Layout Photography in the Digital Age

Between the availability of affordable, professional-quality digital cameras and some sophisticated new software, razor-sharp images with virtually infinite depth of field are now a reality. But why do some model photographers invest so much effort into getting the whole image in focus? Parts of photos of real subjects can be out of focus, right? Well, the operative phrase here is "can be." The vast majority of the time, photos taken outdoors in good light tend to be sharp overall. Consider that, usually, nothing is closer to the lens than about six to ten feet, where most cameras can hold the whole image in focus.

When photographing models, however, the depth of field (the range of sharp focus) is frustratingly slim, because now you're working within inches of a lens, which forces separated by only a foot or so to be acutely out of focus. The result is an image that is instantly recognizable as a model! Indeed, often photographers will deliberately throw parts of images of real scenes out of focus (using a technique called "tilt-shift") to make it look like a model.

The trick for more realistic model photography is to do whatever it takes to extend the depth of field as far as possible. Prior to the digital revolution, this meant pinholes and other mechanical tricks. Today, far better results can be obtained using your home computer using a technique called "stacking," whereby the sharpest areas of a series of images with different areas in focus are combined into one sharp image.

The following example provides a crude demonstration of the principle behind stacking. Here, instead of a normal stack of six to twenty images, I'm using just two. But it still shows just how powerful the process is: the locomotive in the foreground is only two inches from the lens, while the sky backdrop in the upper left corner is over twenty feet away, and in the final shot, everything is clear and perfectly seamless.

My introduction to stacking began with an article in N Scale magazine by Rich Yourstone, who had been tinkering with new (at the time) software called Helicon Focus, an image-processing program created principally for scientific research. A typical application would be microphotography (images taken through a microscope), where the depth of field is inherently paper-thin while the subject invariably is not. When used for model photography, however, "stunning" does not begin to describe the effect.

As it happens, the company I work for was using a competitive freeware product called CombineZ. This application is more powerful, but far less user-friendly (here's a detailed side-by-side comparison). After trying CombineZ, however, I elected to plunk down my $115 for Helicon, and I'm very glad that I did. Bear in mind that Helicon was created for scientific purposes, not general photography, and software made for highly-specific niche markets usually costs thousands of dollars. True! So, consider Helicon a bargain.

Sadly, some modelers can't help but complain about the cost and/or lack of features. "Can't we do the same thing with Photoshop?" Well, yes and no. In Photoshop, the process is incredibly labor-intensive. Also, note that manually altering the focus of an image results in very slight changes in image size, so each frame of a stack must be precisely re-sized before they can be combined, adding even more labor. Trust me, I know, because I've done it in Photoshop, and it can take hours to get right.

By contrast, my very first Helicon image was a stack of seven, and it took about nine clicks and 20 seconds to process. Heck, only $115? I'd have paid more for software that does this. And while Helicon "only" combines the images (it doesn't do color correction or other stuff modelers might like to do), the value of doing the combination automatically cannot be overstated. If you want to get creative with your images, you can pull the combined image into Photoshop (which costs many times more than Helicon!). Remember, once again, this is specialized software created for a specific task, not an all-purpose image processing suite.

Others have commented that using a good lens and stopping down to f/22 solves the problem in a single shot. I don't think these guys have done much model photography (or much pricing—a "good lens" will cost more than Helicon, and that doesn't include the cost of a good camera). Even at f/22, the depth of field is much less than a foot when you're working close to the subject. Sure, a helicopter shot of your whole layout will be good and clear, but just try capturing a close-up of your locomotive on a bridge, and see what happens to the river in the background: right, it gets blurry. That's why you need Helicon, no matter what camera or lens you may have.

For more details on the process and what you'll need to take Helicon photos, see Working with Helicon Focus.

Return to Chapter 8
Return to Clinics Index