backward-moving wheels--what's that called?

[nineiron]

Arrgh! Can someone remind me what the name is for the phenomenon that makes it look like wheels (for example) appear to be spinning backwards when viewed on TV? I know it was discussed here at one point, and I know it has to do with the strobing of light vs. the speed of the film.

Anyone remember? Thanks.

[brianmelendez]

Retrograde motion?

[beagledave]

Here..not sure if it has a name.

[Squink]

Beat frequency ?
Nyquist frequency ?

[Q.E.D.]

It's called the stroboscopic effect.

[RealityChuck]

It's a strobe effect. I don't think there's another name for it.

[William_Ashbless]

Its basically undersampling. It's a result of sampling a signal at a rate lower than the twice the frequency of the thing being sampled (this is called the Nyquist rate). I've also heard it called 'frequency foldover'.

In a nutshell what happens is that if you take a 'snapshot' just barely fast enough, you'll see the wheel 'advance', but if you take it more slowly, the wheel will turn far enough that it looks identical to what would happen if you were taking it 'just fast enough' with the wheel turning backwards.

That is, to the eye, the series of snapshots moving forward but taken at a certain rate just too slow is indistinguishable from a series of snapshots moving backward but taken at a certain rate just fast enough.

I suspect that since the seemingly backwards moving spokes require less 'rotation' to interpret, the brain is tricked into seeing them move backward.

What's interesting is that if you were to speed up a wagon wheel from stationary upwards towards infinity, but held the 'snapshot rate' steady, you'd see the wheel go forwards, then backwards, then forwards, then backwards, indefinitely alternating (assuming an instantaneous and perfectly clear snapshot).

[William_Ashbless]

Er, but I defer to QED for giving it a specific name.

[Desmostylus]

I think the term that Squink and William_Ashbless are looking for is aliasing.

[hobbes730]

I understand how this happens on TV or in films, but how does it happen in real life? i.e. when you look at the hubcap of a car and it appears to be going backwards. Your eye captures images analogously (sp?), not digitally, so the strobe effect shouldn't occur..

[raygirvan]

I second aliasing. I used to work in geophysics, where it crops up in digital sampling, and for the MSc course they always used to cite the 'backward wheels' effect as an example. Here is a reference.

[scr4]

I think aliasing is the most appropriate term here.

Something similar can happen in real life if a pattern in the foreground produces a periodic obstruction to a background pattern. For example, a regular fence or even a wheel cover can cause strange effects. I myself have seen a wheel appear to turn backwards when viewed through a bus window - it turns out that the window had some sort of plastic film with a regular pattern. But without such a foreground pattern, a wheel should not appear to turn backwards.

[Boyo Jim]

I don't think you see this effect in the daylight with an unaided eye. You see it on film, or if you see it live you see it under artificial lighting. Street lights will provide a nice substitute for a 60 hertz strobe.

[hawthorne]

Cecil: How come the wheels of a moving car appear to rotate backward sometimes?

[GusNSpot]

Pilots flying at night can sometimes see this effect if they have multiple strobes or strobes with certain flash rates. If they look at it wrong they also get to try flying while disorientated. That is a lot of fun also. (Ask JFK Jr. about disorientation. It can be done but needs to be practiced.) Helicopters are worse about this due to the overhead slower turning blades as compared to propellers.

[Padeye]

To avoid crossposting http://boards.straightdope.com/sdmb/...hreadid=159083

[Q.E.D.]

I suppose either term would work, however in film, the term stroboscopic effect is used almost exclusively. Aliasing refers more broadly to any data sampling rate errors.

[mrcrow]

movie frames run at the speed to suit the human eye
so strobe effect must be ok if the eye has a 'frame reference' by which images are recorded and is not a seamless transfer of photons via the optic nerve
i go for stroboscopic effect
and you can see this in daylight
spin a bike wheel no lights just your eyes
or watch the fan in a jet engine...on the ground!!

[Popup]

Originally posted by mrcrow
and you can see this in daylight
spin a bike wheel no lights just your eyes
or watch the fan in a jet engine...on the ground!!

That statement is just as false as the grammar and punctuation would lead you to believe.
It is generally taken for granted that in constant light there is no such effect.
(mrcrow might however be correct in stating that you can see it in a jet engine, but that is only because different parts of the engine rotate at different speeds, and thus act as a strobooscope, letting through light at certain intervals. But then we are talking about flickering light again.)

[Q.E.D.]

Originally posted by Popup
That statement is just as false as the grammar and punctuation would lead you to believe.
It is generally taken for granted that in constant light there is no such effect.
(mrcrow might however be correct in stating that you can see it in a jet engine, but that is only because different parts of the engine rotate at different speeds, and thus act as a strobooscope, letting through light at certain intervals. But then we are talking about flickering light again.)

All the compressor and turbine blades are firmly attached to the same shaft which runs the length of the engine. All the parts spin at precisely the same rate.

[Padeye]

The effect is also observable with propellors in sunlight, much easier I think than with the fan on a jet. Turboprops are best as they gradually build up speed from zero much more evenly than a piston engine.

[bbeaty]

The effect is also observable with propellors in sunlight, much easier I think than with the fan on a jet. Turboprops are best as they gradually build up speed from zero much more evenly than a piston engine.

I've looked carefully at propellors in sunlight. Nothing! No strobe effect. There is however an unmoving "shadow" in the blurred disk of the moving propellor. This exists because the propellor blade reflects ambient light differently in each part of it's rotation. (Sometimes the shadow is a bright stripe if the blade catches the sun just right.) But this "shadow" doesn't move forward or backward constantly. It DOES appear to move if your viewing angle changes, for example as the plane moves by you, or as it turns while taxiing.

Also, I've seen highlights in hubcaps which appear to move. Again, this is the same "shadow" as in the propellor blades above. It doesn't move if you drive alongside the moving wheel. It does appear to move if the wheel is going past you (if your viewing angle is changing.)

So, next time you see backwards-moving wagon wheels and aircraft propellors in your environment, it obviously means that you're TRAPPED INSIDE A MOVIE!!! Therefore you can do ANYTHING you want, especially if you're inside a Hollywood style flick. Try leaping from a tall building and see if you automatically land in a passing garbage truck or canvas awning. Try firing your revolver and see if it contains an unlimited number of bullets rather than only six!

[William_Ashbless]

Unless, as Cecil points out there's a light source that, itself, is periodic, nearby. Like a flourescent lamp.

[Popup]

Originally posted by bbeaty
I've looked carefully at propellors in sunlight. Nothing! No strobe effect.

Thank you for confirming this bbeaty. It kept me awake most of the night, and in the end I came to the conclusion that the only times I have seen the retrograde rotatioon in propellers was when flying in the late afternoon, when a lot of the airport is bathed in 100Hz light.

For a long time I had this wacky idea that the finite speed of light would give rise to different patters at higher RPMs (A bit like Michelsons measurement of the speed of light). However, that effect would be negligible at normal speed. In the end I'd rather deny the effect altogether!