How Things Work: Digital Cameras
The Kodak Brownie revolutionized photography in 1900 when it was introduced, promising both cheaper and far simpler means to produce a photograph. It brought photography to the masses.
One hundred years later, the digital camera has supplanted the Brownie as our new form of revolution. In a sense, it’s brought photo development directly to the users: Simply connect your camera to a computer and download the shots. However, it’s also complicated the process far more than the average person wants to think about, using silicon and electrons instead of chemically-coated sheets of gelatin and celluloid. The end process is very different, but at least the basics haven’t changed: Photography is still done by using a glass or plastic lens to focus available light from a scene onto a surface that reacts to that light.
That surface actually changes depending on the type of digital camera that you own, but the two possible options that replace film act largely the same. Most digital cameras today use a “charge coupled device” (CCD) sensor, while others use “complementary metal-oxide semiconductors” (CMOS). Both sensors are found in digital cameras you find every day, and both technologies were developed as early as the 1960s for imaging capabilities. Both are also semiconductors coated with an array of photo sites that each represent a “pixel” of information. The number of “megapixels” that your camera has is essentially just the number of millions of photo sites that the manufacturer was able to cram onto their CCD or CMOS sensor. A 3.3-megapixel camera has 3.3 million such receptors, and so on.
But while both receptors use these photo sites to turn light photons into a string of digital ones and zeros, the ways they accomplish that process is slightly different. CCDs, which are far more common than CMOS sensors, take the analog information that each photo site collects and buffers it; afterwards, it sends it in bulk to a separate analog-to-digital processor. By contrast, each photo site in a CMOS sensor has its own transistor that allows each pixel of information to be read individually. However, these transistors mean that there is less space on the CMOS sensor to actually collect light. It also means that the sensor does the analog-to-digital conversion itself, and outputs ones and zeros without a separate processor.
All of this means, simply, that CMOS sensors are traditionally more bulky than CCDs and produce poorer images because they are less light-sensitive. However, they are also far less expensive to make, and use a lot less power than their CCD buddies. Up until a year or two ago, CCDs were considered the choice for higher-end cameras, while CMOS sensors were limited to cheaper alternatives; recently, however, CMOS technology has improved enough that even some professional digital SLR cameras use them. CCDs still dominate the market, but that may shift in the coming years. Both surfaces replace the age-old film technique in favor of faster usage and quicker processing. While the day when digital photography completely replaces film may be long in coming, there’s no questioning how rapidly it’s found its way into the hands of today’s consumers.