Noncanonicity of volume controls
Jan. 1st, 2011 04:43 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
sniffnoyOK, I think I may have finally figured out the mystery of volume controls.
I'm sure the answer to this question is well-known if I ask the right people, but whoever they might be, I never seem to have asked them before. Either that, or I didn't understand the explanation when I heard it.
The question is: Why do volume controls have no "as recorded" setting?
Consider - why do we have volume controls? So we can adjust volume up and down, duh. But... what if we want don't want to hear it quieter or louder, but rather want to hear it exactly as it was recorded? You would think volume controls would have indicated where the zero setting, the "as recorded" setting is, right? But they don't. It's as if no such thing even existed, as if the concept made no sense.
It certainly seems that it should make sense. In the real world, a sound has a specific volume, just as much as hit has a specific frequency. Imagine if speakers had frequency controls, to make things higher or lower, with no indication of where the "as recorded" setting was! Nonetheless we would all agree such a setting existed, though unmarked.
You might say, well, volume depends on where you're standing. But this argument fails. It has to - after all, frequency depends on how fast you're moving! It fails because I'm talking about an "as recorded" setting, not an "as played" setting. So if the original recording device was 10 feet away from the original sound, and I stand 10 feet away from the speaker doing the playback, it should sound as loud to me as if I were 20 feet away from the original source of the sound. (Up to the abilities of the recorder/speaker to replicate this, of course.)
But I think I've figured it out. Consider the units on these things. Frequency, of sounds or of electrical signals, is measured in units of time. There's a canonical isomorphism between acoustic-time and electrical-time, because they're both just, well, time. But amplitude, well, you've got acoustic-amplitude and you've got electrical-amplitude, and those don't have the same units on them. More importantly, the conversion between them is going to depend on the exact recording (or playback) hardware you're using: The strength of the magnets, etc. Microphones and speakers yield an isomorphism between sounds and electrical signals, but since there's no canonical setting for the strength of the magnets and other such things, so there's not a canonical isomorphism between acoustic-amplitude and electrical-amplitude. Hence, if you want to hear sound at the volume it was recorded, you have to set the isomorphism applied by your speakers to match the isomorphism performed by the device that originally recorded it; since you have no way of knowing what that might have been, you have no way of recovering the original volume.
I'm think the same argument works for phonographs, though I'm less certain. (Is the statement true for phonographs?)
Of course, this forces the question of why, with modern electronic recording devices, we don't have microphones tag what they record with some metadata specifying what volume to set things to if you want to hear it at the recorded volume. Admittedly, as it seems there's no standardization between volume controls currently, such a thing would be kind of pointless, but...
-Harry
I'm sure the answer to this question is well-known if I ask the right people, but whoever they might be, I never seem to have asked them before. Either that, or I didn't understand the explanation when I heard it.
The question is: Why do volume controls have no "as recorded" setting?
Consider - why do we have volume controls? So we can adjust volume up and down, duh. But... what if we want don't want to hear it quieter or louder, but rather want to hear it exactly as it was recorded? You would think volume controls would have indicated where the zero setting, the "as recorded" setting is, right? But they don't. It's as if no such thing even existed, as if the concept made no sense.
It certainly seems that it should make sense. In the real world, a sound has a specific volume, just as much as hit has a specific frequency. Imagine if speakers had frequency controls, to make things higher or lower, with no indication of where the "as recorded" setting was! Nonetheless we would all agree such a setting existed, though unmarked.
You might say, well, volume depends on where you're standing. But this argument fails. It has to - after all, frequency depends on how fast you're moving! It fails because I'm talking about an "as recorded" setting, not an "as played" setting. So if the original recording device was 10 feet away from the original sound, and I stand 10 feet away from the speaker doing the playback, it should sound as loud to me as if I were 20 feet away from the original source of the sound. (Up to the abilities of the recorder/speaker to replicate this, of course.)
But I think I've figured it out. Consider the units on these things. Frequency, of sounds or of electrical signals, is measured in units of time. There's a canonical isomorphism between acoustic-time and electrical-time, because they're both just, well, time. But amplitude, well, you've got acoustic-amplitude and you've got electrical-amplitude, and those don't have the same units on them. More importantly, the conversion between them is going to depend on the exact recording (or playback) hardware you're using: The strength of the magnets, etc. Microphones and speakers yield an isomorphism between sounds and electrical signals, but since there's no canonical setting for the strength of the magnets and other such things, so there's not a canonical isomorphism between acoustic-amplitude and electrical-amplitude. Hence, if you want to hear sound at the volume it was recorded, you have to set the isomorphism applied by your speakers to match the isomorphism performed by the device that originally recorded it; since you have no way of knowing what that might have been, you have no way of recovering the original volume.
I'm think the same argument works for phonographs, though I'm less certain. (Is the statement true for phonographs?)
Of course, this forces the question of why, with modern electronic recording devices, we don't have microphones tag what they record with some metadata specifying what volume to set things to if you want to hear it at the recorded volume. Admittedly, as it seems there's no standardization between volume controls currently, such a thing would be kind of pointless, but...
-Harry
no subject
Date: 2011-01-03 02:56 am (UTC)no subject
Date: 2011-01-03 03:41 am (UTC)Well, even without that being invariant it still seems to be physically meaningful, even if not at all useful. I didn't anticipate it being so anyway.