absolute polarity

[pucho812]

had a guy try and convince me that a mono kick sample sounds different at 0 degrees vs 180 degrees polarity and that it mattered which direction, in or out, the speaker goes first.  This took a long time as he kept interchanging phase vs polarity and finally speaker direction. Having listen to his example of a kick with polarity vs polarity flipped I couldn't denote the slightest bit of difference. 

so I put it up to the guys here. 

does it really matter on a single instance of a mono kick where polarity is at?
I say no because there is nothing else to reference to. More over you can't really say the signal is in phase as phase would mean he is comparing a relationship of more then one signal which was not the case here. I also added that if it truly sounded different their might be an issue with his system... I say this falls into the realm of psycho acoustics  So what say you?

 

[hazel]

this falls into the realm of psycho acoustics

Often called Audiophoolish.

I'll trust any person saying can hear absolute polarity on a mono source just because i think "Why he/she would lie?"

But i tend to react to facts...scientific facts and i strongly believe that there's no acoustical difference on playing a mono source with phase flipped or not.

Of course i'm an open-minded guy. If someone upload a mono audio sample i'll take a listen.

 

[emrr]

Vocals and horns which are highly asymmetric can be obvious. Kick by itself is a different question.  If it's been limited by something that limits asymmetrically, drums can sound better one way or other. 

 

[JohnRoberts]

had a guy try and convince me that a mono kick sample sounds different at 0 degrees vs 180 degrees polarity and that it mattered which direction, in or out, the speaker goes first.  This took a long time as he kept interchanging phase vs polarity and finally speaker direction. Having listen to his example of a kick with polarity vs polarity flipped I couldn't denote the slightest bit of difference. 

so I put it up to the guys here. 

does it really matter on a single instance of a mono kick where polarity is at?
I say no because there is nothing else to reference to. More over you can't really say the signal is in phase as phase would mean he is comparing a relationship of more then one signal which was not the case here. I also added that if it truly sounded different their might be an issue with his system... I say this falls into the realm of psycho acoustics  So what say you?

Well yes there is a real signal characteristic called "absolute polarity" and in some cases the difference is audible. This is generally a pretty subtle phenomenon and a kick drum is not the poster boy sound for demonstrating absolute polarity. While it may be about the easiest signal to visualize absolute polarity with, it is low on the list for audibility.

Higher on the list for audibility are sound sources with asymmetrical waveforms like brass or wind instruments. Another good source useful for hearing absolute polarity are human vocals.

That said this is a fairly subtle phenomenon, and if the audio chain you are listening on is not very precise, the difference may be even less apparent. 

Back before about the early-mid '80s the popular wisdom was that there was no audible difference so equipment designers and studio integrators didn't worry about the details. In fact after people started looking into this, a review of popular recordings found that the absolute polarity was not even consistent from track to track on the same side of some records. I sold a phono preamp back then with an absolute polarity switch, but the market response was underwhelming. More evidence of how subtle the phenomenon is.   

If you search google you may find some specialized asymmetrical sound files that are optimized for audibility of this, but in the meanwhile you might try listening to a very well recorded dry vocal or horn parts that you are familiar with. Be sure to make sure you eliminate all other variables like how loud, and even where you are standing in the room. The difference is not night and day, but one way may sound more correct. Vocals are probably good because we know what real vocals are supposed to sound like.

I consider this important for archival integrity. When designing gear or making recordings why not do it right, even if you can't hear it?

JR

 

[audiomixer]

Inverted polarity kick sucks (air), literally. 

- Michael

 

[Spino]

had a guy try and convince me that a mono kick sample sounds different at 0 degrees vs 180 degrees polarity and that it mattered which direction, in or out, the speaker goes first.

From my days in the critical listening class (it was given by Sean Olive, so the source was rather good) I remember learning that it was not yet proven scientifically that the human ear could recognize absolute polarity.

Please note that this does not mean that we can not, just that it wasn't proven back then. I should also add that those classes were more than 15 years ago, so my memory might just be mixing things up...

 

[Script]

I once met a mixing/mastering guy who looks at stems to nudge and phase-align the waveforms of mainly kicks and bass material (but also higher-register material), so that most transients go roughly in the same direction.

Well, apart from this sounding like hell of a lot of work to me, I think this guy is a bit too anal about the shape of waveforms. But he claims that after spending hours on doing this, the music sounds better... Granted, his mixes and mastering results sound pretty good and clean in the bass -- only objection I'd have: would a bit of eq'ing or HPF'ing to seperate instruments have the same/similar effect? And be a little bit faster, too boot?

 

[JohnRoberts]

I once met a mixing/mastering guy who looks at stems to nudge and phase-align the waveforms of mainly kicks and bass material (but also higher-register material), so that most transients go roughly in the same direction.

Well, apart from this sounding like hell of a lot of work to me, I think this guy is a bit too anal about the shape of waveforms. But he claims that after spending hours on doing this, the music sounds better... Granted, his mixes and mastering results sound pretty good and clean in the bass -- only objection I'd have: would a bit of eq'ing or HPF'ing to seperate instruments have the same/similar effect? And be a little bit faster, too boot?

Time aligning transients is a somewhat different characteristic, but for different sources played by different musicians, he's tweaking the performance as much as mic placement errors.

JR

 

[JohnRoberts]

had a guy try and convince me that a mono kick sample sounds different at 0 degrees vs 180 degrees polarity and that it mattered which direction, in or out, the speaker goes first.

From my days in the critical listening class (it was given by Sean Olive, so the source was rather good) I remember learning that it was not yet proven scientifically that the human ear could recognize absolute polarity.

Please note that this does not mean that we can not, just that it wasn't proven back then. I should also add that those classes were more than 15 years ago, so my memory might just be mixing things up...

I first wrote about this in my Audio Mythology column (as being real, not myth) back in the early '80s  so more than 30 years ago.

Clark Johnson wrote an AES paper in 1991 so over 20 years ago. Johnson mentioned my column in his book "Wood effect" but I was not exactly a pioneer on the subject, just an early observer who was correct. Back then there was much debate about it not being real and dueling experts. I also recall an authoritative article about it from a studio technician (Peter Butt?) in RE/P magazine also 1980s.     

I repeat this is subtle, but real. That said in the real world relative polarity from multiple stem summation of delayed/phase shifted signals can be far more apparent.

JR

 

[Script]

Meticulous mic placement and fixing time/phase, yes, Steve Albini might be a good example -- hardly any effects, mostly choice of mic and placement.

Anyway, this mixing guy I mentioned even time and phase (flip) aligns stuff that's not recorded with mics but straight out of synths and computers etc., claiming that once the first/loudest transients are all aligned (often creating a huger peak in summation), they have to make the speaker cones move outward (! <-- i.e., absolute phase) to sound good and create "the punch" (sic). For him, assymentrical stuff has to make speakers move more outward than inward.

Sometimes it works flipping a kick 180° in a mix before summing. Subtle, but it's there. But a kick just on its own... and synthetic kicks are often sinewave-based anyway.

 

[living sounds]

Maybe I'm particularly sensitive to it, but I can easily hear absolute polarity on asymmetrical signals, especially on kick drums. At least on my current listening system. For instance, on a kick the tonal balance can shift quite drastically, transforming a punchy lower midrange "knock" into a low "oomph", or on a snare drum it might sound as if you just added 3 db at 200 Hz etc.

I do that flipping the phase so the transients point "up" in the waveform display thing, too. 

 

[JohnRoberts]

Maybe I'm particularly sensitive to it, but I can easily hear absolute polarity on asymmetrical signals, especially on kick drums. At least on my current listening system. For instance, on a kick the tonal balance can shift quite drastically, transforming a punchy lower midrange "knock" into a low "oomph", or on a snare drum it might sound as if you just added 3 db at 200 Hz etc.

I do that flipping the phase so the transients point "up" in the waveform display thing, too. 

Be careful when flipping the polarity of one mic in a multi-mic recording. There may be enough leakage into other nearby mics that the polarity flip (or delay tweak) will cause differences in how these multiple versions of the same signal add constructively or destructively.

Are you listening the kick drum stem alone, or in a mix?

Note: I am not arguing about what you hear, i just do not expect this kick drum sound to have a strong polarity identity. Drum kits are notorious for having LF, long wavelength sounds, picked up by multiple nearby mics. It is not unusual to experimentally flip polarity of drum kit mics one at a time for best sound overall while this may conflict with absolute polarity for one or more of the mics.

JR

 

[living sounds]

Be careful when flipping the polarity of one mic in a multi-mic recording. There may be enough leakage into other nearby mics that the polarity flip (or delay tweak) will cause differences in how these multiple versions of the same signal add constructively or destructively.

Are you listening the kick drum stem alone, or in a mix?

Note: I am not arguing about what you hear, i just do not expect this kick drum sound to have a strong polarity identity. Drum kits are notorious for having LF, long wavelength sounds, picked up by multiple nearby mics. It is not unusual to experimentally flip polarity of drum kit mics one at a time for best sound overall while this may conflict with absolute polarity for one or more of the mics.

JR

No, this has nothing to do with interference (=summing of waveforms), just auditioning single tracks. But I work a lot with electronic drums and heavily processed samples etc., which may make it more noticeable.

I also find the effect easily identifieable on complete mixes.

 

[Spino]

JR, far from me the questioning of whether it can be heard or not, what I was saying is that I don't believe it has been scientifically proven; that is a tougher question.

Since I don't want to put in Sean Olives' mouth words he might have never said, I went around the Internet looking for more infos, and the literature out there is quite varied and opposite. As a first example, one site said it could be heard only on mid-high frequencies while another said it couldn't be heard over 500 Hz... Gearsluts is full of threads of people trolling over the terminology and the effects of it, and most of it is OT.

Let's start by noting that in the real world it is nonsense. With our kick drum example, where a kick is supposed to emit a positive sound pressure to the listener, the drummer, being on the batter side, would have been hearing the wrong polarity from his drum his entire life. To a guitar string the concept is even harder to apply: the guitar string moves in a 360 circle, and the first direction of movement varies depending on the direction it's been plucked, and again, absolute polarity would vary depending on where you'd be listening the guitar from.

So this is a matter of pure audio reproduction. As a recording and mixing engineer, I am very aware of the signal path of my recordings, and it is in my usual interests to keep the absolute polarities of what I am recording consistent over the process of record making. But when faced with the websites that let you ABX recognizable absolute polarity waves signals, I am sceptical. Even if I'd be able to correctly recognize them (and I believe I will, I'll try as soon as I have a little time), the only thing that scientifically can be said is that my whole system is not perfectly symmetrical: from the wave file, through converters and tens of opamps, ics, resistors, capacitors, through cables and connections, through transducers,  through the room,  and in the end through my ears, I might state with certainty that somewhere on that path that audio is now polarity-recognizable. Does it mean that scientifically I have proven that I can hear absolute polarity?

Again, this is not to say the thing does not exist and is not hearable or reproducible, I am curious to understand this phenom better, thus the questions.


Cheers

sp

p.s.: obviously this has nothing to do with the flipping of the phase on multi mic sessions.

I first wrote about this in my Audio Mythology column (as being real, not myth) back in the early '80s  so more than 30 years ago.

Clark Johnson wrote an AES paper in 1991 so over 20 years ago. Johnson mentioned my column in his book "Wood effect" but I was not exactly a pioneer on the subject, just an early observer who was correct. Back then there was much debate about it not being real and dueling experts. I also recall an authoritative article about it from a studio technician (Peter Butt?) in RE/P magazine also 1980s.     

I repeat this is subtle, but real. That said in the real world relative polarity from multiple stem summation of delayed/phase shifted signals can be far more apparent.

JR

 

[JohnRoberts]

JR, far from me the questioning of whether it can be heard or not, what I was saying is that I don't believe it has been scientifically proven; that is a tougher question.

This is a pretty old debate and "some" double blind tests have demonstrated a weak proof of audibility. But I repeat this is subtle.

There are also criticisms of this audibility, that the source of the audible artifacts are a secondary effect caused by loudspeaker linearity differences related to driver polarity. 

Since I don't want to put in Sean Olives' mouth words he might have never said, I went around the Internet looking for more infos, and the literature out there is quite varied and opposite. As a first example, one site said it could be heard only on mid-high frequencies while another said it couldn't be heard over 500 Hz... Gearsluts is full of threads of people trolling over the terminology and the effects of it, and most of it is OT.

The internet has never lacked for mis-information.

Let's start by noting that in the real world it is nonsense. With our kick drum example, where a kick is supposed to emit a positive sound pressure to the listener, the drummer, being on the batter side, would have been hearing the wrong polarity from his drum his entire life. To a guitar string the concept is even harder to apply: the guitar string moves in a 360 circle, and the first direction of movement varies depending on the direction it's been plucked, and again, absolute polarity would vary depending on where you'd be listening the guitar from.

I repeat, the phenomenon is subtle and IMO mostly of value for archival purposes. In many cases relative polarity of different mics will trump absolute polarity for sound impact.

I also repeat that the drum sound is not noted for having audible absolute polarity, and drumheads propagate sound in both directions. Arguably with a two head drum there are two sound sources, and two polarities to deal with, but there is no need to over analyze.

So this is a matter of pure audio reproduction. As a recording and mixing engineer, I am very aware of the signal path of my recordings, and it is in my usual interests to keep the absolute polarities of what I am recording consistent over the process of record making. But when faced with the websites that let you ABX recognizable absolute polarity waves signals, I am sceptical. Even if I'd be able to correctly recognize them (and I believe I will, I'll try as soon as I have a little time), the only thing that scientifically can be said is that my whole system is not perfectly symmetrical: from the wave file, through converters and tens of opamps, ics, resistors, capacitors, through cables and connections, through transducers,  through the room,  and in the end through my ears, I might state with certainty that somewhere on that path that audio is now polarity-recognizable. Does it mean that scientifically I have proven that I can hear absolute polarity?

Again, this is not to say the thing does not exist and is not hearable or reproducible, I am curious to understand this phenom better, thus the questions.


Cheers

sp

p.s.: obviously this has nothing to do with the flipping of the phase on multi mic sessions.

Allow me to offer another tidbit. Since the audibility of absolute polarity is associated with asymmetrical signals, and we also know that such waveforms can interact with path non linearity to generate audible artifacts, I would not dismiss the possibility of non linearity inside our ear's mechanical path, affecting different people's experience with hearing or not hearing this. Perhaps hearing this is evidence of distorted/damaged hearing. 

In theory this should not exist as an audible phenomenon. Ironically this is reported to be more audible with more accurate playback paths, odd if the source of this phenomenon is in fact path distortion.

In my judgement it has been demonstrated that some people reliably hear this, making it real, while there may be some complicit factor that was not adequately controlled for in the double blind listening tests (like loudspeaker or inner ear distortion).   

JR

[edit] like so much extreme discussion of audio, this is like listening for the angel's footsteps on the pin head. Audio path design is so mature, and so good, that we must amuse ourselves with esoterica.  [/edit]

 

[PRR]

Look at it from the cave-man point of view. (We were evolved/created to live in the wild before we got side-tracked by houses with hi-fi systems.)

Naked and alone in the jungle.

You hear a sound.

Can you eat it?

Can it eat you?

Pitch and intensity are very significant. A branch-snap that is loud and low is something large. Even if it is not a people-eater, you may need to stay out of the way. Small sharp sounds may be tasty treats.

Reflections, when available, are important. Many of us can drop a screw under the desk and know where to start looking by the echo pattern off sidewalls. Same is true of tiger-snaps reflected from trees or cliffs, and perhaps sounds inside a cave.

Harmony and rhythm have meaning. If sounds are time-related they all come from the same thing; if not, then multiple things are happening. One mastodon versus several mastodons- one quadruped has a step-pace, several quadrupeds have several step-paces that generally do not sync. A tone with 3rd and 5th harmonics is one throat-pipe calling, tones that organize with separate harmonic series suggest several throat-pipes calling.

Pitch, intensity, reflections, harmony, and rhythm are key properties of organized speech, singing, and music. They tickle our brain special ways because we are built to appreciate them.

It is hard for me to see a cave-man need for Absolute Polarity. There's sounds that always start a certain polarity, and sounds which arrive with random polarity for reasons which do not matter for survival.

 

[CJ]

just do some experiments to see if you hear any difference with a polarity swap,

you will hear a difference if you are listening to the kick drum with headphones while you are kicking it, this is because the acoustic signal is getting mixed with the mixing board signal, so you have to record the kick and listen without kicking it,

what kind of mic are we talking about ? a Senn 602 or a speaker mic?

what kind of preamp? remember that a kick signal is kind of like a DC spike,

so if the preamp is sensitive to the polarity of the DC spike, you will hear a difference,

could be that a single ended preamp will sound different than a balanced input,

maybe the direction of the windings of the input transformer will make a difference,

maybe the coupling caps will make a difference if they are polarized (lytic)

sounds like a fun experiment, set up the drum, try a bunch of different mics and pre-amps, record the kick drum and play it back, maybe check it on a scope,

 

[bruce0]

Let's see... In Maine... a small snap, maybe a squirrel, a big snap, a moose.  In both cases the answer to the question "can you eat it" is yes (at least in Maine!) :

Look at it from the cave-man point of view. (We were evolved/created to live in the wild before we got side-tracked by houses with hi-fi systems.)

Naked and alone in the jungle.

You hear a sound.

Can you eat it?

Can it eat you?

Pitch and intensity are very significant. A branch-snap that is loud and low is something large. Even if it is not a people-eater, you may need to stay out of the way. Small sharp sounds may be tasty treats.

Reflections, when available, are important. Many of us can drop a screw under the desk and know where to start looking by the echo pattern off sidewalls. Same is true of tiger-snaps reflected from trees or cliffs, and perhaps sounds inside a cave.

Harmony and rhythm have meaning. If sounds are time-related they all come from the same thing; if not, then multiple things are happening. One mastodon versus several mastodons- one quadruped has a step-pace, several quadrupeds have several step-paces that generally do not sync. A tone with 3rd and 5th harmonics is one throat-pipe calling, tones that organize with separate harmonic series suggest several throat-pipes calling.

Pitch, intensity, reflections, harmony, and rhythm are key properties of organized speech, singing, and music. They tickle our brain special ways because we are built to appreciate them.

It is hard for me to see a cave-man need for Absolute Polarity. There's sounds that always start a certain polarity, and sounds which arrive with random polarity for reasons which do not matter for survival.

 

[JohnRoberts]

just do some experiments to see if you hear any difference with a polarity swap,

you will hear a difference if you are listening to the kick drum with headphones while you are kicking it, this is because the acoustic signal is getting mixed with the mixing board signal, so you have to record the kick and listen without kicking it,

what kind of mic are we talking about ? a Senn 602 or a speaker mic?

what kind of preamp? remember that a kick signal is kind of like a DC spike,

so if the preamp is sensitive to the polarity of the DC spike, you will hear a difference,

could be that a single ended preamp will sound different than a balanced input,

maybe the direction of the windings of the input transformer will make a difference,

maybe the coupling caps will make a difference if they are polarized (lytic)

sounds like a fun experiment, set up the drum, try a bunch of different mics and pre-amps, record the kick drum and play it back, maybe check it on a scope,

I guess this depends on whether you are trying to design an experiment to prove that you can't hear this, or can. I repeat Drums are not noted for being very audible wrt absolute polarity. If you want to experiment to try to hear this phenomenon start with an asymmetrical waveform like a sawtooth (or brass, or maybe vocals)  that is considered more audible.

======
I didn't find any sound files from a quick search but here is a list of extra credit reading.  http://www.audioasylum.com/cgi/t.mpl?f=prophead&m=60282

JR

Craig et al., “Effect of Phase on the Quality of a Two-Component Tone”, J. of the Acoustical Society of America 1962, p.1752

Greiner et al., “Observations on the Audibility of Acoustic Polarity”, J. of the Audio Engineering Society 1994, p.245; Comments, JAES 1995, p.147

“While polarity inversion is not easily heard with normal complex musical program material, as our large-scale listening tests showed, it is audible in many select and simplified musical settings. Thus it would seem sensible to keep track of polarity and to play the signal back with the correct polarity to ensure the most accurate reproduction of the original acoustic waveform.”

Greiner et al., “A quest for the audibility of polarity”, Audio Magazine, Dec. 1993, p.40

Heyser, “Polarity convention”, Audio Magazine, Sept. 1979, p.18

Hilliard, “Notes on How Phase and Delay Distortions Affect the Quality of Speech, Music and Sound Effects”, IEEE Transactions on Audio, March-April 1964, p.23

Johnsen, “Proofs of an Absolute Polarity”, Audio Engineering Society paper 3169 (1991)

Knight, “Report of an Ad hoc meeting on the formation of an AES technical committee on audio polarity”, JAES 1981, p.528

Lipshitz et al., “On the Audibility of Midrange Phase Distortion in Audio Systems”, JAES 1982, p.580; Comments, JAES 1983, p.447

“The important point is that there is a well-established mechanism in the inner ear for detecting waveform asymmetries and hence polarity reversal of asymmetric signals. What is perhaps surprising is how subtle this effect generally appears to be on music and speech. As the above-mentioned experiment [13] indicates however, it is an audible factor, and should be taken into account when performing comparisons of audio components [15], [19], [20] -- acoustic polarity should be maintained.”

Long, “Polarity in absolute terms”, Audio Magazine, Aug. 1989, p.14 [about Clark Johnsen’s book “The Wood effect”]

Long, “Upside down – Find out whether you can hear polarity inversion ”, Audio Magazine, July 1996, p.35

 

[PRR]

> the answer to the question "can you eat it" is yes (at least in Maine!)

That's a survival question too. It is get-eaten, eat, or starve-to-death. Finding food through hearing avoids extinction.

Some folks here will eat anything. Porcupine is easy to catch though hard to prepare. Quill-rattle is not polarity-consistent, the pitch intensity and irregular rhythm are sufficient identification.

The only Maine critter dangerous to a naked human is the Black Fly. Since they may be above or below ear level, Absolute Phase hearing has no survival value.

FWIW: Moose kill people in cars at night. Their tummy is higher than a car's headlights, so you can hardly see the skinny legs, the half-ton body comes in at face-height. But don't blame the moose for speeding in the dark, only people are that reckless.
We have several packs of Coyote here but you never see them, they sense us better than we hear them. Maine Cougar and Maine Wolf are legendary and clearly less likely than meteor strike or lottery jackpot. Though perhaps more likely than the several Bigfoot-like creatures claimed to haunt some areas.