Let’s start this tip with a caveat: I think it’s best to capture full-resolution RAW image files in-camera. This is how I do it, and it’s how I ensure the highest-quality image file with the most options after capture. But sometimes I need to shoot JPEGs, too. Some might say they shoot JPEGs “instead,” but I still prefer to save those high-quality RAW files. In my estimation, the only folks who should be capturing only JPEG files in-camera are those professionals with such tight turnarounds and heavy processing and transmission needs that the very act of shooting RAW files would make their job impossible. Those situations are very few and far between. So, in general, shoot RAW, and if you need JPEGs, too, set your camera to capture them both. Now, on to the issue at hand.
If you’ve determined that you need to capture JPEG files—in my case, this happens when a client needs some immediate turnaround of image files for use on social media—the next step is to determine what resolution and compression options work best for the situation. Here’s how I think about it.
Above all else, I’m primarily concerned with resolution. This is often expressed colloquially as megapixels, though that term doesn’t often show up in camera menus. It’s just another method of expressing area resolution: How many pixels wide and how many pixels tall is the image? A chessboard, for instance, would be 16 squares wide by 16 squares tall. If those squares were pixels, at four pixels per inch (4 ppi), the image would measure 4 by 4 inches. That’s awfully low resolution by any measure.
A 24-megapixel DSLR, however, produces a much higher-resolution image file—roughly, 6000 pixels by 4000 pixels. (Multiply those numbers together, and you get 24,000,000, or 24 megapixels.) At 300 ppi, the commonly accepted standard for print resolution, the 6000 pixels translate to 20 inches wide, while the 4000 pixels work out to about 13 inches tall. So a 24-megapixel image produces a full-resolution image file that will deliver a 13×20-inch print, no questions asked. Now there’s a lot of gray area when it comes to making quality prints from lower-resolution files. But for now I’m discussing raw numbers. If the resolution math works out, you’ll have no problems when it comes time to print or digitally display an image file.
A Nikon D610, for example, offers nine JPEG settings—three each in Large, Medium or Small resolutions, with compression that’s Fine, Normal or Basic. Let’s break down those terms.
Large, Medium and Small refer to the area resolution. The Large setting is 6016×4016. Medium is 4512×3008, and Small is 3008×2008 pixels. This is where I start, the foundation of my file size, based on what I want to do with the image file.
Is it destined to live online only? In most cases, the smallest resolution (3008×2008) is more than enough for full-screen use on most displays. But what if it’s to be displayed full screen on a 4K monitor or a 5K Retina display? In those cases, higher resolution may be necessary to ensure your image file contains enough pixels to fill the frame. When it comes to physical dimensions, whether virtual (on screen) or tangible (a print), it all starts with area resolution. That’s why the first setting I determine is resolution.
For most web use, Small is fine. For images that will need to be cropped, or used full screen on higher-resolution displays, or printed at basic sizes, upgrading to Medium size, 4512×3008, is probably a good idea. And, if I know I need to make big prints or significantly crop to make an image more of a close-up, I want all the resolution I can get, so I choose the Large resolution setting.
Remember, it’s this area resolution that first and foremost establishes the dimensions of the image, whether on screen or on paper.
From there, it’s time to choose the compression, or what Nikon calls Fine, Normal or Basic settings.
Fine, as you may imagine, is the least compressed and highest quality. Edges will appear sharper, and fine details will be easier to make out. The file size on a Large Fine JPEG file from the D610 is about 14 MB. Shoot a lot of those, and you’ll find your cards and hard drives filling up fast.
Normal compression cuts that file size significantly—anywhere from 25% and up, depending on the resolution of the image file. The lower the resolution, the greater the space savings, proportionally speaking. But normal compression does sacrifice some fidelity. The question is, can you tell? The only way to determine this is to use your own camera and put it through the paces at the various resolutions and compression settings. This way, you can make a judgment as to whether Normal compression is too much, or whether even Basic compression is just fine.
In the examples here, it’s all but impossible for me to see the difference between the minimum compression and the medium amount. The maximum compression will create jagged edges and some pixelization, but it’s only evident if you look particularly close, or if the image is particularly low resolution. As long as you started with lots of pixels, you can get away with more compression. And check out the file sizes! For the examples shown here, which are all the same resolution, the lowest compression produced a file of 846 KB. Medium compression trimmed that by about 85%, down to 126 KB, while the largest amount of compression created a file that registers just 56 KB. These aren’t insignificant file size reductions.
The overarching point here is that you can’t put the cart before the horse. I recently had a discussion with a photographer friend who was thinking in terms of Fine, Normal and Basic before he had even set the resolution. That’s a recipe for confusion, if you ask me. Start by determining the physical size at which an image will need to be displayed, set the resolution accordingly, and only then should you determine whether you need to step down from the highest-quality compression (Fine) to either Normal or Basic. I would argue that you probably shouldn’t use Basic unless some extraneous circumstance makes every bit of space savings a matter of getting the shot or not. In most cases, the question will be whether Fine quality works for your workflow, or whether you need to save the space and time (download, processing and delivery time) to Normal.
Canon users will have the same choices, just with different descriptors. An EOS 5D Mark III, for instance, uses L, M and S to stand in for Large, Medium and Small, and a smooth or a jagged curve adjacent to those initials signifies better quality (smooth) or higher compression (jagged).
The next time you need to capture JPEGs along with RAW files, start by determining the necessary dimensions and set the area resolution to Large, Medium or Small accordingly. Then you can compress the image, as needed, to meet your space and turnaround requirements.