The colour wheel on the right initially spins slowly. With the upper stepper you can change the delay between updates (in frames, preset to one frame, typically at 60 Hz). With the lower stepper you can adjust the increment in rotation angle, preset to 5°.
Interesting things appear at ∆=60°, and more clearly at 120°: There the disk appears nearly gray, because each sector is rapidly alternating between the three colours (whose mixture results in gray with the preset colours). If you go 5° faster, a coloured “propeller” rotates rightwards, going 5° lets that propeller rotate backwards. Around ∆=360° the disk seems stationary again, because a full rotation between changes nothing.
Slowing down the time scale by increasing the frame delay (to, say, 20) makes it more obvious what’s happenting.
Motion demonstrations on a computer screen, especially when fast motion is involved, suffer from stroboscopic artifacts. A grander designation would be “temporal aliasing”. So I decided to make a teaching point out of a nuisance – my original goal was to demonstrate additive colour mixture, not a success…
The disk does not rotate smoothly (even if it might look so) but rather is presented in rapidly succeeding still frames, each with a different rotation angle. This produces the percept of smooth motion (within certain limits), the “Phi Phenomenon” of Wertheimer. The current demo is set up with a frame rate of 120 Hz and, initially, an angle increment ∆ of 7° every 20 ms. However, here in your browser another aspect comes into play: the frame rate of your display. Many LCDs have a refresh rate of 60 Hz, CRTs come in rates from 60 to 200 Hz. So what you see exactly depends on the interaction of the two frame rates and the angle increment, and the intermediate effects may not be pretty ;-). Under some conditions you see a slow backwards rotation. This is well known as the “wagon wheel effect”.
Bach M, Meigen T, Strasburger H (1997) Raster-scan cathode ray tubes for vision research – limits of resolution in space, time and intensity, and some solutions. Spatial Vision 10:403–414