As a photographer, I work with light. Forgetting all of the fancy camera ads we’re all bombarded with, it all comes down to light – where it comes from in a scene, its intensity, color quality, reflections and so on.
As such, I compulsively pay attention to lighting when I enter a new room and almost always start looking at the ceiling to mentally calculate the correct exposure of a typical reception where flash is bounced off of (hopefully white) it. Sad isn’t it? Typically white is always good, and high ceilings much easier to work with than low ones.
On the other hand, I can see the effect of different types of lighting in a room without even turning on the lights, so I see it as a plus. Now if I can only stop doing it all of the time and just when I’m actually working on a project.
Here’s your homework if you’re interested. The next time you see a movie at the theater pay attention to the lighting in the film. You’ll quickly understand what lighting is romantic, dramatic, dull, edgy and so on and why the director wanted that effect in that scene. You may even start looking at ceilings when you enter rooms to see if they are white and how high. Sorry. 
I use Speedlights a lot, although I prefer studio flash, but as Speedlights weigh 1lb and studio flash 10, I tend to carry Speedlights. So much, you’d think that I really understand how they work to automatically set the exposure. I don’t. Here are my observations. The Nikon (and most likely, the others brands) have an electronic brain that detects the light bouncing off of the subject and then attempts to calculate how much flash to provide to properly light a scene. Unfortunately, from my experience they don’t work half the time and for no specific reason that I’ve been able to observe.
Let’s start at the beginning. Way back when, you’d but a Speedlight and note its full power GN or guide number, based on the ISO or ASA speed of the film you use. For example, you would use Tri-X which is rated at 400 ISO or ASA. The guide number of the Speedlight would be (fore example) 125 which required some simple calculations to get the correct exposure. If your subject in this example was 40 feet away you had to shoot at about f/2.8 to get the correct exposure. If 20 feet, the f/5.6. You just divide the guide number by the subject distance to get the basic aperture setting. In other words, divide the GN by the distance to find the approximate aperture setting.
I’m assuming that the flash is the only light source, so if you wanted lighten the shadows in daylight, you’d stop down (reduce the aperture) by a few stops to get the correct lighting balance. In the above illustration, to f/5.6 and f/11 respectively. Of course, you would need to modify the exposure based on experience.
Things got more complicated when the flash manufacturers developed methods to change the light output on demand so the GN would vary, and complicate your life. Eventually the GN concept went away in favor of automatic flashes… the only problem being that they don’t work very well.
Broadly speaking, just as you press the shutter release the flash fires a pre-flash (with a sensor) to gauge the subject distance and the surroundings to determine the correct flash output depending on the lens aperture. Sometimes it works well, and sometimes it doesn’t.
What I don’t know is why the flash tries to read the subject distance from the light bounce when it could read it from the focusing mechanism which should tell it how to set the GN. In this example, you’d point the camera at the subject 11 feet away and press the shutter release, the camera should tell the flash that the subject is 11 feet away, the ISO rating is X, the main subject is Y and the aperture is Z and set the corresponding amount of light output. Instead it uses its pre-flash to judge the distance. I just don’t know why.
In any instance, the Nikon Speedlights tend to overexposure the main subject, completely washing them out. Some time ago, I went back to manual flash exposure using either 1/2 or full power and having memorized the GN for each setting. It works fine almost every time.
I happened to visit dxomark.com and was surprised.
The DXO company produces a remarkably good software package for professional photographers that squeezes every bit of quality from the digital negatives, known as RAW files. If your wedding photographer says that he or she only shoots in jpg, doesn’t hire him or her. Jog image quality is severely compromised as the camera converts the image from the original RAW image it produces and saves the photography as a much smaller file, a jpg. Modern PCs are infinitely more powerful in their ability to squeeze out quality, and DXO is one of the best when they do the conversion and not the camera. Back to the story…
The DXOmark website shows the technical variables that go into the quality of the RAW files that each “pro” camera can produce, and the DXO Mark is a simple scale for the bottom line result. Here’s the surprise. The standard wedding photographers camera is in many instances the Nikon D300. It came out a few years ago and costs about $1,600 for just the body, without a lens. The new Nikon D5000 came out a few months ago, and its body sells for as little as $500 or about a third of the cost of the D300.
So you’d think that the image quality of the D300 will be up to three times better than the D5000 as they have the same 12.3 megapixel sensor size? Wrong, the D5000 is nearly 10% better at a quarter of the cost! Technology marches on, and very quickly. I know that the D300 has a better build and is more “heavy duty” (whatever that means to wedding photographers) than the D5000, but the D5000 has features that the D300 doesn’t have either.
So what’s the moral of the story? It’s that newly released cameras offer far better price/performance than the “old” cameras when comparing image quality versus cost. I’m not saying that the D300 is junk, but justifying its cost is getting more and more difficult even though it produces very high quality images.
