I was reading about movie sound the other day and was really interested to find out how pretty much every single bit of space, including space around the sprocket holes, is used to encode the sound in all the various formats. The analog encoding is of course the simplest–just a continuous band that lets through a specific amount of light representing the waveform.
But then I remembered that projectors don’t move the film continuously. They instead pull it forward in little one-frame-at-a-time jerks. It pulls the film forward once, the given frame flashes once or twice on the screen, and then as the shutter is blacking out the image, the frame is advanced again. The key point is that although the film is moving very quickly through the projector, it’s not constantly moving. In fact, it seems like the majority of the time, it’s sitting still.
If analog audio is as simple as it seems to be (and I’m guessing that it’s not) then shouldn’t we just hear little bursts of audio while the film is advancing, and nothing when it’s not?
Apologies in advance if this is a trivial question already answered here before–I did a quick search and didn’t find anything.
The audio pickup is not at the lens. It’s about eighteen frames - three-fourths of a second - away. The film runs smoothly past that point. However, editors have to be careful to adjust for this when they splice broken footage together. That’s why old films sometimes have odd cuts in sound.
I think it’s the case that analogue sound is not recorded on the film exactly alongside the frame to which it pertains, allowing the projected frames to stop-start through the shutter (or whatever the apparatus is properly called), but for the sound to be read continuously as the film travels smoothly through a feeder mechanism (with a bit of slack somewhere to act as a physical buffer to the stop-starting of the projected frames).
Actually, it’s probably a lot more complex than that in real life, as I seem to recall that the film doesn’t actually stop and start through the shutter, but is formed into a pattern of travelling ripples, so that each ripple drops a new frame into the right place in the projector.
Wiki explains it. In short, the sound head gets the film coming by continuously, because there is a constant feed at that later point. This is possible because there is a space where the film is free between the shutter feed and the sound head.
I can follow your reasoning only up to a point. It sounds like one of Zeno’s paradoxes to me. That one about the arrow never getting anywhere since at every instant it is standing still, or the tortoise always being ahead of Achilles (or is it the hare?).
I believe (key word) that experience covers the “jerky” movement quite well. Since sound is produced by vibrations, the very nature of a vibrating medium implies some position of zero-movement, demonstrated visually on an oscilloscope when the waveform crosses the x-axis and reverses direction. Precisely at the x-axis it would be motionless, at that exact instant.
I suspect that if you analyzed the movement of a violin’s bow with stop-motion cinematography, you’d find that it’s also moving in some “jerky” fashion. But we tend to hear the movement as smooth and uninterrupted. Bonus question: does the resin help to smooth out the stroke or to accentuate the jerks?
The jerkiness of the film stopping and starting in the gate is something like 25 ‘jerks’ per second; perhaps higher in some kinds of film projection, but probably not exceeding 100. That’s far too low frequency to act as a carrier for audible frequencies.
24 frames per second normally, according to the Wiki article; sometimes double that.
Zeldar, the Wiki article is pretty clear on how it works. Basically, while the feed across the lamp is “jerky,” the feed across the sound head is not. That is, the sprockets that move the film across the lamp move the film in a stop-start fashion, but the sprockets that move the film across the sound heads do so in a continuous, steady stream fashion. Because they are running at the same rate, and because there is a space between the two mechanisms where the film is left free, the jerkiness of the one mechanism does not affect the smoothness of the other.
Thanks for the additional info, and I have yet to read the Wiki article. My main point was/is that even analog devices tend to have some “digital” aspects if the proper method of breaking down their movements is used. High-speed photography of natural events like waterfalls, windstorms, cloud formation, etc., can assist in detecting those “moments of change.”