When do you really need the 16-bit capabilities of RAW files?
By Rob Sheppard
More isn’t always better, but sometimes it helps. More strokes in a golf game is a bad thing, while more runs is good for baseball (unless you’re on the losing side). More heat is great for baking bread, but awful for storing eggs. Now, what the heck does this have to do with digital photography? Quite a lot, actually, when you look at 16-bit vs. 8-bit images in Photoshop. Bit-depth refers to the way digital data handles tonal and color information for a photograph. There are a lot of absolutes going around about using them, which are sometimes misleading and confusing. These two bit-depth choices are important to understand so you can make the best choice for your needs. They influence whether you shoot JPEG or RAW and how you process RAW files.
JPEG is an 8-bit file, while a RAW file is 16-bit, though it really isn't that simple. Digital SLRs and compact digital cameras typically start with 12-bit files coming from the sensor (it can't capture 16 bits of data). A 12-bit camera file is smartly converted to an 8-bit JPEG image inside the camera. For a RAW file, this 12-bit original data is prepped and placed into a 16-bit file, but you get no more data than the original 12-bit image (the 16-bit file is simply a big space in which to put the 12-bit image).
Without going into all the math, let's look at how these bit-depths affect color:
1 8-bit RGB can represent 16,777,216 distinct colors (millions of colors), which is plenty for making a true photographic print.
2 12-bit RGB represents 68,719,476,736 distinct colors (billions of colors), which adds adjustment room in which to work (by a factor of over 4,000!). 3 True 16-bit color (which isn't what comes from most digital cameras) represents 281,474,977,000,000 colors (trillions of colors), which is way more than we humans can distinguish.
On the surface, and especially since some think the bigger the better, the 12/16-bit RAW file seems to be a much better deal. And it is, if you need the data. Think about finding a friend in a small party of 20 people versus at a football game with a crowd of 80,000. If your goal is to find only that person, those extra people are of no help. On the other hand, if you're trying to get signatures for a referendum, the small party of 20 does little good compared to the football crowd.
A JPEG file is quick and easy to adjust, and for many scenes, may be all you need. On the other hand, some scenes simply don't do well with the 8-bit data of JPEG and need the 12 bits of information in the 16-bit RAW file. These are typically images that need strong adjustments across subtle gradients. The big advantage of JPEG is speed and ease of work; the big advantage of RAW is a hugely increased adjustment space.
It can be difficult to know if a particular scene will have problems when shot with a particular bit-depth. I can tell you from experience of shooting thousands and printing and publishing hundreds of images in JPEG that you can shoot JPEG alone and get quality images almost all of the time. However, I can also tell you from experience that RAW files will provide better images in many situations and will output the best images in certain challenging conditions (which might not be what you expect, as we'll note in a moment).
The way to deal with this is to shoot RAW+JPEG with any camera that offers the choice. It guarantees that you have both JPEG images that can be dealt with quickly and easily (plus they include special internal camera processing that RAW does not) as well as RAW files for when you need them. You gain the choice of either.
This wasn't so much an option for most photographers a few years ago. RAW+JPEG took up a lot of storage space and slowed down cameras. Cameras can deal with this today, plus memory cards and hard drives are quite reasonably priced for a lot of megabytes and gigabytes, respectively. Now RAW+JPEG is a viable way to go. Some of the newest cameras (such as Canon's EOS-1D Mark II N) allow you to record these shots to separate memory cards (e.g., RAW to the CompactFlash card, JPEG to the SD card) for more convenience.
There are certain images that I'll open first using the JPEG file. These typically are properly exposed photos with a good range of tones and colors. There also are certain pictures that I'll open right away into Adobe Camera Raw. These usually are scenes with very bright highlights, images with strong detail in the darkest areas, photos shot under very low-contrast situations and scenes with subtle gradients.
Almost everything I shot a few years ago was JPEG. As I increased my RAW shooting, I was pleased to discover that more and more cameras offered the RAW+JPEG setting, and I began to use it (now I always use it if it's available on a camera). Today, about 60% of my photos come directly from the JPEG file and 40% from RAW.
The photo with this column is a good example of RAW helping to get more from the image. I saw these immature lubber grasshoppers hanging out on a bromeliad leaf in Loxahatchee National Wildlife Refuge in southeastern Florida. The day had started with some light at dawn, but clouds had taken over the sky. I was on my way back to the car, because raindrops were starting to fall, when I saw the grasshoppers.
I got out a telephoto zoom and lined up the camera parallel to their perch. I used a wide lens opening to blur the background and shot a few variations of the scene until the raindrops got to be just too much. I knew this could be a challenge because of the low-contrast conditions and the out-of-focus background. The latter typically results in nice gradients of tone and color for a background, but it also stresses the bit-depth of an image.
I didn't even bother with the JPEG on this one and went right to the RAW file. I did what I felt was optimal processing in Camera Raw and opened the image in Photoshop. Here's where it got tricky. I usually set Camera Raw to open the image as 8-bit. This has no effect on the processing of the image inside of Camera Raw (it's still done with the original 12-bit camera data), but it does affect the processing within Photoshop.
The reason I usually select 8-bit is that, most of the time, you can complete the needed 16-bit work inside Camera Raw, and a 16-bit image is a big file in Photoshop. While Photoshop CS2 is perfectly capable of dealing with 16-bit on many levels, the file size of a 16-bit image will climb in a hurry. An 8-megapixel RAW file opened at 8-bit is approximately 23 MB, but at 16-bit, this doubles to 46 MB. Add layers or open more than one image and you quickly can stress your computer, which will slow it down significantly. This is needless if the 8-bit file can handle the finishing work (and usually it can).
When I opened the grasshoppers in 8-bit and tried to further enhance the image to fully realize the original shot, however, I found the background tonalities started to show signs of banding (caused by incomplete data for the gradations). I reopened the image in Camera Raw, which remembers how the image was opened before, so I didn't have to do any readjusting. I simply selected 16-bit in the Depth box.
When I opened the photo and made the same adjustments as before to enhance the image, the background tonalities held their gradations quite smoothly, with no banding. In this case, 16-bit processing throughout was a necessity.
The choice between 16-bit or 8-bit, RAW or JPEG should never be an either/or proposition. I believe it needs to be strictly related to the photographer and the photograph's needs. This is why shooting RAW+JPEG is such a great way to go with today's gear.
OP editor Rob Sheppard's latest book is Adobe Camera Raw for Digital Photographers Only. His new website, www.robsheppardphoto.com, features photo tips and more.