Car photography
I took a few pictures of my car right when I got it, but they didn't turn out as well as I'd hoped. Also, since then I've slipped down the waxy slope of learning a thing or two about auto detailing. Since I work for 3M, I bought a handful of their fine products from the company store.
Since I'm an engineer, I got to thinking about what makes things shiny. For example, if you keep using finer and finer grit sandpaper (and eventually polishing compound), when does a rough surface become optically smooth? The surface roughness really just needs to be much finer than the wavelength of light, according to the wave theory of light. Light is small, but it's not that small; a human hair is only about 200 wavelengths in diameter. From a quantum perspective, each photon hitting a nanometer-scale scratch can be thought to try bouncing off every point in a neighborhood of the scratch and constructively interfering with itself. Feynman has a great discussion of this sort of thing in QED.
Also, what's the limit to how shiny you can polish something? Well, for dielectric (nonconducting) materials such as paint and wax, the amount of light reflected from a smooth surface is given by the Fresnel (Fra-Nell) equations and is limited by the index of refraction of the material. (This is why diamonds and leaded crystal are so shiny.) Wax and plastic and glass all have a refractive index in the neighborhood of 1.5. As such, you won't get more than about 4% of light bounced back at you if the sun is at your back.
Fortunately, there's the GIMP to magically change my car wax from n ≈ 1.5 to something much more. For example, one of these two is streight out of the camera:
A few more, all low-pass filtered by subtracting a Gaussian-blurred version (σ=200px) and then adjusting brightness and removing some artifacts by hand.
Since I'm an engineer, I got to thinking about what makes things shiny. For example, if you keep using finer and finer grit sandpaper (and eventually polishing compound), when does a rough surface become optically smooth? The surface roughness really just needs to be much finer than the wavelength of light, according to the wave theory of light. Light is small, but it's not that small; a human hair is only about 200 wavelengths in diameter. From a quantum perspective, each photon hitting a nanometer-scale scratch can be thought to try bouncing off every point in a neighborhood of the scratch and constructively interfering with itself. Feynman has a great discussion of this sort of thing in QED.
Also, what's the limit to how shiny you can polish something? Well, for dielectric (nonconducting) materials such as paint and wax, the amount of light reflected from a smooth surface is given by the Fresnel (Fra-Nell) equations and is limited by the index of refraction of the material. (This is why diamonds and leaded crystal are so shiny.) Wax and plastic and glass all have a refractive index in the neighborhood of 1.5. As such, you won't get more than about 4% of light bounced back at you if the sun is at your back.
Fortunately, there's the GIMP to magically change my car wax from n ≈ 1.5 to something much more. For example, one of these two is streight out of the camera:
A few more, all low-pass filtered by subtracting a Gaussian-blurred version (σ=200px) and then adjusting brightness and removing some artifacts by hand.