An amazing experiment

I've learned an amazing experiment on the morguefile website. The results are shown on the side here. If you don't think the pictures look spectacular, read on - you'll change your mind after reading!

The two pictures aren't sharp. First of all, it isn't necessary for what I'm trying to show, and second of all, the camera wasn't able to focus on the 'subjects' anyway.

What are the subjects? Two sheets - one white, one black. You can tell them apart because the black one is slightly darker - but not as much as you'd expect! If you take your time to setup the experiment (I shot the pictures with different textures, and under bad light), you'll end up with two identical pictures. What happened?

Keep in mind that the metering of the camera has no clue what the light conditions outside the camera are. All it knows is what it sees through the lens. The camera than has to make a decision on how to choose aperture and shutter speed to get a correct exposure. And that's where things 'go wrong'. The camera will try to expose the image in such a way that the median shows 50% gray. So, for a monochromatic image - wether it is white or black - the camera will adjust so that it will show at a gray level of 50%. There's an important lesson to be learned: automatic metering aims for average exposures. Whenever the subject is not average, compensate - or pay the price!