HOME
TOPICS
SEARCH
ABOUT ME
MAIL

 
Believe it or not, your printer doesn't care what the DPI of your photo is.
 technofile
Al Fasoldt's reviews and commentaries, continuously available online since 1983

T e c h n o f i l e
Debunking the DPI myth: Why you can ignore dots per inch in photo editing and printing


Dec. 21, 2003


By Al Fasoldt
Copyright © 2003, Al Fasoldt
Copyright © 2003, The Post-Standard

   It's time to debunk the last great mystery of digital photography.
   It's the myth of DPI -- dots per inch. Like 2X4 pieces of lumber, dots per inch aren't really what they seem. That 2X4 you hauled home to fix your closet isn't 2 inches thick by 4 inches wide -- measure it if you think I'm kidding -- and the DPI measurement your printer uses is likely to be just as fictional.
   Professionals who work with image-editing software day after day probably think I've lost my mind. But that's OK. To professionals, dots per inch are extremely important. They couldn't do any serious work without coping with DPI.
   But to folks like you and me, dots per inch might as well be craters on Mars. You can ignore them when you're printing your pictures.
   Dots per inch is one of those terms that seems to need no explanation. A rating of 300 dots per inch means there are 300 dots in an inch. But what are we measuring? Usually, all we mean by dots per inch is the number of separate picture elements, or pixels, in a digital photo, measured across the image. Here's a simple example: You print an image on paper and lay a ruler on it, then count the dots, then divide by the width to get the number of dots in one inch.
   People who are very serious about getting the best quality from their inkjet printers usually think they need to match up the DPI of their printer with the DPI of their photos. They think they have to edit their pictures to give them, say, 1,200 dots per inch because their printer is rated at 1,200 dots per inch.
   But that's a waste of time and a sure way to get bad results. There are two reasons why this is so.
   Your printer's actual DPI is never what it seems. For example, color inkjet printers use at least three inks for colors and one for black. Each ink is put down in a dot pattern. A 1,200-DPI printer that uses four different inks might actually have only 300 DPI. (Multiplied times four inks, it's 1,200 DPI, but that sounds like cheating to me.)
   How can you know for sure what the real DPI is? Most times, you can't. It's a numbers game, and printer manufacturers don't want you to know too much.
   Your printer doesn't care what the DPI of your photo is. The printer creates its own, new pattern of dots when it prints your picture. It always changes the dot pattern of each photo it prints, so trying to match up DPI settings between your images and your printer doesn't help. (And don't forget, as we saw in Reason No. 1, you probably can't tell what the real DPI of your printer is anyway.)
   Are there exceptions? Yes, but only regarding Reason No. 1, and then only in semi-professional printers. Some semi-pro models, especially the ones made by Epson, behave differently, and users of those printers know exactly what the printer's DPI is. In that case, making sure your images have the same DPI as your printer makes sense.
   There's a secret to that, too. If you want to change the DPI setting in your images to match your printer's DPI, don't resample the image. Just change the DPI rating.
   For example, here's how to do it in Adobe Photoshop Elements 2.0, the software I recommend as the best consumer image editing program: Open the Image menu and choose Resize, then Image Size. Type the DPI setting without checking the "Resample Image" box. Photoshop Elements will then show you size the image would be if you printed it on a printer with an honest DPI setting that matches the setting of the photo.
   But telling Photoshop Elements to resample the image instructs the software to do much more than change the DPI setting. It tells the program to add or remove parts of the picture so that the image has the "right" number of dots in each inch. This can turn a good image into a lousy one in the most extreme case. In a less extreme situation, it can make the picture look a little fuzzy.
   Programs other than Photoshop Elements 2.0 might work the same way. Remember, changing the DPI setting on your photos usually doesn't produce better prints because your printer changes the photo's dot patterns anyway. But if you still want to do it, avoid resampling unless you're sure it works for you. In most cases, you should be able to ignore DPI entirely when printing your images.