We had the first real snowfall of the winter today. My dog, who is only 15 months old, probably doesn’t remember last year’s snow, so this fall came as a whole new experience.
During our morning walk, it started to come down more heavily, with bigger flakes, and that set the dog off on a mission to catch as many as he could.
These photos pose a particular problem: black dog, white snow. This is an example of an issue that faces all photographers at some point—when the scene does not match the light meter’s simplistic assumptions.
Most of the time, a scene is composed of things with a range of brightnesses. Most objects will be in the middle of this range, and those things at the extremes will be less important, so can be left to go a bit dark (don’t let bright things go pure white, though; that’s frequently a lot worse than letting dark things go black. Bright lights and specular highlights are among the few things that you can afford to let go to pure white).
Here, though, I’m photographing a black dog on a white snow background. There are no mid-brightness objects in the scene, only things at the extremes of the brightness range. The snow dominates, so the camera’s exposure meter under-exposes on the assumption that the scene has an overall average brightness. For all of these photos, I have overridden the camera’s exposure by at least one stop of overexposure to get the snow looking white, rather than mid-grey.
Now, you can either tell the camera to add one stop to whatever it thinks the exposure should be, or you can boost the exposure later in Lightroom†.
Mostly this doesn’t matter, except that doing it in camera alters the shutter/aperture setting, which can have consequences for the image. Note that these pictures contain falling snow. I had the camera on aperture priority (so it selected the shutter speed), and +1 stop exposure compensation. That means that it slowed down the shutter to increase the exposure, so the snowflake trails lengthened, and the dog was just a blob of motion blur. Can you spot the point where I boosted the ISO setting to compensate for this?
It’s also worth noting that, even with the exposure boost, the dog still came out as a black, featureless blob, so I needed to selectively boost him in post-processing as well, to bring out some detail in his coat and face.
This is a situation where you simply can’t get it entirely correct in camera.
† Other (inferior) image processing applications are available.