One of the most overlooked aspects of a digital camera's design is the CCD imager itself. Most people simply look at the number of horizontal and vertical elements, and the size of the imager, and are satisfied with that. A lot of science and technology goes into making an imager that delivers truly high quality pictures.

One of the first aspects to consider is the geometry. Overall, the vast majority of imagers used in digital photography are rectangular. This is mainly because the processes that produces the imagers are borrowed from the world of microchip manufacturing, microchips also being rectangular While this factor seems a bit of a copout, it doesn't hurt that the pictures we take, print, and publish are also mostly rectangular.

As a general rule (but not a strict requirement), imagers follow an aspect ratio of four to thee (expressed as 4:3). This ratio closely matches the aspect ratio of normal human vision. Humans have binocular vision, which means we have two (bi) eyes (ocular). When we view the world, our brains guide our eyes to a single point, and the line of site from each eye converges. The result is two separate but partially overlapping images that our brains assemble into one single image. Where each eye would have a vertically and horizontally symmetric field of vision separately, the combined image is wider than it is tall. This trick of human sight affects our perception of the world, and we are often unaware just how much more comfortable it is to see images that have this or a similar aspect ratio. This is why movie theater screens are shaped the way they are, and why TV's will soon be similarly shaped.

The next consideration is the symmetry of the individual elements of the imager itself. When you look at an imager, it appears to have a glassy smooth surface. But when you look at it under magnification, it is actually thousands of tiny elements. Each of these elements is an individual CCD that produces an electronic signal when light hits it. That signal is interpreted into a digital number to represent a part of the image. The shape of the individual element is important. The shape must cover as much area as possible, while allowing the least amount of area to form as a gap between it and other elements. Light that falls into those gaps is lost information that is not recorded.

To this end, most elements are square, or as nearly square as possible. Like the tiles of a bathroom floor or wall, this simple design makes it easy to build the elements together and pack them in tightly. However, this is not necessarily the best arrangement, and many other element designs have been tested. Most notably, Fuji uses an octagonal element array on their imagers. Since the octagons are more nearly round, but still extend to fill the spaces between elements, the octagonal elements yield a better diagonal resolution. Since humans relate to our world in horizontal and vertical planes, some question whether this increased resolution in the diagonal plane makes a noticeable difference.

Finally, one of the biggest factors affecting CCD performance, and therefore one of the biggest considerations, is quality of materials. Without getting too heavily into the science of the materials, one should know that inexpensively made imagers using cheap or impure semiconductors and less than adequate processes are prone to a number of visual artifacts such as poor color depth, blooming, low dynamic range, and slow readouts. Since you cannot be expected to know all of the various compounds used in manufacturing, it is always prudent to go with the "safe bet", and stick to using equipment known for quality imagers. Whether you are purchasing a scanner, digital still camera, or digital video camera, you should stick with brand names you are familiar and comfortable with.