A while back, on The Online Photographer, I linked to a video segment from a British television show called The Gadget Show. It documents a not-so-rigorous test of film vs. digital in which the two hosts dress up as the main characters from the old “Avengers” TV show, get their pictures shot with a film 35mm camera (a Nikon F5) and a digital 12 MP camera (a Nikon D700), and then have 17-meter-high prints made from both, which they hang up on the side of a building to evaluate. Not the best test (not to mention the fact that the whole film vs. digital thing is sort of a “who cares” issue these days), but I was impressed by their budget!
After I posted the link, several TOP readers posted comments along the lines of, “Well, I’m never going to worry about the enlargeability of 12 MP digital files again.”
Really? Not so fast.
One thing those commenters might have been forgetting is that print size scales with viewing distance. Looking at a print that covers the side of a building from 200 feet away might be little different than looking at a 6×9” print placed eight inches from your nose.
Years ago I sat in on a class at a college I was thinking about attending. During the lecture, which was about reality and illusion, the instructor tried to claim that images in mirrors are not accurate. He told his students that if you stand in front of the mirror and trace the outline of your head on the mirror in grease pencil, you’ll see that the tracing is too small—and that, therefore, the mirror “miniaturizes” what it reflects.
At the time, I had a tendency to be a gadfly, terrorizing instructors with sharply-pointed questions, but on that occasion, as a visitor, I was trying mightily to be on my best behavior—so I didn’t raise my hand and innocently ask, “So, is the moon miniaturized because I can cover it with my thumb?” What he wasn’t taking into account (probably because he didn’t know) is that the virtual image in a mirror is as far in back of the plane of the mirror as the object it’s reflecting is in front of it—for a very simple reason: if you’re standing two feet in front of the mirror, the light reflecting from you has to travel two feet to get to the mirror and two feet to get back to your eyes. Therefore the virtual image is four feet away from you not two, and the grease pencil tracing on his mirror is not miniaturized—it’s exactly the same size as a same-size head would be on a person four feet away, or twice as far away from you as you were from the mirror when you made the tracing.
Another way you can think of this is with TV or movie screen sizes. Consumers tend to treat TV screen sizes as though they are absolute, but they also have to take into account how far away they like to sit. If you buy a 64” screen and sit fourteen feet away from it, you’re getting quite a similar viewing experience to sitting seven feet away from a 32” screen. There are a few other factors involved, of course—the distance at which your eyes focus comfortably, how big your TV room is, how many people have to be able to see the TV at once, and so forth. (Not to mention that big TVs can be status objects with which to show off wealth.) When considering big flat-screen TVs, buying a somewhat smaller one and simply sitting closer to it might be a good way to save some money. In a movie theater, I tend to sit the “same” distance from the screen—but it’s only the same in relative terms: in absolute terms, I sit closer to smaller screens and farther away from larger ones.
Or consider all those billboards along the side of the road. The pictures on them weren’t shot with enormous banquet cameras—they were shot with Hasselblads and Nikons, and, in some cases now, DSLRs, just like the ones you can buy. The only reason they look good is because you’re far away from them when you drive by on the highway. I guarantee that you wouldn’t be satisfied with the “enlargement quality” if you were the paper hanger who put the billboard up.
Generally, the gold standard for print viewing is still what it looks like on a wall—that is, a wall where you can decide where you want to stand to look at it, such as a museum or gallery wall with nothing physically enforcing your chosen viewing distance. In such cases, people will naturally tend to take two standpoints—the first at a distance where they can comfortably take in the whole work, and the second a closer position to better scrutinize details. Observe your fellow museum-goers the next time you’re in a museum and you’ll see what I’m saying.
If you do, you’ll also observe that people naturally take a standpoint that is relative to the size of the artwork. If they’re proceeding around a room looking at one painting at a time and come across a particularly large painting, they’ll stand back. If they come across one that’s much smaller than the rest, they’ll step forward.
Not only that, but the “scrutinizing” standpoint adjusts as well. People will “peer at details” from a closer distance with smaller works of art and a farther distance with larger ones. Note, for instance, in the “Avengers” test, the Gadget Show hosts first look at their 17-meter-high prints from a comfortable distance:
...Then the fellow in the red coat says, “Right, let’s have a look at the details,” and he scurries in (with the video appropriately speeded-up) for a closer look, to roughly this distance:
So notice where he goes to “peer” at the prints—closer, and too close for him to take in the whole work comfortably, but he’s still fifteen or twenty feet away from the prints! (The two more distant figures are walking in from their first standpoint, which was a little farther out than they are in this second frame-grab.) Now, ask yourself, how many photographs in museums, on walls, do you ever look at from fifteen feet away?
What I think you’ll notice if you watch people looking at prints is that even the distance at which they scrutinize details scales to the size of the whole work. The distances will vary with individuals, but even when most people look closely, they still tend to be farther away from bigger works and closer to smaller ones.
So the Gadget Show test doesn’t really prove much about what sort of enlargeability you can get from a certain size sensor, because print size has both an absolute aspect (the prints on the building really are 17 meters high) but it also has a relative aspect (you’ll stand much farther away to look at them than you would to look at smaller prints). To evaluate the enlargeability of your prints—film or digital, it doesn’t matter—you need to make prints the size you want them, then look at them in the way that feels most natural. If they look good to you that way, then they’re good enough.