Photography

Sensor Size, Speed & Resolution

Digital Cameras use an array of photosensitive electronic sensors or imaging chips in place of the traditional film to store images. The two popular imaging chips are the Complimentary Metal-Oxide Semiconductor (CMOS) and the Charge-Coupled Device (CCD). Both types of imagers convert light into electric charge and process it into digital signals which are then sent to the camera's computer system for further processing and storage. The way they do this however is different which gives them both advantages and disadvantages over the other.

CCD sensors gather all electrical signal from each pixel and outputs it through a single node where it is converted to voltage, buffered and converted to an analog signal. This allows it to produce a very uniform output which is a key factor to better image quality. CMOS sensors on the other hand processes each pixel's signal individually, converts them to voltage and often uses amplifiers, noise correction and digitizing circuits to output a proper digital signal.

Earlier sensor designs proved that CCD generally has better quality while CMOS is cheaper and consumes lesser power. However, the technology behind these sensors have evolved greatly during the past few years and no imaging chip is any longer superior than the other in terms of quality, complexity, price and power consumption. There are other factors that are worthwhile checking when looking for a good camera other than the type of sensor. It is equally important to understand the size, speed (sensitivity), and resolution.

Sensor Size

The sizes of imaging chips on digital cameras vary on the type of camera. Compact and sub-compact cameras are generally fitted with smaller sensor to fit their smaller bodies while DSLR cameras have much bigger image sensors. Most high-end DSLRs like Canon's 1Ds, 1DsMkII & MkIII, 5D and the Kodak DCS 14n are equipped with 36mm x 24mm sensor sizes, the same size as 35mm film. This size is the basis as a full-frame sensor. Other less expensive cameras use much smaller sized sensors with widths ranging from 30mm (approx. 1.18") to as low as 8.5mm (approx 1/3"). Some low-end webcams and phone cameras even use much smaller sensors to make them smaller, lighter and cheaper. Sensor sizes are sometimes measured as a factor of the width compared to a full-frame sensor (crop factor). Canon 300D, 350D, 400D, 30D and 40D have a 1.6x crop factor which means that its sensor is only capable of seeing 1/1.6 of the image that a full-frame can see. Nikon D70 and D100 have a crop factor of 1.5x and Canon 1D has 1.3x.

Some sensor sizes are based on Kodak's Advantix film which uses the Advanced Photo System (APS) format. This format has three sizes: APS-H, for high definition is 30.2mm x 16.7mm (16:9 ratio); APS-C, classic is 25.1mm x 16.7mm (3:2 ratio); and the APS-P, for panoramic is 30.2mm x 9.5mm (3:1 ratio). APS-C is very common among a lot of DSLR models although the sizes vary a little bit on different manufacturers. APS-H is also used but not very commonly and is even rumored to be phased out soon. APS-P is not used for practical reasons, the same output can easily be achieved by simple cropping.

The size of the sensor is critical to the quality of the images since it determines the amount of light that the camera absorbs. Imaging sensors with large areas can receive more light and can therefore have a sufficient amount of information to sample and convert into digital data, thus producing a more accurate and high-quality output. Smaller sensors have to perform certain levels of amplification, balancing, and guesswork to produce the best possible output, but due to the limited information it can gather out of its small area the outputs are often inaccurate and noisy especially in poorly lit environments.

Sensor Speed (Sensitivity)

A Film's Speed or Sensitivity determines how quickly it reacts to exposure to light. The lower the sensitivity, the finer the grain, but more light is needed. This is ideal for outdoor photography, portraits or occasions where the subject is not moving and the lighting is very good. For low light conditions or action photography however, where faster shutter speed is needed, a "fast" film with high sensitivity is more useful. The disadvantage of having a high-speed film is that the output tend to be more grainy.

Imaging chips on digital cameras are also capable of varying sensitivity levels or "speeds" rated with the same sensitivity levels for film measured in ISO values based on the old American Standard Association (ASA) values. ISO 100 is considered to be a norm and is ideal for most outdoor photography. Higher ISO values are used if there is inadequate lighting or for capturing fast moving objects such as in sports photography.

The speed of the sensor is increased by amplifying its electronic output, allowing it to obtain equivalent sensitivity levels of ISO 200, 400, 800, 1600, 3200 or even 6400. Boosting the signals may increase the sensitivity of the imaging sensors but it also introduces more noise into the images specially on small sensors. Some cameras employ certain degree of noise reduction to compensate for the noise but this often results in loss of detail and sharpness of the images, so using the lowest ISO setting possible is always desired.

Resolution

If you look closer at the digital images taken using digital cameras or scanners, you will notice that it is actually made up of very fine blocks of varying colors. These blocks are called pixels (short for picture elements) which are the basic element of a digital image. The resolution of a digital image refers to the number of pixels that composes the image. A typical VGA image for example is comprised of 640 columns of pixels with 480 pixels each column or 307,200 pixels (640 x 480). A 5 Megapixel image with 4:3 ratio (width:height) is 2,560 pixels horizontally and 1,920 pixels vertically or 4,915,200 pixels, rounded off to 5 million pixels.

Digital cameras manufacturers are increasingly offering models with high resolutions. 6 and 8 Megapixels are becoming more common. There are even high-end models that has resolutions of 10 and 12 Megapixels. Canon's latest 1Ds Mark III has a 21.1 Megapixel resolution.

An image with a high resolution correspondingly offers more details. In many cases however using a very high resolution is a disadvantage as it makes the image occupy more storage space and makes it slower to load. This also pushes the limits of an imaging sensor as it makes the sensor's pixel size smaller and thus less sensitive, increasing the tendency to introduce more noise. It is always good to determine the right resolution for a particular purpose. For typical 3R or 4R prints a 2MP resolution is already very good. Higher resolutions are ideal for larger prints such as posters and billboards or if you intend to crop certain portions of an image.

Personally I would consider a printing resolution of 300 pixels per inch (PPI) to be superior although lower values up to 200 PPI is very much acceptable. Higher resolutions up to 600 PPI are excellent -- beyond this is an overkill and is just a waste of resources and provides no significant quality improvement. With this basis, at 300 PPI, a 4" x 3" image must have 1200 pixels by 900 pixels or 1,080,000 pixels which is practically what a 1 Megapixel camera can provide.