When Does Digital Match Film?

Film and digital can't be simply compared by using math. There's much more to the images than numbers.

 

Wings of Spring

Wings of Spring

Which picture is from film? Does it matter here? Not really. Both photos above work just fine on the printed page. The Minnesota forest scene (top) is from Kodachrome, while the Great Smokies stream (bottom) comes from a 7.1-megapixel Olympus EVOLT E-330. Both images make excellent large prints. You can’t compare film and digital simply by arbitrary use of numbers—the technologies are just too different.

About 15 years ago, digital imaging started to capture the attention of the press, including photography magazines. At that time, a lot of this was gee-whiz stuff, and most photographers saw it mainly as a curiosity or something that might work for scientists or other specialized use. Many pundits at the time made rash pronouncements of the technology, using all sorts of techniques to compare film and digital, but mainly they all came to the conclusion that it would be a very long time, if ever, before digital image capture could match film. And they all said that film would be around for a very long time. Obviously, that has turned out to be wrong.

But the idea of comparing film and digital capture still remains. The problem is that these two technologies are being compared by math rather than reality, so the comparisons often are dead wrong, which has caused problems for some pros.

The most common evaluation is done by comparing 35mm film with digital images. Some people still do this to “prove” the superiority of film over digital, even though no actual tests are done. They have used a lot of formulas to mathematically compare film and digital. Usually, the result is that for digital to equal 35mm film, something around 22 megapixels is needed, though I’ve seen comparisons that go as high as the 30-megapixel range.

Anyone in the publishing industry using a lot of digital images over the past few years quickly discovered this simply wasn’t true. No one did any math, but the results were obvious. You could enlarge digital images onto the printed page much more easily than working with film and a 6-megapixel camera would match 35mm except at the largest sizes. Of course, size was one element of the math formulas, so this only gave a hint of what digital might or might not do.

Personally, I’ve worked with several photographers who started printing large prints of 16×24- to 24×36-inches with stunning results. Even though I had been following the photo industry for quite some time, I was amazed and impressed by the quality in these prints that came from…6-megapixel cameras. This wasn’t supposed to be possible. Yet here at the OP offices, some of these photos were put up on the walls next to fine prints from 35mm film and, as a whole, they looked very close to the film prints.

Now cameras have gained better sensors, not simply with more megapixels, but with better noise characteristics, better tonal renditions, better color and more. These are now translating into amazing prints. I’ve seen 3×5-foot prints that are incredible from 10- to 12-megapixel cameras. These prints are from images shot with high attention to gaining image quality, such as shooting from a tripod, proper exposure, careful RAW processing and low ISO settings. They exhibit image quality above and beyond what we ever expected from 35mm film.

Recently, I scanned some old slides for a book project because I needed the subject matter in them. These were from Kodachrome and Velvia, and were shot using the best technique I could. I used to think some of those images were terrific shots from a quality standpoint. Yet, when I scanned them at high resolutions, I was disappointed. They didn’t match my digital work in sharpness or grain/noise.

So what’s going on here? How could the math used to compare film and digital be so wrong? There are a number of factors involved, not the least being noise/grain—pixels simply capture cleaner, better quality than sensitive bits of film grain.


But that doesn’t explain it all. I think partly that too many people are trying to make comparisons as if the technologies are equal. Consider this: Would you expect a car engine from the 1950s to match the horsepower of one of today’s engines, even though the displacement of both was the same? Actually, most people would be surprised and disappointed if the car industry hadn’t developed more than that. Or consider refrigerators, a simple piece of equipment, right? Take one from the 1960s and try to match its efficiency in cooling and energy use with even the cheapest refrigerator available today. You can’t do it.

The point is that technology changes. The technology used for film and the technology used for digital capture are two entirely different things developed in different ways. Plus, remember that although film has greatly evolved over the years, it’s still based on a technology that was developed many years ago, and there are limitations to that technology. Digital is a very recent development, on the other hand, and has very little history to its technology, so it can be based on very new ideas and engineering.

The race for megapixels has overshadowed some essential points. One of these is that it’s not just megapixels that are important, but how the individual pixels are constructed. This is one reason why a small, 10-megapixel point-and-shoot camera will never match a 10-megapixel D-SLR—the pixels aren’t created equal. (Though, truthfully, even small digital cameras can deliver amazing prints when the photography is done with care.)

Some people may remember that the Mars Rover camera originally had a camera with a 1-megapixel sensor. Admittedly, the big prints from Mars were essentially stitched panoramics, but still, the performance of the sensor was amazing. A lot of effort had been put into making the actual pixels capture absolutely the best quality images (some specialized lenses for the pixels were used, too).

Yet the silliness of using math to compare film and digital still continues in some places. There are a few diehards who love film (and film is still a beautiful medium) who want to put down digital rather than simply accept their love of film, so they use old-school math to “prove” how much better their images are than digital.

And there are the stock agencies. I’ve heard a lot of complaints about them from photographers. The agencies arbitrarily used old-school math to determine what they would or wouldn’t accept from digital. They examined digital images much differently than they ever did film. Their excuse was that they needed the highest-quality images for the most buyers, which I can tell you from our experience is nonsense. We’ve purchased stock photos from film for our other magazines that came as digital files that “met” the agency’s “math standards,” yet those files were crappy scans that we couldn’t use at a full-page size. We ended up having to go to the photographers to get new scans of the original images. Yet these were purported to be “agency quality‚” even though any 6-megapixel digital image file (which didn’t meet their “standards”) would have worked.

So where do 35mm film and digital equal each other? How many megapixels do you need? This is a subjective question, of course. How big are the photos? How are they used? Megapixels affect how large an image can be reproduced. If you were making 8×12-inch prints and smaller, you wouldn’t find much difference from film or cameras with 6 megapixels and up. If you made 16×24-inch prints, you’d find that a high-quality 10- to 12-megapixel sensor would easily match 35mm film. And, as noted, the pixels in the megapixels also affect image quality. One reason for buying a newer digital camera today isn’t necessarily for higher megapixels, but for better pixels and other camera technology (such as improved translation of analog data from a sensor to the digital data needed for the image file). In everything we’ve seen at OP, digital capture truly has arrived and has passed 35mm.