to show up as higher dynamic range. In the Sony sensors, for instance, the column based ADC is relatively sophisticated and is lowering read noise, and additional transistors at the photosite's electron well can lower noise, as well. We're going to see more of this type of advance moving forward. And we still don't have a BSI sensor in a mirrorless product yet, so there's room for advance in the coming generation, especially for the small sensor mirrorless cameras such as the Pentax Q, Ricoh GXR, and Nikon 1. (If m4/3 is to go beyond 24mp, for example, it, too would probably benefit from BSI.)
- EVFs: I found multiple high resolution, high response EVF prototypes at Photokina. Both things are steps forward. A 60 fps an EVF can have as much as a 200ms lag when everything is said and done, which puts it behind the shutter lag of even low end DSLRs. A 240 fps EVF gets you down close to the blackout realm of current DSLRs if done right. One million dot EVFs are basically about VGA status, which is actually a pretty poor resolution these days. 2.4m dot and higher EVFs get us closer to the level we'd really want, with at least XGA levels of quality.
- WiFi: as I predicted, communication is something that every camera will need to do in the future. Samsung and Sony are already there with their latest models, but everyone will soon follow. It's low hanging technology fruit. Plus, done right, it's something a consumer actually understands and wants.
So it's not difficult to predict a lot of phase detect, 16mp+, reasonable high ISO capable, low-lag EVF cameras with WiFi starting to appear in the coming year to eighteen months. The Nikon V2 almost gets there today. I'll bet an OM-D EM-7 gets there as might a Fujifilm X-Pro2 and NEX-7X (or is it NEX-8?). As we've already seen, those things will trickle into the lower end models fairly quickly (witness the 16mp Sony sensor on the Olympus m4/3 models and the high resolution EVFs now available for the NEX-5 and on the NEX-6 and NEX-7).
Put another way, we're one generation away from near DSLR-capable mirrorless cameras. By DSLR-capable, I mean that in virtually no aspect of performance would the mirrorless camera lag the DSLR significantly, all else equal. Of course, that comes at the high end first, so expect to pay for it.
By mount:
- m4/3: Olympus seems to have focused most of their camera R&D squarely on m4/3 now, and it shows. They're iterating as fast, if not faster than the rest of the pack. Let's hope that they concentrate on changing the performance aspects and not the physical aspects. The OM-D design is basically good. What most users want is better focus, even better sensor, better EVF, and cleaned up menus, not more or different buttons and body shapes or even additional camera features. I'm a little worried that Olympus is spending a little too much time trying to save 4/3. That's partly because Olympus has lingering lens inventory and capacity they don't want to write off. That's a distraction from m4/3, and another 4/3 body takes resources away from another m4/3 body plus has the potential to have Olympus take the time to try to make the new 4/3 body better than the top m4/3. Wrong answer, because a 4/3 DSLR doesn't get you very far above a next generation m4/3 mirrorless (see discussion above). Best scenario: dump 4/3, redesign some of the 4/3 lenses for m4/3, and concentrate on m4/3. Okay scenario: dump 4/3 bodies, figure out how to make 4/3 lenses work on m4/3 bodies better, and concentrate on m4/3. Worse scenario: continue 4/3, invest R&D time and money on new 4/3 bodies and lenses.
Panasonic has already publicly said that they're likely to go more upscale with their future m4/3 products. I take that to mean more GH3 type products and GX1/G5 products with higher end performance and more features, with less emphasis on entry and mid-level products like the GF5. But frankly, I think they're seeing the wrong problem (and thus the wrong opportunity). To put it simply: Panasonic's distribution, sales, and marketing groups suck. It's not that the GF5 is a weak camera; it's a very competitive one that should be selling in far higher quantities than it is. One reason Panasonic is not making money off the low-end cameras is the way they've handled them in the marketplace. Going upscale actually is the wrong answer because with poor distribution, sales, and marketing they'll take on a smaller niche and still underperform. Worse still, Panasonic's president, Kazuhiro Tsuga has declared that any business failing to earn margins of at least 5% will likely be shut down. Currently, cameras don't earn 5% at Panasonic. In other words, the camera group is under both internal and external pressure at the moment, and they have a systemic problem with structure that's a critical issue now. - NEX: Sony is iterating like mad. To the point where they're starting to get discontinuities in their products. The NEX-6 is better in many things than the NEX-7, for example, and the NEX-7 is now out of whack with the application and WiFi future Sony is headed for. Sony is starting to have the same problem with NEX that Nikon has with their DX DSLRs: the iteration has gotten ahead of the inventory sell off. Just like Nikon will still be selling D3000 DSLRs in the US this Christmas, Sony dealers will still have significant stock of older NEX-3 and NEX-5 models to get rid of with new ones (and the NEX-6) coming in to muddy the waters. This is a problem Sony needs to fix, or else they'll get undermined by their own models being sold below cost.
As for how Sony is iterating: technology full speed ahead. You'll see more and better EVFs, GPS, communication, and programmability (unfortunately only by Sony) in the future models, and the sensor progression will continue (mostly in late 2013 and 2014) as well. Sony has basically said that they're going to throw all their expertise in electronics at the camera problem. Don't expect much in the way of retro from them—especially with NEX—but rather much more modern, progressive, technology-driven designs. That's not everyone's cup of tea, as it shows up in user interface as well as capability, but personally I think Sony's on the right track for them. In short, they're headed for making "the alternative camera that's different."
Unfortunately, like Panasonic, Sony has company-wide issues. Like Panasonic, Sony's debt rating is just above junk status at the moment and the company is straining to remain a consumer electronics giant. Analysts following Sony say that they don't expect Sony to show a profit outside its financial services group any time soon. Unlike Panasonic, Sony has made specific comments that they want to make the imaging market one of their core on-going product areas, so coupled with NEX's decent current position in the market, I don't think we'll see any major changes to their strategy in the near term. - NX: Samsung hasn't made any mirrorless offering stick so far, so they're still experimenting. It appears that they've decided to take that experiment into Android, so expect the first Android-based mirrorless camera to come from them, and sooner rather than later. I know some of my friends working at Samsung will wince at this, but I think that Samsung has two problems: (1) they haven't figured out who to copy yet; and (2) they haven't actually managed to get their marketing expertise to work on cameras yet. Android isn't the answer. But we'll get it, anyway. I also expect Samsung to be the slowest to get to some of the other solutions, in particular phase detect autofocus.
- CX: Here's a dirty little secret: CX (Nikon 1) won't be Nikon's only mirrorless system. They have no choice but to come out with an APS (DX) sized sensor in a mirrorless body. Why? Because the competitors they worry about—Canon, Samsung, and Sony—are all there. The CX system isn't designed to compete with those. It's designed to get natural Coolpix upsells, especially with women. Indeed, the CX system—other than the FT1 option—really seems like the original Panasonic/Olympus mirrorless play targeted towards compact users, only executed three years later.
To their credit, Nikon continues to try to push the ability of the CX system, though they're trying to do so with an undersized engine. CX is ripe for a BSI sensor created with a smaller process size, but it isn't there yet. I think we'll continue to see CX explore the area it's already established for itself, and Nikon will let some DX-based mirrorless system try to truly capture the performance-oriented user. The only problem is that this is still a far future thing and Nikon is already having a hard time trying to define why they have three mounts. So adding a fourth won't be easy for them, short of giving up one of the existing ones. That's led to a lot of speculation that DX DSLRs will be what Nikon gives up, but that's the meat of their interchangeable lens camera lineup, responsible for the majority of their sales and profits.
There's no question in my mind, though, that Nikon will have to match Canon's move into APS mirrorless. The question is how well they'll execute that. No doubt Nikon is watching the EOS M and how it affects Rebel sales very, very carefully. But you'll note a common theme in what I'm writing about the future: it's the marketing and sales problem that is the real issue with most of the camera companies' future. Nikon would be at five product lines of cameras with an APS (DX) mirrorless system: Coolpix, Nikon 1, New DX mirrorless, DX, and FX. They can't fully manage to get the marketing messages aligned for the existing four, so adding a fifth will be a real problem for them. Much more of a problem than the technical side. - EOS M: I'm a little surprised by the EOS M. First, that it exists so soon, and second that its technology doesn't move the bar at all. While it has phase detect autofocus, it sure doesn't seem like it ;~). And Canon's large sensors (APS and 35mm frame) are getting long in the tooth (still good, but not great and not moving the bar forward).
At this point it's near impossible to tell where Canon is going with the EOS M. At Photokina they were pretty tight-lipped about anything other than presenting the actual camera and two lenses. They didn't want to talk about futures, they were reticent to talk about how it fit into their lineup, they didn't want to discuss performance aspects. It was mostly "here it is" and not much more. Now perhaps someone in a motel room briefing got more than I did out of Canon, but if they did, they're not talking, either. Some of this is a cultural thing. Canon's been missing their financial goals, and in Japan that means side-shuffling of current management to nice window offices with no real responsibility while the system tries to reach consensus on what faction to promote to change things. If you know anything about how the way China's political system works, Japanese corporate management tends to be similar. And we're at the changing of the guard point at Canon, thus no one wants to talk to outsiders yet. EOS M could have a bright future at Canon, or it might not. We won't know until the white smoke has appeared at the chimney. - Fujifilm XF: Fujifilm's all in on this one. X is one of the few highlights of Fujifilm's continually contracting camera group, so X marks the spot for them. Phase detect is coming on the imaging sensor (dare I predict it on an X-Pro2?). Unlike most of the others, Fujifilm thinks that the X factor is the highly retro design, so they won't be fast to add other technologies into the mix, such as WiFi. The EVF will get more work, though. Basically, Fujifilm's on a "more of the same" approach, except for autofocus, where they've been maligned and thus know they need to fix. We're likely to see lower and higher end models from where we're at now, too. The higher end replaces the X-Pro1, the XE-1 holds the middle ground, and they'll try to figure out how to build a lower end model so that they can get into the pricing game with everyone else (that last is probably a mistake).
At Photokina the Fujifilm folks were amongst the ones most open about talking about both the deficiencies of their current product and the basic path towards future products. There were no secrets here: improve focus performance, make the EVF even better, extend the system with new models, and don't mess anything up while doing that. One other thing: a lot of the key engineers and executives I talked to at Fujifilm actually use their cameras and are aspiring photographers (doesn't seem to be true of Nikon and Sony, where it's mostly about technical engineering expertise). That all bodes very well for their future. - Leica M: Leica also is all in. Indeed, I saved Fujifilm and Leica for last in this list because I found both companies welcoming, open, and practical. They both aspire to build "best possible cameras." They both have some restrictions in doing so, as neither has a huge R&D group nor exhaustive resources to draw on. They both have targeted a higher end, more traditional approach to their mirrorless entries.
What impressed me most with Leica was the bet they're making on sensors. Rather than continue to use off-the-shelf sensors and deal with the limitations of those, they've made the correct decision that having something state-of-the-art that's in their control is a better choice. It's the power of "small group being efficient and focused" versus "large group with unlimited resources exploring everything." I didn't get a lot of time to spend with Leica at Photokina because of my being sick, but I hope to make up for that in the future. I liked what I heard. Leica has to be the most attuned to actual photographers of any of the mirrorless entrants. Whether they can fully deliver what we want is another story, but they're far closer today than they were with the original Leica M8 digital. Of course, Leica has made a choice: the M doesn't need phase detect autofocus or WiFi or a high-end EVF because none of those things are part of the base upon which they're building. Eventually, though, I think we'll see Leica move from manual, rangefinder focus to something else. But not yet. Most of their engineering cycles are at the sensor in the current generations, which isn't a bad spot to put it.
Footnote: Some may see my comments about the financial issues facing Panasonic and Sony (and actually quite a few other Japanese companies, at least in particular certain divisions, including imaging divisions), as "Thom is predicting the demise of Company X." Not exactly.
Much like the US Auto industry was in the 90's and 00's, the Japanese consumer electronics industry is in a slow motion collapse and has been for some time. Japanese companies don't tend to lay people off or do dramatic pruning of divisions as Western companies tend to. The Japanese companies also tend to have interlocking relationships—especially with the banking industry—that isolate them from quick and outright failure. In Silicon Valley we call companies that have businesses that can stay alive off of cash flow but not return anything on investment "walking dead." While it's an apt term, the "dead" part gets overblown by people reading that term.
We've seen this game before and the end is inevitable, however: you can restructure while you're still moving and jettison underperforming assets to save performing ones, you can fix the underperforming assets, or you can wait until some market dynamic pushes you over the edge (e.g. as the recent recession did for most of the US auto industry).
Here's the problem: for many of these companies jettisoning the underperforming assets—that would be compact cameras—would make the ongoing camera groups so small in some companies as to be invisible on their bottom line. Ricoh is already like that: cameras represent so little of the Ricoh business that the Pentax/Ricoh camera group is reported almost like a footnote. At some point, small groups, even profitable ones, aren't worth the upper management time and effort and you can't put enough money in them and get an ROI that will save the company. If you're going to stop making compacts, you might as well stop making all cameras at Fujifilm, Olympus, Panasonic, Pentax, and Samsung. But compacts are shrinking, and the smaller players are shrinking faster so there's a lot of pressure to perform.
In the case of Canon and Nikon (and to a lesser degree Sony), they can't give up any camera types lest they lose the ability to say they are one of the leading camera brands. In Nikon's case, cameras are their main business, to boot.
Unfortunately, what I see is that all of the Japanese camera companies are going to stay the course, much like most of the US auto industry stayed the course right into requiring government intervention. Thus, they'll continue to do what they've been doing right up to the point where an outside force stops them. That might the rise of a disruptive camera, it might be continued advances in camera phones, it might be a recession, it might be a quick devaluation of the dollar or yen, it might be any number of things. But a "big storm" will dislodge some of these camera operations. Multiple "big storms" could dislodge them all.
Short term—basically as far as can be predicted reliably—it's status quo, though. No one is going out of the camera business any time soon, all of the companies are picking and choosing strategies they think will win them more sales in mirrorless, and thus the overall news is okay for us users. Not great (except perhaps for the likelihood of price drops), but okay. We'll get another generation or two of mirrorless cameras from pretty much the entire set of players that continues to push mirrorless performance into new and higher ground.
