Friday, October 9, 2015

Light L16 Camera

There's this new camera announced. For sale in a few months, they say. It's 16 different cell phone camera modules banged into a small tablet format, and then computationally combined. It gives you loads of pixels, DoF adjustable in post, digital zoom that is allegedly not terrible, in-camera HDRish, etc. It's a bit like the high end Lytro. In some ways it resembles this thing Ctein suggested in 2011 but in other interesting ways it does not.

Update: To my surprise, Light sent me an email requesting that I add some backlinks to their web site. I see no reason not to, and indeed it seems like a gap in my original remarks. So, here's their corporate web site, and here's a link to the camera's page. I'll be sure to let you all know if they start sending me truckloads of money, or fly me anywhere on a press junket.

Allow me a brief vignette into the marketing of high tech products. The bible here is Crossing the Chasm and the extremely brief executive summary is that there are these people called innovators and early adopters who try new stuff out. You can sell stuff to them for a while. Then there's a gap in the system, and the next thing you have to do is start selling to the early majority, and there's a bunch of work you need to do.

The early adopters "get it", they buy the new thing for its newness. They're technology enthusiasts, and they understand the unique problems that the new thing solves. They will accept a remarkably sketchy and expensive product, because they "get it". It's usually not even a whole product, it's just a widget. The "whole product envelope" typically has huge gaps in it the early adopter has to fill in by himself.

Lytro, and now Light, are in the throes of the early adopter phase. What's interesting here is that the go-to guys on the internet are not early adopters. Thom Hogan and his compadres think of themselves as early adopters, because they embraced digital a decade ago. Now, however, they are serving an audience of enthusiasts and are themselves enthusiasts who are locked in to the DSLR and its followons. What they think of as innovation is a better DSLR, and "a better version of the thing I already have" is almost literally the opposite of innovation.

The most obvious tell is that the operative question always boils down to Ok, how can I pull these files in to Lightroom and start whacking on them with Adobe's toolchain?

This is to completely miss the point. These things capture before the Adobe toolchain becomes relevant. One can think of them as capturing light "in flight" and computationally forming the image in post. Adobe's tools are helpless here.

Furthermore, the innovation isn't in getting better stuff into Photoshop and then applying better Photoshop effects. That's buggy whip stuff. That's the old way. We don't yet know what the new way looks like, but we do know it's not the old way.

This begs the question, who the hell are the early adopters?

You need the early adopters to fund development, to fund the drive to the whole product and to fund the drive to enough volume to bring costs down. You need these people to buy the product before it's ready so you can figure out how the software is really supposed to work, what the UI ought to look like, and so on. They're integral.

They have money. They're interested in photography, in pictures perhaps. I'd like to say that they're not DSLR users, except that the previous two items pretty much guarantee that they are DSLR owners. Therefore, they're unsatisfied DSLR owners, frustrated DSLR owners. At a guess they hate all the fiddly options, all the "shoot RAW and develop in ACR and then import to Lightroom and and and OH MY GOD WHAT THE FUCK" garbage.

Lest we think of these people as losers, let me point out that they are precisely analogous to film photographers who neither develop nor print. They might, perhaps, be interested in shooting pictures, but not in the technical details necessary to produce the final product. There have been some pretty fair photographers in the past who fit this mold. In this modern era of digital, you almost can't work this way. There isn't any good way to have other people do that stuff. Nobody shoots and hands the SD cards to someone else. Do they?

I don't know if these people exist in sufficient quantity to fund this stage. There are at least two companies that have gotten some money to try it out, though.

How does this affect you?

If you're interested in the new technology, and you think you might like to be an early adopter, stop asking the questions "Does it shoot RAW? How do I import to Lightroom?" The new thing is new and a good part of the time the $1500 you spent will simply vanish after a while as that company's strategy does not prove out. It's all gonna be different, though, from shutter-press to final print.

Being an early adopter is expensive and frustrating, but it has its rewards.

Also, anything Thom Hogan or Lloyd Chambers says about the L16 is almost certain to completely miss the point.


  1. I looked at the sample images on the L16 site, and watched their videos. To me, the still images shown looked very good, very good indeed, but the on-line image files are small, and it's impossible to know how much post work it took to arrive at the final images. The HDR-like Golden Gate at night shot looked wonderful in terms of DR, but the near pilings looked like they had the dreaded watercolor effect, but it was impossible to know for sure. The shot of the boy with his father at the turntable had that "worked on for two days in PS" look of a high-end magazine editorial image. In a word, it looked FABULOUS! No doubt, the promotional materials the L16 people put up were slick, and the still images looked slick. I thought the camera lacked one necessary thing: lugs for some type of strap, to keep the camera secure, and carryable without need of a holster, or to be shoved into a pocket or purse. I thought the stills they showed looked pretty good. Hogan's concern about the company not being able to bring this thing to market, as evidenced by their need to beg for cash to fund development, is a serious issue though. He didn't mention it, but I think the early adopter market for a what is essentially a phone-sized camera, has been hurt tremendously by the development of successful, incredible new smartphones with already established track records--and this upcoming "camera" lacks the multi-function capability of something like an iPhone 6S or a Samsung S6...the ability to blur out the background of a pan-focus shot...ehhhh...that's pretty shaky ground on which to build. Overall, I like the look of the device, but again, it seems to be only a camera. I am also wondering how it will fare with the new DxO One and its much lower price impacting the early adopter camera segment.

    I get the feeling that the company pushing this new L16 is under-capitalized. It has never been easier to find venture capital for ill-advised ideas and products. Things like expensive modern brass-barreled Petzval lenses and $100 iPhone light meter attachments often flop terribly, even though the ideas sounded great to those who slapped up a site with a couple of videos.

    Of course, early adopters do not care much about anything except the cool factor that a brand-new, rare product conveys upon them. And so, the L16 might have its day in the sun. I think if this technology were introduced by a proven company with a lot of experience (Apple, Sony,Samsung,etc) it might stand a chance of going mainstream. If it were part of a smartphone, its chances of succeeding would probably be boosted twenty-fold. Sales of single-purpose things like compact-sized cameras are not very promising, and the idea of a $1699 to $1299 no-brand camera doesn't sound to me like it has a lot of marketplace appeal. The fact that the company doesn't even have the basic tech specs worked out or settled makes this look like the next "Silicon Film" solution; a great pitch that never makes it to the retail channel. I hope I am wrong though! The idea sounds appealing to me: gather a LOT of data, with the idea of being able to manipulate the image in software to get vastly different images from one original capture file. Having a wide-angle, normal, and a tele shot of each subject might be pretty cool. But again, the limited information available from the Light16 site is very difficult to draw any hard conclusions from.

    1. I certainly don't think either or both of Light or Lytro are sure things, by any means. Both will probably fail.

      What they demonstrate is that computational photography is real, and possibly the future. The discussion surrounding then also tends to illustrate how out of touch the go-to commentators on the internet are.

      You can have breathless, thoughtless, cooing about new tech, or equally thoughtless and slightly desperate dismissals.

      Nobody seems to try to actually think through the implications. Except, of course, me. But I am remarkable, as we all know.

  2. I'll take L16 seriously when you can actually buy one