The Three Variables of Exposure

A photographic image is created by exposing film, or a digital sensor, to light. However, if too much or too little light reaches the film or sensor, the image will be overexposed or underexposed, respectively, possibly to the point of rendering the image unrecognizable.

Prior to the advent of quality digital cameras, most serious wildlife photographers shot color reversal (slide) film. This required more precise exposure control than color print film because minor exposure errors in the latter could be corrected when printing from the negative. On the other hand, with reversal film the final product was the film that had been in the camera and, once developed, could not realistically be modified. Quality digital cameras permit greater exposure flexibility than either type of film because of their sensitivity to a greater brightness range and because changes to RAW digital images can be made during processing.

There are three variables that you must balance to achieve a desired exposure: ISO, shutter speed, and aperture. ISO is the measure of how sensitive the film or sensor is to light. Less light is needed to achieve a desired exposure with a higher ISO, while more light is needed to achieve a desired exposure with a lower ISO. Shutter speed determines how long the shutter is open. The slower the shutter speed (i.e.: the longer the shutter remains open), the more light will reach the film or sensor for a given aperture. The faster the shutter speed, the less light will reach the film or sensor for a given aperture. Aperture determines the size of the hole (its area) through which light is permitted to pass on its way to the film or sensor. The larger the aperture, the more light is permitted to reach the film or sensor in a given unit of time. The smaller the aperture, the less light is permitted to reach the film or sensor in a given unit of time.

While these three variables apply equally to fim and digital, the ISO is essentially fixed (not realistically changeable) once you put a roll of film in your camera. However, indigital, you can easily change your ISO whenever you want. (This may not seem like a big deal, and many film photographers newly switched to digital do not quickly appreciate it, but it is one of the most significant advantages offered by digital.)

Generally, both with film or digital, the lower the ISO the greater the image quality. Image quality in film is generally determined by the size of the film’s grain (better images have finer grain), while image quality in digital is generally determined by the amount of noise present (better images have little or no noise). Film grain is typically the result of the film manufacturer increasing the film’s ISO (its sensitivity to light) by using larger silver salt particles (larger grains). Digital noise is typically the result of lower quality sensors that produce discolored (color noise) and/or improperly exposed (luminance noise) pixels. One of the advantages of better digital cameras is that they produce quality images at higher ISOs than do lower quality digital cameras. Indeed, many digital cameras today are capable of ISOs far in excess of films. For example, my Canon 1D Mark IV is capable of an ISO of 102,400, while the highest ISO of any film I ever used was 400! This means that the Mark IV’s sensor is 8 times more sensitive to light than film with an ISO of 400. Were I to take a shot with 400 ISO film at 1/50th of a second, the Mark IV could take the same image (without changing any other settings) at 1/12,800th of a second (faster than most cameras can operate). As a wildlife photographer higher ISOs give me a much greater ability to freeze a moving bird in flight. (True, outrageously high ISOs, even in high end pro cameras, are noisy; but their ISOs that are a stop or two lower often yield remarkably low noise images.)

Your shutter speed, especially if you are shooting wildlife, is often determined by your desire to freeze a moving subject. For example, without panning, a shutter speed of at least 1/2500th of a second is usually needed to freeze a flying bird. As noted in the last paragraph, it is easier to do this with a higher ISO than with a lower one. If the lighting requires a shutter speed of 1/125th of a second with a lens set at its widest opening to properly expose an image with an ISO of 100, it would require an ISO of approximately 2000 to achieve a shutter speed of 1/2500th of a second at the same aperture. This essentially excludes the use of film (unless you are willing to endure very large grain, very expensive film), and requires a digital camera of sufficient quality (i.e.: cost) to have a sensor capable of yielding low noise images at such an ISO.

Your aperture is often dictated by the depth of field you desire to achieve.

Depth of field refers to how much of your image is in focus. Remember, the larger the aperture, the greater the amount of light that will reach the film/sensor per unit of time. This is because a wider beam of light is permitted pass through your lens. Wide apertures soften your focus. The converse is also true: the smaller the aperture, the narrower the beam of light that reaches your film/sensor, and the crisper is your image. (There is a limit to this at very small apertures, but explaining that is for another time.) This is a fancy way of saying that the smaller your aperture the greater your depth of field, and the larger your aperture the less your depth of field.

Wildlife photographers need to balance a shallow depth of field so as to create a relatively clean background with a depth of field great enough to have the entirety of your subject in focus.


RAW vs JPEG: The Pros and Cons of Each

A basic question arising from digital photography is whether to shoot in RAW or JPEG format. As with much of photography, the is an inherently personal one. Having said that, I am unaware of any reasonably serious digital photographer who does not shoot RAW except on rare occasions.

RAW is an unprocessed format. In other words, the light entering your digital camera hits a sensor that converts the light into a series of numbers (digits – hence the name digital photography) based upon the intensity of the light. The typical sensor is setup in what is called a Bayer Pattern. This takes each square element in your camera’s sensor and divides it into four equal sections. Two of the sections, diagonal to each other, are sensitive to green light. One of the remaining is sensitive to red light, and the fourth is sensitive to blue light. The reason that twice as many sections are sensitive to green light as to either of the other primary colors of light is that the human eye is more sensitive to green than it is to red or blue. This is because red is at one end of the visible light spectrum while blue is at the other with green in between. As a result, extreme reds (infrared) and extreme blues (ultraviolet) are not visible to the unaided human eye while the full range of greens are visible.

(At this point some of you may be saying: “Hey, wait a minute. The three primary colors are red, blue, and yellow. Furthermore, you get green by mixing equal amounts of blue and yellow, so green cannot be a primary color.” While it is true that there are three primary colors, they are (in alphabetical order) blue, green, red, and yellow. No, I do not have trouble counting. Rather, red, green, and blue are the primary colors of light while red, blue, and yellow are the primary colors of pigment. Much more can be said on this subject, but that is a topic for another time.)

These unprocessed RAW data are then sent to the camera’s recording medium, a CF card for example, and ultimately downloaded to a computer for processing. The computer, via its imaging software (without which you cannot view digital images), places the data into color channels and then projects onto the computer’s monitor varying intensities of the three primary colors of light based upon the numbers in each channel at each picture element (each pixel). Your eyes then blend these signals into the color you see when looking at the monitor. You then process (develop, or enhance as some prefer to call it) each of the RAW images and save them in one or more of any of a number of file formats. These images (really, computer files) can be quite large and the processing can take a fair amount of time.

On the other hand, images captured in JPEG format are processed in the camera and thus almost invariably look better when loaded onto a computer. Many JPEG shooters do not further process their images, saving a significant amount of time. Also, the files are substantially smaller than the RAW files or those processed from the RAW files, often very much smaller.

Why, you may ask, would they do that given the extra time required to process the images as well as the extra space needed to store them?

The answer is simple, to create better images.

Unless your objective (and there’s nothing inherently wrong with this) is merely to make adequate (often marginally so) visual recordings of events, JPEG images are a waste of the money you spent on your camera.

Let me explain. Suppose you bought a Canon G10 point-and-shoot (one without an interchangeable lens) digital camera for $480.00. This is a 14.7 megapixel camera capable of shooting in both RAW and JPEG formats. Realize that the megapixel count is a measure of the amount data the camera is capable of recording. When you shoot in RAW with this camera, you get to use all of the data these nearly 15 million pixels record every time you click the shutter release.

When you shoot in JPEG, even at the highest quality possible, you end up with an 8-bit image. This means that the amount of data you are left with is 2 to the 8th power for each of the three channels. However, when you shoot in RAW, the camera uses all of the data it recorded, or 2 to the 14th power for each of the three channels. In other words, having spent nearly $500 for a camera capable of recording 4096 variations of light intensity per channel, you are now throwing away all but 256 variations; a loss of 7/8 of all of the data your camera recorded. Furthermore, you do not even get to choose which of those data are to be discarded.

I have had long-time film shooters tell me that I am somehow cheating when I process RAW images, and that I should at least shoot JPEG and then not process them. Unfortunately, such an assertion does little more than demonstrate that these film shooters believe that because he/she did not process that the image, that it was not processed at all. This is simply false. Someone programmed the camera, which is after all a specialized computer, to process the images based upon a mathematical formula, an algorithm, created by someone who has no idea what kind of images you would be shooting.

Another area of misunderstanding arises from the fact that many studies have shown that it is essentially impossible to see any difference in the quality of a printed 8-bit per channel photo from a 16-bit per channel one. Thus, some argue, the extra pixels are a waste of space and money. However, this position ignores the fact that RAW files need to be processed to reveal images that look the way the original scene did. This is because the process of converting the light hitting the sensor into digits necessarily results in a loss of image quality. As a result, the more data there is to work with, the greater the potential of creating a better image. However, once the image has been processed, there is no need to save it as a 16-bit per channel file. At that point, I convert all of my processed files to 8-bits per channel.

All right, you say, but what if you decide to rework the file at a later date?

That is why I NEVER discard the RAW file from which I processed the final image. In fact, I think of my RAW files as my negatives, and always save them at 16-bits per channel.

This is not to say that JPEGs serve no purpose. Indeed, anyone who tries to email images that are not JPEGs is likely to cause a great detail of frustration to everyone involved, including him/herself as the images will be so large as to take a very long time to transmit and are not likely to fit on the viewer’s screen. However, keep in mind that JPEGs are smaller than similar files because everytime (and I mean EVERYTIME) you close one, an algorithm over which you have little control discards part of the data. Likewise, when you open that file, an algorithm over which you have no control constructs data that it adds to the file in an attempt to compensate for the data previously discarded. This discard/construct process has nothing to do with whether or not you made any changes to the file. Again, this happens whenever you open or close a JPEG!

For the reasons discussed above, and because it is my intent to create the best images I can, I shoot in RAW format.





Page 1 2 3