This is one of those subjects that makes people’s eyes glaze over, but it’s important for learning how digital cameras process images and what that means in post-processing. Most people know it has something to do with color and that more bits are better, but that’s about as far as it goes.
The basic terminology is fairly simple. A bit is a basic unit of data processing and is expressed as either a 1 or a 0. Eight bits, also called a byte, can be used to express 256 different states or 2^8.
Most digital pictures, inkjet printers and most color monitors express colors at an 8 bit color depth per channel. So that’s 256 shades each of red, green and blue. Each pixel is then represented by the color which is the sum of all three color channels.
As you can see, in a 16-megapixel chip there’s an aweful lot of processing going on, even for a simple 8 bit image. But modern cameras don’t shoot 8 bit anymore, most are shooting 16 bits per channel, 65,536 possible representations of each color channel, or 48 bit color depth. A 48 bit color image is capable of producing billions of colors.
Many cameras are pushing 22 bits per channel, or 16,777,216 shades of each channel. With a 20-megapixel chip, if nothing else you should gain new respect for what’s going on inside your camera just to record the image. Hopefully, these comparison specs, will mean more to you now.
When it comes to working with images in post-processing, generally working at higher bit rates will yield more subtle variations in tone. But don’t feel that you have to chase ever increasing color depth. The human eye can only discern about 10 million different colors, about 24 bits per pixel. So, for display, anything beyond that is not going to look that much better to viewers.