8-bit Versus 16-bit in Photoshop (or How I learned to stop worrying and love pictures with smooth skies)
The first in a continuing series published at Big Swing called Zen Photoshop
So, what’s the difference between working with an 8-bit image and a 16-bit image? Short answer: a lot of color.
Most articles about 8- and 16-bit imagery begin by explaining what a bit is, what bit depth is, and what you get when you add up all the numbers.
Well, I can hardly keep my eyes open reading the technical stuff. I’m an artist, not a mathematician. Let’s leave the mechanics to the engineers, and focus our attention on how a basic understanding of bit depth can enhance your aesthetic vision and potentially speed your workflow. (For a good explanation of the numbers, see http://www.photoshopessentials.com/essentials/16-bit/)
The big difference between working in 8 and 16 bit is how many unique colors and tones your image contains. Think of it this way: 8 bits can hold a gallon’s worth of color and 16 bits can hold more than one million gallons!
Why does this matter? Because the more unique colors your image contains, the greater flexibility you have to change the aesthetics of your image. With 8-bit images, even modest contrast and color adjustments can produce unintentional breaks in tone and loss of image quality, whereas working in 16 bit allows you to really push your image around and maintain good detail and smooth transitions between colors and tones.
For example, how many times have you worked for hours on a photograph, making big tonal adjustments, subtle color changes, and retouching only to end up with what look like “stair steps” throughout your image? This “stair stepping” (also called “banding” or less frequently “posterization”) is an artifact of your retouching and digital darkroom work, resulting from working in 8 bit. The problem is you’re asking your image to express more colors than it contains.
Illustration 1 presents the classic scenario, but certainly not the only one, where working in 8 or 16 bit can really make the difference between a beautiful image and one that is unusable. Notice the color breaks in the sky. These breaks could have been avoided by working in 16 bit.
Illustration 1: An 8 bit image. The sky has been darkened using a series of very soft selections, Curves Adjustment Layers, and Layer Masks. Note the banding apparent in the sky. This banding would be even more apparent when printed. image by Stephen Yadzinski
Detail of Image 1
So, if you had worked in 16 bit, the stair stepping would not have occurred? Yes, that’s what I’m saying. But remember, when working in Photoshop the preview on your desktop is 8 bit. Even when you’re working in 16 bit, you will still see the banding until you flatten your image. Once you flatten your file, Photoshop re-draws the preview, and your image should appear as smooth as a baby’s bottom—check out illustration 2.*
Illustration 2: A 16 bit image. The same image as in Illustration 1. In this version all of the adjustments applied are identical to those in the 8 bit version, but using 16 bit we now have a beautiful smooth sky! image by Stephen Yadzinski
Detail of Image 2
And don’t worry, if you’ve got and image in 8 bit that has banding or other loss of detail issues, your hours of hard labor are not lost. To fix the image, you have (at least) a couple of options:
1. Re-process your original RAW file in 16-bit. Move all your adjustment layers from your original 8-bit file to your new 16-bit file. Remember to hold the shiftkey down while dragging the adjustment layers from one file to another—this will keep the layers in proper registration. Once you flatten the new 16-bit image, any banding issues should disappear.
2. You can also try converting your image from 8 to 16-bit. With your 8-bit image open, go to Image > Mode > 16 bits/Channel. Try flattening your image, and in many instances the image will be smooth, with no apparent banding.
That’s pretty much it. The discussion of whether to work in 8 or 16 bit is often more confusing and complicated than necessary. Simply put, always work in 16 bit.
There are a few downsides (files take longer to save, not all filters will work) but considering the substantial increase in image quality, the few limitations are more than worth it. And remember, if you come across a situation where 16 bit is not an option (for example, your favorite filter is unavailable), there is always a workaround.
*An Adobe person at some point told me this. Unfortunately, I was unable to verify that Photoshop creates an 8-bit preview of your image to display on screen. However, it is correct that when working in 16 bit, you will often see banding until you flatten your file.
Filed in How we get stuff done, Workshops, Zen Photoshop



4 responses to 8-bit Versus 16-bit in Photoshop (or How I learned to stop worrying and love pictures with smooth skies) (Add yours)
1. Marc — Nov 11 2009 03:20 PM
Great article, Steve. Is there a way to take advantage of 16-bit color if the result must be a jpeg?
2. yadz — Nov 11 2009 03:31 PM
Hey Marc,
Assuming you are shooting RAW to start, you can:
1. Process your RAW image as 16 bit
2. Make all your adjustments in Photoshop
3. Flatten your image
4. and as a last step go to Image > Mode > 8 Bits/Channel
This will convert your image from 16- to 8-bit, and will allow for JPG file saving.
If you must start with a JPG image (as opposed to a RAW) then you may first convert to 16-bit from 8 and make your adjustments. Then, as a last step, convert back to 8-bit and save as a JPG.
3. Mihai Bocsaru — Nov 12 2009 06:36 PM
Great Article!
4. darrell joslin — Jul 18 2010 11:13 AM
great article. i am glad you left out all the boring tech stuff.this made the whole thing a lot clearer.