Tuesday, September 13, 2016

Whither Nikon?

Kirk Tuck recently took a swing at this one so now, of course, I cannot resist.

First a little review of some market realities. In the olden dayes of yore, there were a lot of people who Just Wanted Pictures. They used Polaroids and Brownies, and they went to Sears for portraits and so on. A lot of them just shrugged and decided they didn't want pictures that bad after all. Then there were enthusiasts and professionals, who needed or at any rate wanted a "real camera" with knobs and dials and stuff you could adjust. Full control, etcetera and so on. Two quite different markets.

When digital photography came along, the major "real camera" makers jumped in early and were able to deliver some products that hit the right sort of nerve with the "I just want pictures" market. People of that stripe could buy a Rebel for a few hundred bucks and snap away endlessly and effortlessly. Nikon wasn't able to nail that Rebel vibe quite as well, with stodgy things like D3000 instead, but they followed along and made a ton of money as well. There was, briefly, a bubble experienced by the mainstream makers of "real cameras" with knobs and dials and stuff.

Then phones arrived, and that was that. The "I just want pictures" people now have phones and they're out. One could argue that a lot of those buyers are now somewhat more sophisticated, having held a Rebel. They have some slight notion of what differentiates "professional looking" pictures from phone camera pictures. But don't worry, the future is here. We have two different phones out there now that use two lenses to construct a 3D model. Right now they're doing a second-rate version of simulating fast lenses, but there's so much more. In a year or two we'll see "soften light" sliders as well as "more bokeh" sliders (yes yes I know what bokeh "means", but it's just Mike, you, and some guy in Iowa who are fighting that battle now).

So even the newly more sophisticated "just wants pictures" mom is going to remain satisfied with her phone, as will many enthusiasts.

So Nikon's market is contracting. I've been over all this before. The name of the game is to defend and grow market share in the shrinking (back to 1970s levels, or something like that, not to zero) market, and to look for opportunities to expand into new markets.

Nikon is a more or less storied name. Canon means printers and copiers and cameras and all kinds of shit. Nikon means Photography.

So, this is extremely obvious, but I have not seen it said. Nikon ought to be working on a product that is Very Very Much Nikon, but also Very Very Much the Future, For their core market, yes. Screw the consumers, they're gone. Nikon shouldn't care what moms who just want pix want. Making the perfect buggy whip for a market that wants cars is idiocy. Nikon also cannot follow, they cannot just pump out the Perfect MILC and seize command, that's just dumping another hybrid crossover semi-SUV thing into the already crowded hybrid crossover SUV thing market, and way too late.

Remember the Df ad campaign? Note perfect. Too bad the camera was a snore, eh? Retro, it turns out, is not Nikon. We sort of think of them as Retro, because we remember the old Nikkormats and the FE2 and so on, but those were not retro. They were entirely contemporary. Nikon has always been entirely contemporary.

Nikon basically has two choices. The first is to stay the course of endlessly refining the existing products, and hoping to manage supply chains and manufacturing costs and so on better than the other guys, and so survive to the new, future, smaller world. The second is a moonshot kind of thing. Something radical, something new, something Nikon, which is to say, something Photography. It can be a bit fiddly, the target market likes fiddly. They'll probably do the former, it's safer. But what if?

Grab that 100 megapickle Sony sensor, bang a light-field microlens array on the front of it, and deliver me a box that looks like a big MILC (cf. Hasselblad X1D) but yields 3D models that can be rendered out usefully big. With a 3D model backing up my picture, a huge amount of crap becomes trivial. Drop out backgrounds, erase objects, compositing, this is all trivial now. Relight the thing, at least in theory and within certain limits.

Don't give me the actual model unless I ask for it. Give me a 24 megabit flat image file with a set of perfect layer masks, for starters. It's just like a DSLR except you never miss focus, the lenses are as fast or slow as you want them (since you do all that in post), and you get perfect masks for every object. Roughly. I think the implementation would be slightly more complicated, but that's the sizzle you're selling.

I can practically guarantee you they're not doing this, since it doesn't protect the F-mount lenses, but that's the class of move we're looking for.

With that in place, a line of products built around this Nikon/Photography/Future system which, with luck, solidifies their position with the enthusiast, then they can start to look for oppostunities to actually grow. Note that they don't have to actually sell a camera to every enthusiast, they merely need to be on the radar of every enthusiast. As long as everyone's Uncle Bob sighs and says "man, that new Nikon.. I wish.. someday" then maybe, just maybe, you can make something of a Nikon branded phone camera, or a Nikon branded anything else that makes pictures.

Then you can start asking yourself "if every Thing is supposed to be on the internet, why shouldn't every Thing also take pictures?"

1 comment:

  1. Hmm, in my opinion the problem for Canon, Nikon and Sony is that while they manufacture extremely competent cameras and fantastic lenses, they fail to provide objects of desire. Especially in the low- to mid-range segment, cameras suffer from the following:

    - Nerdy and partially dysfuctional user interfaces. Autofocus and autoexposure have been introduced to make photography easier, but instead made it more complex - users need to know about AF-S, AF-C, single-point vs. focus point group AF etc.. Exposure: Instead of just matching a needle to a circle, they have to decide between P, A, S, M, iA and scene modes. In software engineering, we call this "leaky abstractions". On the other hand, there is only a single, modal command wheel for shutter speed and aperture, which brings me to

    - obvious attempts to cut costs. This leads to plastic bodies, lens mounts and lenses and tiny, dim pentamirror viewfinders. In conjunction with short product lifecycles, this has a bad impact on the perceived value of the camera.

    - and lastly, dated industrial design from the nineties.

    This was not a problem for many years as the new technology attracted technology nerds. Now that the technology isn't new any longer, these folk have moved somewhere else since they probably didn't care about photography in the first place. What remains as customers are in fact people who want to take pictures and with better quality than from a smartphone. Many of these are probably turned away by the issues outlined above. Why bother with an incomprehensible UI when all you want to do is take pictures, and are smart enough to deal with manual exposure and focus? Why pay a substantial sum for an ugly, cheaply-made lump of plastic which will cost a third in two years?

    I don't think that even more technical features like light-field would solve this problem. A simple yet functional user interface with less automation in a body which suggests "value" by its haptics is what I would try if I had to decide at Nikon. Something like a 70s SLR, but digital - oh, wait - the Fuji XT1 ...