PMA 2006 produced less than overwhelming news. What does that mean?
The Photo Marketing Association (PMA) trade show is one of the primary places to evaluate where the photography industry stands (Photokina is the other). At past PMAs we've had lots of action in digital camera announcements and new product introductions, but the 2006 PMA seemed almost dead by comparison. It was much more conspicuous by what wasn't announced:
- Fujifilm broke its schedule (the S1, S2, and S3 were announced on a regular schedule at previous PMAs) and didn't announce an S4.
- Mamiya didn't show up with the ZD (though it popped up in a few places in the world a week or so later).
- Olympus and Nikon jumped way ahead of PMA this year with their E330 and D200, respectively. In Olympus' case, it may have been to allow Panasonic a clearer intro path for the LC1. For Nikon, it may have been to separate from the Canon 30D announcement, though, quite frankly, the D200 would have overshadowed the 30D in a side-by-side fight at the show.
- No new significant, push-the-boundaries, all-in-one bodies appeared (I'm talking of bodies like the Coolpix 8800).
- Pentax showed prototypes under glass, but gave no meaningful specs or timeframe.
- Sigma didn't replace the SD10, which is getting long-in-the-tooth feature wise.
- Sony didn't launch their DSLRs, Samsung only brought what they already announced, and Panasonic only announced theirs.
- New lenses were few and far between, except for Sigma's decision to move some of their lens designs to the 4/3 mount.
So, is the digital photography industry taking a break from the constant new designs, has it hit a ceiling, or was this just an idiosyncratic year? Let's take a look by looking at some clear trends I see.
It's not Megapixels
I think the die is cast. As Michael Reichmann has written about the 2006 PMA, it does indeed seem that camera makers have gotten the correct message: more megapixels aren't what we what. We want better pixels. Less noise, fewer artifacts, more dynamic range, plus more consistent and predictable color with better in-camera tools for achieving these.
Here's the reason why: digital cameras have now hit their stride for each market segment. I count four distinct markets, and they can be defined by megapixels:
5-8mp (group A). Compact digital cameras that are generally pocketable and budget DSLRs. These are snapshot cameras. The images will be viewed on screens and printed at 4x6 for the most part, with an occasional blowup to a larger size. These cameras are used in hostile lighting environments, which is their primary drawback at the moment. By that I mean few are capable of noise-free high ISO shooting and most have autofocus limitations in low light. What this class of user doesn't want is the same or worse pixel action at 10mp. They want higher ISO capabilities, VR, and faster focusing. Special note: this group breaks and loses cameras and ends up buying a new one every few years. Loyalty at replacement is low--they'll jump to another maker who gives them what they want. You could have 50% market share in this group and someone could come along solving all the point-and-shoot problems and knock you right off the pedestal. This does not bode well for current market leaders.
8-12mp (group B). DSLRs aimed at the advanced amateur. These aspire to be more than snapshot cameras, as the user typically aspires to fine art photography. That means that they actually post process and print their images, often as big as an Epson 2400 will allow them to go (13x19"). They need enough pixels to retain detail at that size, but more pixels than that generally only give them cropping ability; the smart ones will get lenses or use their feet to obtain the right cropping, if only the camera would do exactly what they wish of it. This group is probably the most satisfied (at least if they're a Nikon or Canon owner), as the current crop of cameras in this range don't have any major liabilities. Expect others to come play in this realm soon (Sony, Pentax, and Samsung). Special note: this group really only wants to buy once and concentrate on shooting, not keeping up with technology. Once they dial in with something that works for them, they stop buying. On the other hand, this is also the group that's most likely to switch back and forth trying to find that special product that's everything they need in the first place. However, once they're happy with a product, when they eventually get around to replacement, they will tend to be loyal because of their investment in lenses and accessories. So while they may have gone D30->D100->20D->D200 in the past, if they've got a Canon 30D or 5D today and are happy with it, they'll stick with the Canon brand in the future; likewise if they've got a Nikon D200 or D2x today and are happy with it, they'll stick with the Nikon brand in the future. This does not bode well for new entrants.
12-16mp (group C). The pro DSLR user that used to shoot 35mm film. These folk have specific uses, so this is where cameras need to be thought out a little more. We see a bit of that in the 1DII and D2x, which cater to someone who needs fast autofocus and high frame advance to capture action. But we need more of that dedication to usage. Wedding photographers have slightly different needs than sports shooters. Landscape photographers need different things than photo journalists. And every maker in this realm needs to think seriously about bulk and weight versus features and performance plus make sure they have the lenses and accessories that pros demand. Personally, I want two cameras that have the same image quality: one is smaller and lighter and gives up features and perhaps a bit of speed for that size advantage; the other is packed to the gills with features and can be as big as the 1DsII or D2x. But there are other ways to carve this space. All of us shooting in this group complain about minute aspects of image quality, particularly artifacts, antialiasing, and the extremes of the sensor's dynamic range. The top end, 16mp, is enough resolution for us, as it matches well the lenses we're using, produces a very nice double-truck image for magazines, and is just enough resolution to keep the stock agencies happy. Like group A, this group wants higher ISO capabilities and faster focusing, but our demands are much more specific and higher level. This group is highly loyal at one level, not-at-all level on another. By that I mean that as long as their brand provides basically what they need and doesn't get carried away with new UI, they stick to it, as they have no interest in spending time learning new tools. On the other hand, because their livelihood depends upon their tools, if that brand doesn't give them what they need, they'll switch in a second. Currently only Nikon and Canon serve this market and there's no hint of anyone else coming to play. Canon has high loyalty because they've provided what the user needs; Nikon has low loyalty because they haven't. Thus, until Nikon manages to up their game for this group, Nikon will continue to see its pro market share whittled away.
24mp+ (group D). The pro digital user that used to shoot medium format (MF). Note that this group starts higher than the last group in pixels. I'm tempted to put the bar at 29mp, but I'll be nice today and leave it a little lower. This group is shooting for resolution and detail. Thus, a bump from 16mp to 18mp isn't enough to distinguish themselves from group C. Because of the nature of rectangular areas, it takes a substantive boost in megapixels to get a clearly visible change in capture resolution. We just don't see that until we get into the mid-twenties. This group is a tough one to win. On the one hand, they're loyal, but that's only to a body (box) and lens system. Because most MF bodies allow interchangeable backs, their loyalty on the digital back side is far less. This poses an enormous problem for the companies involved. If you make MF cameras (Mamiya, for example), this small group already has your body and lenses, so your ability to push new product is low unless you get into the digital back game (ala Imacon and Hasselblad getting together). If you're a digital back maker, the interchangeability is both a good and bad thing. It's good in that you can push product to existing body owners. It's bad in that the smart body makers are realizing they need to be making backs and designing away from simple interchangeable backs (witness Mamiya with the ZD totally going away from interchangeable backs and Hasselblad changing the body/back integration in ways that favor them).
So with that in mind, how are things after perusing this year's PMA?
Group A: A lot of action here, with some focus on the higher ISO, VR, and faster focusing and no attempts to push out the megapixel boundaries. That's the good news. The bad news is that we've got WiFi and other trendy technology things that aren't really useful showing up. (WiFi may very well be useful, but not in the camera for shooting. HP has the right idea with a WiFi enabled camera dock--you want WiFi as a transfer mechanism, not as a shooting need.) Most makers have too many cameras with too little differentiation. They're playing a typical consumer retailing game: isolate features so that the store has plenty of up sell opportunities. "Oh, you want VR, then you'll need to move up from the X2 to the X3. Oh, you also want 6mp instead of 5, then you'll need to move from the X3 to the X4." And so on. What I find mostly missing from this category are truly distinguished cameras. For instance, I'm curious why Nikon hasn't produced the Coolpix equivalent of the 28Ti. Prediction: the company that figures out how to make the Stylus XA in digital form will have a big winner. In the DSLR portion of this group we now have a crowd: Canon, Nikon, Olympus, Sigma, Pentax, Samsung, Sony, and at the "deluxe end" the new Panasonic. I still think this is a war that will whittle down to only three or four survivors and will be driven more by price than features, though features will be used as a tool towards keeping the price floor from dropping out from under everyone.
Group B: Nikon solidified themselves in this category with the D200, Canon less so with the 30D. Still missing in action are Fujifilm, Pentax, Sony, and whoever else decides this is a spot they want to be. Olympus is likely to have a difficult time in this category, as I suspect that the E1 replacement will be more pro-targeted but due to 4/3 more likely to have a pixel allotment in this category. I would have said that maybe the E330 or E500 might stretch into this category until I saw the new Panasonic version of the E330 (the LC1, which will also be marketed with a Leica designed Vario-Elmarit 14-50 mm F2.8-F3.5 ASPH that has vibration reduction). Surprisingly, Panasonic seems to grasp what's needed in this category; certainly more so than Olympus. Remember, this group really wants to buy one camera and then use it for a very long time. In the Nikon film world, the equivalent camera was the F100. Once a group B user bought a F100, they stopped buying cameras (though they continued to buy accessories for it). Thus, Canon and Nikon are well set in this category: if both cameras continue to sell in large quantities, it simply reduces the market for the others.
Group C: Surprisingly, this is the area where nothing is happening, but the one where the most bragging rights are available. Canon owns this market right now, with Nikon having made some inroads with the D2x. I expect Canon to shore up the 1DsII beachhead later this year, but I don't expect Nikon to make any strides here until late 2007 at the earliest. From a Nikon pro standpoint, any new camera that pushes beyond where the D2x currently is would be snapped up in a heartbeat, pretty much regardless of price. I know I'd jump in a second, and every other Nikon pro I talk to would, too. Meanwhile, there doesn't appear to be anyone sneaking up on the two from behind: the 10mp Pentax targeted for Photokina announcement is group B, as is the 10mp Sony likely to be announced at the same time. As noted earlier, those could be too late to the game (and it's the wrong game for this category).
Group D: PMA had nothing much to offer here (let's hope Photokina rekindles some life in this group). Pentax had an 18mp 645 prototype under glass, but that's not going to appeal to anyone other than those that already have a film 645 and want to use those same lenses. If you're buying from scratch, the Canon 1DsII is a fine choice in that megapixel range and available now and for probably less than the Pentax will sell for. Without a megapixel infusion, the Pentax 645D will be dead in the water. The Mamiya ZD seems conspicuously absent from both PMA and the marketplace (supposedly some have been sold in Japan; we now have sightings in the UK and a few other places, but not here in the US). We've lost Bronica, Contax, and a few others, probably forever. Kodak punted. No one seems to know where Fujifilm wandered off to. Meanwhile 29+ megapixel Phase One backs are in backorder and slowly shipping out of the factory and Hasselblad continues to sell at modest levels. This game seems over. If you've already got a MF body, you're buying a Phase One back. If you're just entering and buying new, your only ubiquitous choice, and the only choice that seems financially healthy and shipping new innovative bodies is Hasselblad.
Future Action is in Software
Apple Aperture and Adobe Lightroom continue to push the boundaries in the software realm. Nikon Capture NX gets a nod in that respect, too. All three were up front and center at PMA. Finally we're seeing more photography-centric software rather than technology bits and pieces that solve a specific problem or plug into Photoshop. This trend will continue, but of the 18 projects that I talked about in a speech about the future of photography software last year, less than half have any equivalent currently being talked about. Plus Aperture and Lightroom are very unfinished and incomplete projects, in my humble opinion.
I talk about "hubs" in photography-related software. To be fully realized, a digital photographer needs three hubs: an acquisition hub, a manipulation hub, and an output hub. At present, many photographers use something like Nikon View or Photo Mechanic as their acquisition hub, but none of these quasi-catalogers come close to the feature set a photographer really needs, and none seem to even have any idea of how photographers work in the field (example: while Photo Mechanic can suck all the files off my Epson P4000, it doesn't respect card boundaries--cards sucked into the P4000 on different dates, and probably different locales, are all sucked into PM the same way). Photoshop has long been our manipulation hub, but as Aperture and Lightroom clearly demonstrate, Photoshop forced us to learn how to work like graphic artists, not photographers. And I really don't think anyone's given much thought at all to the output hub side of things. Heaven help you if you want to integrate an alternative RIP into your workflow, and now that everything is digital why the heck can't all my submissions and jobs be handled directly from my computer in my workflow?
Group A: Nikon has some momentum again, though I really think the WiFi thing is misplaced energy. Canon still has it, but only because they iterate the same designs so rapidly it looks like real change. Fujifilm has it for a couple of cameras. Kodak and HP also have a little bit of it. Sony, Olympus, Casio, and Pentax are losing it.
Group B: Nikon has it because of the D200. Canon had it with the 20D and probably still does have it with the 30D. Olympus reached for this realm and missed. Panasonic might have reached for it and grabbed the bottom rung with the LC1 (but it needs to ship in a near time frame to complete the grab). The Sony R1 is an interesting reach into this category, and has plenty of momentum since it stands alone in all-in-one design. No one else other than perhaps the Leica back, which is a specialty and low-quantity item, is playing the game yet.
Group C: It's a two-player game. Canon is living off 1DsII momentum. Nikon has a little bit of impact with the D2x, but it's tenuous due to its being at the bottom end of the megapixel range necessary to compete.
Group D: Phase One has momentum for the time being, as they're the megapixel king apparent. Hasselblad is in neutral, neither gaining nor losing momentum. Everyone else is fast losing it.
Software: Adobe has it. Apple looks like it has it, though they really need to move iPhoto and Aperture to cross platform if they're serious. Nikon has a bit of it through the new relationship with Nik, which produced Capture NX. Bibble has it, managing to push boundaries on raw conversion the others haven't all gotten figured out yet. But from there things fall rapidly. Lots of me-too programs and plug-ins, and one-trick ponies. A photographer is as interested in buying 50 one-trick ponies to shore up his or her digital darkroom as they are in buying 50 lenses to cover their primary focal lengths. Consolidation and bundles are going to start appearing soon. Unless you're ahead of the curve, as Eric seems to be with Bibble, that's a sign of weakness, not strength.
Wither the All-in-One
The all-in-one market will wither down to two or three viable products. The Sony R1 looks like one. Perhaps we'll see one or two other distinguished cameras show up eventually. But note that the R1 sells for more than most consumer DSLR kits. That's why I used the word "distinguished." An all-in-one is somewhere between that snapshot and serious amateur camera, so it either needs to be smaller than most current iterations, or it needs to have distinctive, desirable features that aren't me-too. The R1 wins in this respect with a better sensor and a fantastic lens choice. Indeed, so much so that it manages to break into group B.
All-in-ones do have some advantages. For someone who rarely changes lenses, they have dust advantages (if designed properly). Since they use electronic viewfinders (EVF), they can be used at odd angles without stooping or ladders. Some designs are lighter and smaller than DSLRs. There's usually nothing to lose or misplace other than a lens cap. They offer more control and manual override than consumer digital cameras (group A), which appeals to some.
In short, you're trading expandability for convenience. But you don't want to trade off image quality or performance. And there's the rub. Most all-in-ones to date have had image quality or performance liabilities, or worse still, both. To survive in the middle world between group A and B, they can't have either. Who's capable of designing such beasts? Well, Sony has been at it for awhile and finally seems to have honed it down pretty well. Canon and Nikon could if they set their hearts into it. Olympus probably could, too, but the E330 shows they've chosen a different direction, trying to combine some of the advantages of the all-in-one with a small consumer DSLR. Fujifilm and Kodak have tried, but IMHO missed.
Its going to be Pigment on Paper
Epson has owned the group B printing market (and group C and D proofing market) with its pigment printers. Introductions of pigment-based printers from HP and Canon show that the future for serious desktop printing is going to be pigment-based for the foreseeable future.