I’m always curious about new camera technology. So I was intrigued when announcements came out last year about a camera where you didn’t have to worry about focusing when you took the shot…you could focus later. “Shoot now, focus later”.
I had to wait about half a year for the software to be released for Windows, but I was finally able to preorder in July, 2012 and the camera arrived on August 13th. I read through the information, checked the camera’s functions, took a few photos, posted some, and now have some observations.
How Light Field photography works, from the photographer’s perspective.
Light Field photography encourages you to produce an image with multiple subjects, or multiple points of interest on a single subject (other wise there is not much point in using Light Field). You take the picture and when you post it on the Lytro site, viewers can click on different points in the picture. The selected point will come into focus, and other areas will go out of focus. Imagine a person standing with one arm towards you, elbow bent. The viewer could focus on fingertips, wristwatch, elbow, shoulder, eyes, nose, ears, etc. The selected part would be in focus, and other parts would be out of focus.
Here are a couple of examples. The first is Scout. Click on her nose, or the background to change focus.
This is just a bush with a spider web. But you can click around for different focus points.
Without a doubt the worst ergonomics I have ever encountered in a camera.
In over 40 years as a serious photographer I’ve used cameras ranging from 4×5 press cameras down to Minox subminiatures. Form really does follow function, and with some cameras you experience how the design of the camera influences and enhances the picture taking experience. Hasselblads, the Olympus OM series, the Olympus E-1, Minox, Leica rangefinders, Arriflex cine cameras – to name just a few – use design to support the workflow and “thoughtflow” of the photographer. Not the Lytro.
I can tell that the designers struggled with a way to make the camera “handy”, but still manage to fit in the zoom lens and provide a functional display while still ending up with a pocketable camera. One of the challenges is that Light Field photography, despite the hype in the photo and computing media, isn’t about not ever having to worry about your shots being in focus. Good Light Field photography demands that you think about what you are trying to communicate in your photograph, and then take advantage of the Lytro “gimmick” so that the online viewer can select the the different points of focus in the photograph. And in practice that turns out to be a miniscule percentage of the photographs shot by real people in the real world.
To make the Lytro “sing”, great care must be taken in the composition of the photo. That means that the Light Field camera will need an LCD display with a wide viewing angle, is adjustable for different lighting conditions, and is large enough that you can be sure of your composition. The Lytro is none of those. The workaround for the limited viewing angle is to rotate the camera 90 degrees at a time to see if your eyes can get a better angle on the LCD. There is no workaround for the other shortcomings.
That camera rotation, in turn, brings out another set of problems: The dreadful controls. This is the point where the designers should be fired. The zoom control especially will change just through accidental handling, since its design seems to be more influenced by the iPod wipe screen control paradigm than by the needs of a camera user. Though probably more expensive to manufacture, this control should have been in the form of two buttons that require “pressure”, not “presence”.
The Lytro cameras in the photo below look cool, and Lytro states that “form follows function”‘ The problem is that the function apparently isn’t real world photography of any type. Want to find the zoom control? Look at the gray camera…On the upper face, the fourth row of little rectangular “nubs” from the bottom. You might be able to see some kind of shape there. Each of those little basic raised nubs is about 2mm x 4mm. On that fourth row are some even smaller raised things, each about 2mm long x 0.2mm wide, and probably less than 0.1mm high. Those identify the zoom “control” area. This works by pressure as you swipe you fingers sideways. The location of you finger bears no relationship to how far the zoom lens is extended. And you have to swipe, and swipe, and swipe to extend or retract the zoom fully.
This isn’t hip, or leading edge…This is just plain stupid.
Normally I’m not one to go on and on about flip-out displays on cameras. My favorite conventional camera is a Leica M9 – where the issue doesn’t even come up. But camera position is absolutely critical for Light Field composition and that means that the photographer is going to find him/herself in unusual positions relative to the camera. That speaks to the need for a display that can actually serve the photographer – who otherwise has to just guess and shoot. The solution is some kind of articulated or hinged display.
Shoot now, focus later?
This is really deceiving. It implies a more light-hearted, spontaneous shooting experience. In fact, the horrible ergonomics might be excused for a simple snapshot camera with a zoom lens — if the quality of snapshots wasn’t so poor. And I’m not just talking about pixel count here (which wouldn’t have been too bad 12-15 years ago). As a snapshot camera, the image quality is very poor. You’d have to smear a lot of something over the lens of your smart phone to get results as bad as the Lytro. Take a look at these two pictures — One shot with the Lytro and the other shot with my Blackberry.
The photo below was shot with a Blackberry Bold. It was downsampled to 27% of the original size to fit here. There was no other post processing.
This next photo was shot with the Lytro. The only post processing was to downsample the photo to 66.66% of the original size.
The reason for this comparison is to show that if you just want to take a quick picture, “Shoot now, focus later” is, frankly, B.S. The depth of field with a smart phone camera is so deep that focusing isn’t going to be an issue. And the quality of the Lytro image is so poor that you’re going to have a very bad looking picture from a $400 camera. For a quick snapshot, use your smart phone, or go to your local camera store and get the cheapest zoom lens digital camera you can find, along with the cheapest 4GB memory card you can find.
What to do?
Lytro needs to decide what it wants to do. The form and ergonomics of the first generation cameras points towards early adopters, snapshooters, and folks looking for a nifty $400 gadget. These users will be less likely to produce compelling images. Good Lytro images need careful composition, consideration of light, contrast, shape, and other factors. Because the Lytro camera needs to be close to the subject in most situations, even small movements of the camera position can have a large impact on the final image. To accomplish this, a serious tool (camera) is needed: 2 – 2.5″ hinged (not even fully articulated) display, fast fixed focal length lens (35mm field of view) — perhaps with a flip-down 1.4x converter, truly ergonomic controls, a few more controls for the display, exposure, etc. That’s not a sexy camera to market.
And all of this might be a moot point. How do you catch a person’s attention so they’ll be compelled to spend time mousing around an online picture looking at the different focus points? If you were presented with 100 Lytro photographs online, would you look at each of the first ten images? What about the next ten, and the ten following that. Would you even get to the 100th image?
Meanwhile, I’m sending my Lytro back.