Dynamic Range and Sensitivity
One way to do this is to give the photosites a larger surface area, which increases the volume of the bucket and allows collecting more photons. In fact, the jumbo photosites in larger dSLR sensors (so-called full-frame models) allow greater sensitivity (higher ISO settings), reduced noise, and an expanded dynamic range. In the past few years, a metric called pixel density has provided an easy way to compare the size/density of pixels in cameras, even those with different absolute resolutions. For example, two cameras from one vendor share the same 12 megapixel resolution. One has the typical APS-C sensor size (more on that later) measuring about 24mm × 16mm, and with 3.3 megapixels packed into each square centimeter of sensor area.
The other camera is a full-frame model with a 24 × 36mm sensor, and with more room in its larger sensor to fit in 12 megapixels worth of photosites, it has a pixel density of only 1.4 megapixels per square centimeter. Clearly, the camera with the larger sensor has larger pixels that are more sensitive to light, and which can provide better image quality, particularly at higher ISO settings.
Pixel density allows you to compare non-similar sensor sizes and resolutions, too. The same vendor offers a 24 megapixel camera with a 24 × 36mm sensor. It must squeeze 2.8 megapixels into each square centimeter. So, as you might guess, this camera’s pixels are smaller than those of its 12 megapixel stablemate (and therefore not as sensitive to
light). Its resolution is higher, but high ISO performance is not as good. Indeed, the 24 megapixel camera’s 2.8 MP/cm2 pixel density and light-gathering power is more similar to that of the 12 megapixel APS-C model with 3.3 MP/cm2 pixel density. Dynamic Range and Tonal Values While we’re always interested in the amount of detail that can be captured under the dimmest lighting conditions, the total dynamic range of tones that can be grabbed is also important. Dynamic range can be described as a ratio that shows the relationship between the lightest image area a digital sensor can record and the darkest image area it can capture. The relationship is logarithmic, like the scales used to measure earthquakes,
tornados, and other natural disasters. That is, dynamic range is expressed in density values, D, with a value of, say, 3.0 being ten times as large as 2.0.
As with any ratio, there are two components used in the calculation, the lightest and darkest areas of the image that can be captured. In the photography world (which includes film; the importance of dynamic range is not limited to digital cameras), these components are commonly called Dmin (the minimum density, or brightest areas) and Dmax (the maximum density, or darkest areas).
Dynamic range comes into play when the analog signal is converted to digital form. As you probably know, digital images consist of the three color channels (red, green, and blue), each of which has, by the time we begin working with them in an image editor, tonal values ranging from 0 (black) to 255 (white). Those 256 values are each expressed
as one 8 bit byte, and combining the three color channels (8 bits × 3) gives us the 24 bit, full-color image we’re most familiar with. However, when your digital SLR converts the analog files to digital format to create its RAW (unprocessed) image files, it can use more than 8 bits of information per color channel, usually 12 bits, 14 bits, or 16 bits. These extended range channels are usually converted down to 8 bits per channel when the RAW file is transferred to your image
editor, although some editors, like Photoshop, can also work with 16 bit and even 32 bits-per-channel images.
The analog to digital converter circuitry itself has a dynamic range that provides an upper limit on the amount of information that can be converted. For example, with a theoretical 8 bit A/D converter, the darkest signal that can be represented is a value of 1, and the brightest has a value of 255. That ends up as the equivalent of a maximum possible dynamic range of 2.4, which is not especially impressive as things go.
On the other hand, a 10 bit A/D converter has 1,024 different tones per channel, and it can produce a maximum dynamic range of 3.0; up the ante to 12 or 16 bits (and 4,094 or 65,535 tones) in the A/D conversion process, and the theoretical top dynamic ranges increase to values of D of 3.6 and 4.8, respectively.
These figures assume that the analog to digital conversion circuitry operates perfectly and that there is no noise in the signal to contend with, so, as I said, those dynamic range figures are only theoretical. What you get is likely to be somewhat less. That’s why a 16 bit A/D converter, if your camera had one, would be more desirable than a 12 bit A/D converter. Remember that the scale is logarithmic, so a dynamic range of 4.8 is many times larger than one of 3.6.
The brightest tones aren’t particularly difficult to capture, as long as they aren’t too bright. The dark signals are much more difficult to grab, because the weak signals can’t simply be boosted by amplifying them, as that increases both the signal as well as the background noise. All sensors produce some noise, and it varies by the amount of amplification used as well as other factors, such as the temperature of the sensor. (As sensors operate, they heat up, producing more noise.) So, the higher the dynamic range of a digital sensor, the more information you can capture from the darkest parts of a slide or negative. If you shoot low-light photos or images with wide variations in tonal values, make sure your dSLR has an A/D converter and dynamic range that can handle them. Unfortunately, specs alone won’t tell you; you’ll need to take some pictures under the conditions you’re concerned about and see if the camera is able to deliver. (Which is another good reason to buy your camera locally, rather than purchase a pig in a poke from a mail-order or Internet vendor.)
Controlling Exposure Time
This wonderful process of collecting photons and converting them into digital information requires a specific time span for this to happen, known in the photographic realm as exposure time. Film cameras have always sliced light into manageable slivers of time using mechanical devices called shutters, which block the film until you’re ready to take a picture, and then open to admit light for the period required for (we hope) an optimal exposure. This period is generally very brief, and is measured, for most pictures taken with a hand-held camera, in tiny fractions of a second.
Digital cameras have shutters, too. They can have either a mechanical shutter, which opens and closes to expose the sensor, or an electronic shutter, which simulates the same process. Many digital cameras have both types of shutter, relying on a mechanical shutter for relatively longer exposures (usually 1/500th second to more than a second long), plus an electronic shutter for higher shutter speeds that are difficult to attain with mechanical shutters alone. (That’s why you’ll find a few digital cameras with shutter speeds as high as 1/16,000th second: they’re electronic.)
Mechanical shutters can work with any kind of sensor. One important thing to remember about a digital SLR’s echanical shutter is that its briefest speed usually (but not always) determines the highest speed at which an electronic flash can synchronize. That is, if your dSLR syncs with electronic flash at no more than 1/160th second, that’s probably the highest mechanical shutter speed available. Some special flash systems can synchronize with electronic shutters at higher speeds, but I’ll leave a detailed discussion of syncing for Chapter 3.
Next Topic-- How We Get Color.