When gathering images to pass to you designer for a project, she may request that you find images that are “high resolution,” “300dpi,” or something similar. Here’s how to check the size of an image, quickly and easily:
On a MAC
Open Image in Preview.
Go to Tools // Show Inspector, or press Command+I.
Image size (in pixels, width by height) and DPI will show up.
On a PC
Right-click on the image filename or icon.
Select the Summary tab.
Width and Height will be shown, as well as vertical and horizontal resolution, which should be the same number. If something has a horizontal resolution of 72 dpi and a vertical resolution of 72 dpi, then you’d simply say its dpi is 72.
So, what differentiates a big image from a small one? It depends on how it’s going to be used, really. In either program, press Command+0 (zero) to see the image at full screen size. That’s the biggest it should appear on any screen application (web, tv, any digital medium). Any larger, and it’ll be pixellated. In print, however, its optimal size is much smaller than its screen size. So if it’s small on screen, it'll be smaller in print, and it likely won’t print well.
Why does it matter?
On screen, it's only size that matters—dpi is irrelevant. If a picture is 3000 x 1000 pixels, its actual size depends on the size and density of the pixels on the screen. The smaller and more tightly packed the pixels are, the smaller the image will appear. Think about a high-density screen, like Apple’s iPhone retina display, with pixels too small to be detected by the naked eye. Compare that to a low-density screen, like a digital billboard, whose pixels are so big, you can sometimes see them from the street. A 3000 x 1000 image will be much smaller on the retina display, because the pixels are so much more dense than on the billboard.
So why do people (maybe even your designer) say screen images should be 72dpi? Back in 1984, when the first Macintosh was introduced, the height of an inch on the screen measured 72 pixels. The screen was literally 72dpi. As technology has evolved and screens have gotten way more dense, we continue to measure digital images by pixels—that 72dpi rule-of-thumb is just residual (get nerdy and read more about that here).
For print, however, dpi is critical. Whether your project is being printed on a laser printer, a digital press or offset, the process is the same—tiny dots of ink are being printed onto paper to create areas of color and shape. You’ve probably seen a close-up of a CMYK print before, like this. Print images are just dots of vivid cyan, magenta, yellow and black. But when viewed with the eye, those dots are indistinguishable (this works the same when printing with spot colors, by the way).
Three hundred dots per inch is the necessary resolution to print images whose individual dots of ink aren’t evident to the human eye. At a higher DPI, more detail may be printed, but our eyes can’t notice it. At a lower dpi, the printed image will appear blurry or grainy.
So why can’t you just “res it up”? I love in crime tv shows when the detective is watching the surveillance video and he waves his hand dramatically and says, “enhance,” and the tech presses some button and it zooms in and what was blurry automatically becomes clear and the killer is identified and all is well (seriously, the video below is hilarious). Read my lips: this is not possible. The digital file doesn’t know what the pixels in between the pixels look like. If the image is blurry, it’s blurry. When you’re actually taking a picture with a camera and you zoom in, you get a closer image because the lens is actively focusing at a new depth. But once that digital file is recorded, there’s no changing it. It is what it is.
What I'm saying is, it doesn't work like this. Sorry.
But there will inevitably be times when you have an image that’s not big enough to print, and it’s the only one you have, and it absolutely has to be printed at a larger size. Your designer can make this happen, within reason. The technical term for the task is resampling, because the existing pixels have to be spread out, and new pixels are created for the in-between places. The software literally has to guess at what the new pixels should look like, based on the ones around them. This results in a blurry or pixely image, and it should only be done when necessary. Good designers have smart ways of doing this to produce results that maintain a smooth, natural feel, and good software can go back and add contrast and sharpness after the resampling. But it’s never as good as the real thing. So when you take pictures, take big ones. It’ll pay off down the road.
Resampling down is easy. It’s also necessary for a designer to do—if she prints a 300 dpi image at half its original size, she’s going to lose detail and degrade any filters or effects she applied while editing unless she resamples. Good designers should always use smart methods to do this (like ALWAYS saving the smaller file with a new filename). But it’s easy. So send her the biggest file you’ve got.
One last thing to bring it all together: Resizing is when you change the dpi of an image, and the width and height adjust accordingly (or when you change the measurements of an image, and the dpi adjusts accordingly). So if an image is 4” x 3” and 600 dpi, it can be resized to 300dpi, and it’ll then measure 8” x 6”. Resizing just spreads those dots out to be 300dpi, so the image is much larger. No resampling involved. Cool, huh? In the same way, if you have 4” x 3” image that’s 72dpi, resizing it to print at 300dpi results in an image that’s just 0.96” x 0.72”! That's why something can look good on screen, but be basically unusable in print.
And that, dear clients, is why your designer keeps pestering you to not take pictures with your flip phone to include in the annual report. Size matters!