hermanv said:
Bass--particularly drums--will have more impact when the phase is correct.
KSTR said:
This is such a mess I'm not even sure where to begin.
Where did you get the idea that 'most drum types' don't have an identifiable polarity, and why would some have an identifiable polarity and others not, given that all drums work the same way, i.e. a stretched membrane is set in motion by a strike of some sort. At that point the drum head, be it animal skin, paper, or plastic, will begin a complex damped series of vibrations. The initial transit of the drum head when it is struck will not be retraced to the same extent on the return. As a result, even within the first full cycle of the fundamental, it is already asymmetrical. For it to be otherwise would violate the laws of thermodynamics as there is no further input of energy.
Unless you wish to posit a perpetual motion machine...
'Electrical instruments of any sort surely don't have a correct phase.'
Oh, dear.
1) Any stringed electric instrument functions precisely the same way its acoustic counterpart does. I play electric bass. When I pluck a string with a finger and release it, it stretches, then begins a series of damped resonant oscillations in exactly the same way that the strings on an upright acoustic bass do. You can study the way the body, neck, bridge, etc. modify that vibration, but that's all they do...modify. The string sets the phase. To attempt to set electric instruments apart from acoustic ones isn't going to get you anywhere.
2) Your fallback position is--or should be--the synthesizer. True, you can make a sine wave with a synthesizer, and no one would be able to tell the difference, but sine waves are not the only things synthesizers can do. They can also imitate pianos. Do you claim that pianos have symmetrical waveforms?
Your paragraph regarding recording drums is completely wonky and you left out your one and only possible valid point, which is that the listener might find himself faced with a record where the drums are recorded correctly, but the bass might be inverted, due to any number of devices inserted into the recording chain, say a flanger.
This isn't an argument against absolute phase, it's an argument for simpler recordings, or at the very minimum more attention to detail on the part of those who design amps, effects pedals, mixing boards, and so forth.
Please, stop and think this through.
For what it's worth, I don't regard absolute polarity as the most important thing in the universe. It isn't, as has been noted, an in-your-face sort of thing. It is, however, an attention-to-detail thing. If I worked in a restaurant, I could take a recipe for a cake and start changing the amount of sugar I put in. Would anyone notice if I took out one pinch if the total amount was a cup? No. Would they notice if I took out two pinches? No. Would they notice if I took out an ounce? I submit that anyone paying attention would notice a difference from yesterday's cake if I took out 1/8 of the total amount of sugar. True, some wouldn't notice. But to take that as an excuse to do so, whether because I was a sloppy baker or because I was trying to save money, would matter to at least some of my patrons. Saying "some people can't taste the difference" wouldn't carry any weight with a loyal customer who asked what was wrong with the cake. Nor should it.
Audio is the same way. You should always pay attention to the details. Not that any one detail will necessarily stand out, but the sum of ten or twenty little things adds up to a much bigger thing. It's synergistic, in other words.
Grey
Gents,
Polarity testing/proving using music is filled with hidden dangers, and often cannot be easily shown.
1/ The speakers may have wildly varying phase response as previously mentioned from reversed drivers and crossovers designed only for best measured flatness.
2. What do you use for the music? Unless it is a live recording with very simple miking and no later eq-ing, all sorts of phase differences may already be encoded on the recording.
As an extreme example, many multitrack studio recordings are impossible to use as the vocals may have been recorded in LA and the band in NY, and hence can be out of absolute phase to each other.
Absolute phase/polarity first came to my notice when, having read about it, I sold a basic system (but with time coherent speakers) to a Symphony musician and told him to try reversing the speaker wires on both speakers to see if there was one way he preferred. He reported back a very clear preference on his first test record, but his hifi buff friend couldn't hear a thing!
Like JC, my big differential (tube) preamp has had a front panel phase switch since 1980, and I am now addicted to it But I also use full range elctrostats, which tend to show up such differences..
Regards, Allen
Polarity testing/proving using music is filled with hidden dangers, and often cannot be easily shown.
1/ The speakers may have wildly varying phase response as previously mentioned from reversed drivers and crossovers designed only for best measured flatness.
2. What do you use for the music? Unless it is a live recording with very simple miking and no later eq-ing, all sorts of phase differences may already be encoded on the recording.
As an extreme example, many multitrack studio recordings are impossible to use as the vocals may have been recorded in LA and the band in NY, and hence can be out of absolute phase to each other.
Absolute phase/polarity first came to my notice when, having read about it, I sold a basic system (but with time coherent speakers) to a Symphony musician and told him to try reversing the speaker wires on both speakers to see if there was one way he preferred. He reported back a very clear preference on his first test record, but his hifi buff friend couldn't hear a thing!
Like JC, my big differential (tube) preamp has had a front panel phase switch since 1980, and I am now addicted to it But I also use full range elctrostats, which tend to show up such differences..
Regards, Allen
Grey,
Oh, man, it gets boring. Yet another time when you competently put things into my mouth which haven't been said.
You are talking about asymmetry, which I never questioned. Asymmetry is everywhere in practically any musical signal. The question of "correct" polarity (I'm talking about polarity and not assymetry) is now which half (referenced to zero) of this asymmetrical waveform we declare to be positive. This decision is, for many instruments, arbitrary by virtue of the way these instrument produce sound and by the way how these sounds reach our ears or the microphone. For other instruments we can declare the correct polarity of a playback system, being the same pressure change polarity (at the eardrum) that we would have when listening to these sources in reality.
- Klaus (active musician, too. Plus I've spent considerable time in recording studios, on both sides of the control room window).
Oh, man, it gets boring. Yet another time when you competently put things into my mouth which haven't been said.
You are talking about asymmetry, which I never questioned. Asymmetry is everywhere in practically any musical signal. The question of "correct" polarity (I'm talking about polarity and not assymetry) is now which half (referenced to zero) of this asymmetrical waveform we declare to be positive. This decision is, for many instruments, arbitrary by virtue of the way these instrument produce sound and by the way how these sounds reach our ears or the microphone. For other instruments we can declare the correct polarity of a playback system, being the same pressure change polarity (at the eardrum) that we would have when listening to these sources in reality.
- Klaus (active musician, too. Plus I've spent considerable time in recording studios, on both sides of the control room window).
KSTR said:
many instruments begin producing sound with a positive overpressure.
If the end system produces a negative/underpressure, it's audible.
john curl said:
That's a shame many friends at MIT worshiped him, and actually understood what he was saying (not me).
Klaus,
If it's asymmetrical, it matters. The positive and negative halves of the sound won't sum to zero, at which point you're losing information that was there when the instrument was played. It made it onto the recording. You have it within your power to make it right on playback. Whether you choose to do so or not is up to you.
Claiming, after the fact, that I'm 'putting words in your mouth' won't work. What you wrote is there for anyone to read.
Noting that you're a musician doesn't necessarily mean that you listen carefully or that you've thought everything through. Claiming experience in recording is almost damning, in that most of the folks involved in recording music wouldn't know decent sound quality if it bit them on the behumpus. Want proof? We're hip deep in recordings that sound perfectly awful. Pick one. Ask yourself how it could all go so dreadfully wrong if those involved in the process gave a rip about the sound quality. The answer...they don't.
Every once in a while, you get someone who's prone to arguing that the compression/rarefaction must sum to zero. This statement is often accompanied by the observation that the woofer cone returns to its rest position, ipso facto since the woofer cone has returned to rest (i.e. zero), then what came before must also. And therefore (their logic goes) the only way you can hear sound is if it sums to zero.
Not true.
A simple demonstration of this is to place a 1.5V battery across the driver terminals. If you hook the positive terminal of the battery to the positive terminal of the speaker and negative to negative then the woofer cone will lurch forward. On disconnecting it, the cone will return to its rest position. What it will not do is swing to an equally negative position before returning to rest position. That's a near-perfect demonstration of a positive-only asymmetrical waveform. And yes, it's audible, even though you will sometimes run into people who argue that it can't be so.
Nelson Pass goes to great lengths to argue that the very asymmetry of the air itself comes into play in that if you drop from 14.7psi to 0, that's all the negative swing you're going to get. On the other hand, you can theoretically increase pressure to infinity (and beyond...sorry, couldn't resist) and still have a clean positive waveform.
Scott,
Brass instruments are the classic demonstration of oddball waveforms because they break every rule in the book. Taking what I was saying earlier about wide bandwidth into account, brass instruments frequently exhibit steeply rising waveforms, so they're good test material for wide bandwidth circuits.
Grey
If it's asymmetrical, it matters. The positive and negative halves of the sound won't sum to zero, at which point you're losing information that was there when the instrument was played. It made it onto the recording. You have it within your power to make it right on playback. Whether you choose to do so or not is up to you.
Claiming, after the fact, that I'm 'putting words in your mouth' won't work. What you wrote is there for anyone to read.
Noting that you're a musician doesn't necessarily mean that you listen carefully or that you've thought everything through. Claiming experience in recording is almost damning, in that most of the folks involved in recording music wouldn't know decent sound quality if it bit them on the behumpus. Want proof? We're hip deep in recordings that sound perfectly awful. Pick one. Ask yourself how it could all go so dreadfully wrong if those involved in the process gave a rip about the sound quality. The answer...they don't.
Every once in a while, you get someone who's prone to arguing that the compression/rarefaction must sum to zero. This statement is often accompanied by the observation that the woofer cone returns to its rest position, ipso facto since the woofer cone has returned to rest (i.e. zero), then what came before must also. And therefore (their logic goes) the only way you can hear sound is if it sums to zero.
Not true.
A simple demonstration of this is to place a 1.5V battery across the driver terminals. If you hook the positive terminal of the battery to the positive terminal of the speaker and negative to negative then the woofer cone will lurch forward. On disconnecting it, the cone will return to its rest position. What it will not do is swing to an equally negative position before returning to rest position. That's a near-perfect demonstration of a positive-only asymmetrical waveform. And yes, it's audible, even though you will sometimes run into people who argue that it can't be so.
Nelson Pass goes to great lengths to argue that the very asymmetry of the air itself comes into play in that if you drop from 14.7psi to 0, that's all the negative swing you're going to get. On the other hand, you can theoretically increase pressure to infinity (and beyond...sorry, couldn't resist) and still have a clean positive waveform.
Scott,
Brass instruments are the classic demonstration of oddball waveforms because they break every rule in the book. Taking what I was saying earlier about wide bandwidth into account, brass instruments frequently exhibit steeply rising waveforms, so they're good test material for wide bandwidth circuits.
Grey
GRollins said:
No problem with that, it's my own personal "anyone can hear it test". The massed horns during Reiner plays Wagner, the guys that made the CD got it all WRONG!
Wanna bet?
Take out a mike and do your experiment. Whaddaya know, when you disconnect the battery, the SPL waveform is negative. Huh!
Grey,
I'll rest my case, again the things you bring up seem out of context, IHMO...
All I said was this (boiling it down to the essentials):
1) there is no "correct" absolute polarity with a drum. Kick drums just happen to be recorded with the initial attack being positive (assuming a polarity preserving recording chain), while with floor toms it's the opposite way round (unless the tom has only a batter head and is mic'ed from "down under inside" -- not very common today). The "correct" polarity for these kind of instruments is that one which sounds better, this is a completely valid personal choice. Other instrument (yes, brass as a good example) have correct, or better natural, absolute phase, no doubt about that.
2) *I* could, so far, not positively test absolute polarity with *my* sources, *my* system and the ones I had access to, plus *my* ears (hope you get the "I-message"). I do *not* claim the phenomenon doesn't exist (it does, I know).
- Klaus
I'll rest my case, again the things you bring up seem out of context, IHMO...
All I said was this (boiling it down to the essentials):
1) there is no "correct" absolute polarity with a drum. Kick drums just happen to be recorded with the initial attack being positive (assuming a polarity preserving recording chain), while with floor toms it's the opposite way round (unless the tom has only a batter head and is mic'ed from "down under inside" -- not very common today). The "correct" polarity for these kind of instruments is that one which sounds better, this is a completely valid personal choice. Other instrument (yes, brass as a good example) have correct, or better natural, absolute phase, no doubt about that.
2) *I* could, so far, not positively test absolute polarity with *my* sources, *my* system and the ones I had access to, plus *my* ears (hope you get the "I-message"). I do *not* claim the phenomenon doesn't exist (it does, I know).
- Klaus
Scott, I have many of his books, but they are not easy reads. I can believe that if you lived physics, day and night for years, you could easily follow him. He is more of a graduate student's ideal. By the way, you won't believe this, but Jack Bybee used to work with him as a consultant in low temperature physics (Jack was the consultant) and Jack has the greatest respect for him. Jack tells me that Feynman would at least try something, before condemning it. Jack likes Landau as well.
Polarity depends on the mike, SY Many mikes have output transformers and they can get the polarity back easily.
Exactly. But for every SPL "in" on the leading edge of a displacement pulse, there's an SPL "out" on the trailing edge of the pulse.
I trust you're joking, SY. The air pressure in your listening room never goes negative. That you can see a negative sweep on a scope is an artifact of the microphone's inability to hold at the peak position the same way the woofer cone is doing. In that sense you're losing information that was present in the room; it wouldn't make it onto a recording and would thus technically be bandwidth limiting, or in more general terms a type of distortion.
If you were to arrange a digital on/off pulse from 0 to 5V and back to 0, you'd accomplish much the same thing, but the microphone could track the waveform more accurately.
All of which gets me thinking about low end response--that being the problem here--and the fact that it would be nice to have a microphone that was reasonably flat down to DC. Yes, yes, yes, there are other ways you can read DC and signals down to a fraction of a cycle, but they're not microphones. Many of the circuits available can reproduce to DC, so how much music are we losing because the recordings are limited to, say, 10Hz? As Ivor Tiefenbrun used to say, all musical waveforms start at 0Hz.
Grey
If you were to arrange a digital on/off pulse from 0 to 5V and back to 0, you'd accomplish much the same thing, but the microphone could track the waveform more accurately.
All of which gets me thinking about low end response--that being the problem here--and the fact that it would be nice to have a microphone that was reasonably flat down to DC. Yes, yes, yes, there are other ways you can read DC and signals down to a fraction of a cycle, but they're not microphones. Many of the circuits available can reproduce to DC, so how much music are we losing because the recordings are limited to, say, 10Hz? As Ivor Tiefenbrun used to say, all musical waveforms start at 0Hz.
Grey
GRollins said:
The absolute pressure can't be negative, but the instantaneous sound pressure sure can be, and by definition it has to be negative if at some point it is positive.
john curl said:
Who? Martin Landau as in Space 1999. He and Barbara Bain were famous objectivists. I think there should be a corrolary to Godwinn's Law for Ayn Rand.
Who said it did? No such thing as negative pressure.
What's the second derivative of a simple out-and-back square pulse, like that of your proposed experiment?
I just found something from Richard Heyser published almost 30 years ago on absolute polarity. This was in 'Audio' magazine, September 1979.
Leaders in the audio field, such as Hansen and Madsen have point out the audible difference caused by audio polarity, and recently Moncrieff has taken the important step of alerting the listening public to this effect and has identified a number of amplifiers which invert signal polarity. I now publicly call upon the entire audio industry, from computer composer through loudspeaker and headphone manufacturer to acknowledge polarity as a psychoacoustic parameter and identify either the polarity or the phase convention of the product.
Leaders in the audio field, such as Hansen and Madsen have point out the audible difference caused by audio polarity, and recently Moncrieff has taken the important step of alerting the listening public to this effect and has identified a number of amplifiers which invert signal polarity. I now publicly call upon the entire audio industry, from computer composer through loudspeaker and headphone manufacturer to acknowledge polarity as a psychoacoustic parameter and identify either the polarity or the phase convention of the product.
- Status
- Not open for further replies.