The Camera We Want

Tue 05 June 2007 by psu

I like Thom Hogan. He is an intelligent writer, photographer and fellow technical geek and his web site is one of the better collections of information about Nikon cameras. As a bonus, the overall design and layout of the site shows a lot of taste and restraint, unlike almost every other such site in the universe.

This week, Thom posted a nice rant about the sad state of the compact digital camera. This is a subject that is close to my heart. I love small cameras but very few are well executed. In fact, one could argue that no one has yet built a nice small digital camera.

I don’t really need to repeat Thom’s main points here, but I will anyway. Compact digital cameras suffer from a proliferation of useless differentiation. The result is that they do not serve any of their potential markets all that well. People who want simplicity get a choice between hundreds of models all of which have complicated user interfaces supporting dozens of complex features. Photographers who want a useful tool get a choice between hundreds of models all of which have complicated user interfaces supporting dozens of features which are all useless photographically. I would suggest, for example, that no one who wants to buy a useful camera really wants any of the following features:

  1. Face detection.

  2. A zoom lens that is a 400/F11.2 on the long end.

  3. 18 bazillion point autofocus.

  4. Digital Zoom (e.g. automatic stupid cropping).

  5. Auto modes that generally do nothing but pop the flash when you are taking a night shot from the top of the Eiffel Tower. Not only do people not want this, they don’t want to have to decide between 50 models of cameras at various price points all of which appear to have this same list of features that they don’t want. This is especially true when the people you are talking about are enthusiastic photographers.

What we want from a camera is an effective tool that is predictable and easy to control. We want:

  1. Good focus and exposure controls.

  2. A good viewfinder.

  3. A good fast lens.

    And most of all:

  4. A good sensor with a high quality fast imaging pipeline.

Thom’s extensive feature list basically boils down to these four points. You put these four things into a small camera and you would basically have a digital version of some classic point and shoot cameras like the Olympus Stylus, Ricoh GR-1, or the king of them all: the Konica Hexar. All of these cameras had the ability to capture pictures every bit as well as a big Pro SLR if you knew how to use them.

Instead what we get are cameras like the dozens of Canon SDXYZ, the Panasonic/Leica point and shoots, 50 different ugly cameras from Fuji and 89 different cameras in various odd shapes from Sony, Casio, Samsung and all the rest. A few of these come tantalizingly close in terms of lens quality and controls. The Ricoh GR digitals are nice. The PanoLeicas are OK. Sadly, few of these cameras can generate a decent ISO 100 JPEG file and none have an image pipeline that is even 1/10th the speed of a D-Rebel shot to shot.

This, in the end, is the ultimate insult of the digital point and shoot. Even if you learn to work around the control and interface problems and you get a decent shot, in the end the sensor will let you down because the manufacturers have spent the last five or six years in an irrational penis waving contest to see who can get the largest number of noisy pixels onto a thumbnail sized sensor, rather than building a machine with a decent imaging pipeline in it.

I’ve whined about this before, but Thom’s article depressed me even more than usual, and led me to think that we who want decent sensor performance at ISO 400, or a fast shot to shot time are doomed to carry big cameras (even as small as it is, you can’t fit a D-Rebel in your pants). He made me think this because here he was laying out an intelligent design agenda for a compact camera any photographer would love to use, and any point and shooter could just point and shoot, but he had made one fatal mistake which all intelligent engineering types tend to make. He assumed that his interests were those of the mass market, and that building such a camera would bring people and money in droves.

I don’t think this is true. The market for an intelligently designed, relatively expensive, enthusiast camera is vanishingly small. Most people don’t want an easy to use photographic tool. They want a box with a single button that takes pictures. It doesn’t really matter to them that the latest Canon PowerSnoot SD5867 IS VR AFS also has 59 other buttons and takes video that can be sent directly to Youtube. They just put the camera into the “this button takes a picture” mode and go off and start hitting the button. When they hit the button at the right time, they get a JPEG file that can make a print at Costco that is approximately 15,000 times better than what they used to get from the film mini-lab. And thus, they are happy.

Where this leaves the poor photography enthusiast is with hundreds of models of compact cameras all of which have complicated user interfaces supporting dozens of features which are all useless photographically. The feature sets will not go away, because the camera companies think that it’s the lists that sell. The badly designed cameras will continue to sell, because most users will just work around the bad design by ignoring it and hitting the one button. What this means for the photo geeks is that there will never be a small camera that makes us happy because the industry does not know how to manufacture and market such a device. It requires that you make a camera that does “less”, but is better in a way that is difficult to explain. All such products are usually doomed, although there are a few exceptions (hello iPod).

All of this, in the end, leaves me melancholy. I would actually be willing to pony up a lot of cash for a high quality fast small camera. I’ll be the first in line if the heavens open up and deliver such a miracle to us. But I’m not holding my breath.

Category: Photo