Dude, Where’s My Photo? Maximizing Megapixels For Print
by John Downey
The past couple of days I’ve been cleaning out my hard drives. During the purge, I found this image of a Mallard among fallen cherry blossoms. The original was in landscape format (3:4), standard for most 35mm cameras. Lately however, I’m interested in the square frame particular to medium format cameras (1:1). While things look ok on the screen (image resolution tops out at a mere 75dpi), cropping from 3:4 to 1:1 reduces the maximum size an image will print well.
For some time, camera manufacturers raced to produce sensors that packed the most megapixels and unfortunately consumers bought into the hype. The bottom line is that most of us are not going to print even our best photos greater than 13×19 unless you have a house with more wall space than Oscar Niemeyer. So do you need a 26MP SLR? Probably not. 10MP will print 13×19 well. And most people on the planet will rarely print at all (enter cel phone cameras which have created an explosive interest in photography, but try to print an 8×10 taken with an iphone). The image above is 279k (little more than a 1/4 megapixel). How big would it print?
Dude, where’s my photo? Compared to the first image, the one above is what you’d get on paper (and this is simulated by the way). Withstanding further geek speak, when images are cropped, precious megapixels are lost and limit the maximum size prints you can squeeze from them. So if you’re into printing or think you’ll eventually want to enlarge, always get in as close as possible to fit the important elements – composition, in-camera, is key. You’ll know you’re getting the hang of things when you stop cropping in post-production. And if you want square, you’ll need to crop unless you’re using a square sensor.
Last point. Most pictures of landscapes are horizontal. Most portraits are vertical. When shooting a headshot in landscape orientation, unless the background provides context or artistic style, you’re wasting the pixels to the left and right of the subject. Rotate the camera ninety degrees and fill the frame with a face and now all of the pixels that sensor is capable of reproducing are representing nothing but the subject (unless your subject has a head like a piano). If a toaster is 3×6, would you insert the long side or short side of a 3×6 piece of bread? Don’t know about you but I want the whole slice toasted. Pass the butter.