Absolute Polarity in Mastering

[Jerry Tubb]

(Originally Posted by Jerry Tubb)
........................
BUT once in a blue moon, when previewing a project that sounds noticeably lifeless and grey, I'll flip it, and if it gets better, I'll check with the producer, and we'll probably go with it.

Originally Posted by clarkjohnsen

Exactly right. It "sounds noticeably lifeless and grey" is a perfect description of the phenomenon I call "the muffling distortion". Whichever, I must again be at pains to point out that that's over your system. When the CD reaches the customer's home, either it or his system may well be the opposite! You might as well have left it the way it was.

Huh? It would still sound better, because at that point the record had been mastered.

Well in this one case, flipping the polarity was just enough Butterfly Effect for the band to say Wow! as we previewing the mixes, at the beginning of the session, before I had turned any EQ knobs. Proceeded to master the record, which was easier because of the "flip".

As a "meat and potatoes" style ME, I usually don't go for a lot of nebulous esoteric BS, but in this one instance, about five years ago, it actually helped.

JT

[clarkjohnsen]

Originally Posted by Jerry Tubb

Huh? It would still sound better, because at that point the record had been mastered.

Sound better where? Maybe where you are, but by the time it goes through several more stages of processing and through God-knows-what-polarity the listeners' systems have, it's a wash which way you make it.

Well in this one case, flipping the polarity was just enough Butterfly Effect for the band to say Wow! as we previewing the mixes, at the beginning of the session, before I had turned any EQ knobs. Proceeded to master the record, which was easier because of the "flip".

I understand and I agree. But then... it leaves your hands.

clark

[clarkjohnsen]

Originally Posted by bob katz

Clark, you are destined to get completely random results using a Polarity tester in a manner which it was not designed.

That sounds like the FDA defending their chosen uses for medicines. But the fact remains, some drugs have a profound and positive effect on other, unlisted diseases.

Besides you do not know that it was not designed for this application too; that's just an assertion you made.

It is not possible to technically examine a complex musical waveform whose origin you do not know and determine by measurement if its polarity is correct or incorrect or even just one way or another.

Talk about assertions! One by one:

-- ...not possible to technically examine. Is it then not possible at all? Or is there a nontechnical way I'm unfamiliar with?

-- ...a complex musical waveform. Some are less complex than others; are you saying there is no point along the complexity scale where analysis is possible?

-- ...whose origin you do not know. I can't see what knowledge of origin has to do with it. On the other hand, a rim shot is a rim shot.

-- ...[you cannot] determine by measurement if its polarity is correct or incorrect. Says you! In fact, I can, and have. Not only that but, as revealed in The Wood Effect, there are several devices on the market that do the same job. I can't think where you get these assertions.

-- ...or even just one way or another. The SMART thing of course differentiates between two polarities (au contraire!) and also manages to indicate which instrumental transients are compressive and which rarefactive, in the acoustic field. Naturally it's up to the user to know which is "correct" -- as determined by the instrument in question.

clark

[dcollins]

Originally Posted by clarkjohnsen

-- ...or even just one way or another. The SMART thing of course differentiates between two polarities (au contraire!) and also manages to indicate which instrumental transients are compressive and which rarefactive, in the acoustic field. Naturally it's up to the user to know which is "correct" -- as determined by the instrument in question.

I think you have to use the built-in SMAART test signal, not music, or you are likely to get a 50% result.


DC

[clarkjohnsen]

Originally Posted by dcollins

I think you have to use the built-in SMAART test signal, not music, or you are likely to get a 50% result.

Sorry, wrong thinking! First, there's no built-in test signal, it relies on an included CD to generate the test sequence. More to the point, I can testify as a user that it is sensitive to music -- say, a drum thwack, or anything with a pronounced transient. And it's always correct!

Of course you do get a "50% result" -- because the two polarities are equally present, split 50/50.

clark

[Nordenstam]

Originally Posted by clarkjohnsen

More to the point, I can testify as a user that it is sensitive to music -- say, a drum thwack, or anything with a pronounced transient. And it's always correct!

Of course you do get a "50% result" -- because the two polarities are equally present, split 50/50.

Now that's some dizzyingly fast circular logic!

I recall that even you admitted that most gear does get the polarity right. That would make the statistical chances of 50/50 in the final production very very small..

[clarkjohnsen]

Originally Posted by Lupo

Now that's some dizzyingly fast circular logic!

You're too kind.

I recall that even you admitted that most gear does get the polarity right.

First, what's with this "even [me]"? More importantly, I made no such statement. It is however true that most, but far from all, home reproduction electronics are non-inverting. (In a sense that is "right", but not I think in the way you intended.) Studio gear, alas, is another matter entirely.

That would make the statistical chances of 50/50 in the final production very very small..

After taking such pains to explain this over and over, I am still surprised that this bunch of competent, intelligent guys doesn't get it yet. So once more unto the breach dear friends, once more.

There is no extant standard for acoustic polarity direction on tape. There is no extant standard for acoustic polarity direction on LP (nor was there ever on 78s). There is no extant standard for acoustic polarity direction on CD. Therefore which signal direction comprises a compression transient, which a rarefaction, in musical acoustic space, is unspecified -- with the result necessarily split between the two choices... 50/50!

To make matters worse, there is not even consistency across the bands of any given album, nor across any one label's various releases.

Until mastering engineers (inter alia) figure out why this should be, the public will remain as confused about polarity as most people here seem to be.

clark

[dcollins]

Originally Posted by clarkjohnsen

There is no extant standard for acoustic polarity direction on tape. There is no extant standard for acoustic polarity direction on LP (nor was there ever on 78s). There is no extant standard for acoustic polarity direction on CD. Therefore which signal direction comprises a compression transient, which a rarefaction, in musical acoustic space, is unspecified -- with the result necessarily split between the two choices... 50/50!

Haven't we been over all this material before? The standard is no polarity inversion. Full stop. The media is immaterial.

As a physicist, certainly you must see that a 50% (null) result is trying to tell you something..............


DC

[24-96 Mastering]

Originally Posted by dcollins

As a physicist, certainly you must see that a 50% (null) result is trying to tell you something..............

Having used polarity testers before, I strongly second this notion.

[glassmaster]

Originally Posted by clarkjohnsen

What clarkjohnsen actually wrote was, "No standard exists for impressing acoustic polarity consistently on any recorded media." Does anyone not see the difference?

Actually Clark, there IS a standard, and it encompasses the ENTIRE electroacoustic signal chain.

AES26-2001 (r2006): AES recommended practice for professional audio -- Conservation of the polarity of audio signals (Revision of AES26-1995)

Abstract
This document standardizes the polarity of the signals at the various interface points between different items of equipment, in particular from the acoustical, electrical, mechanical, digital, and magnetic aspects. Each item of equipment complies separately with the polarity requirements for the input and output signals.

[Adam Dempsey]

“I borrowed some albums from a friend of mine. I was gonna record them onto tape cassettes. I had the wiring all backwards on my stereo and accidentally erased all the records. I gave him back these blank, black things.”
- Steven Wright

The whole thing gives me ridiculous ponderings about the planetary standard... the Coriolis effect and polarity... if it's a hemispherical thing (ignoring, perhaps, universal electromagnetic laws)... cyclones/hurricanes/low pressure systems/water down a plug hole all naturally spin clockwise here, as opposed to northern hemisphere.
Perhaps less "work" and greater design efficiency would be involved if records & CDs behaved the same...

Maybe another good coffee is not in order before my late session this eve...

[Lagerfeldt]

Originally Posted by glassmaster

Actually Clark, there IS a standard, and it encompasses the ENTIRE electroacoustic signal chain.

AES26-2001 (r2006): AES recommended practice for professional audio -- Conservation of the polarity of audio signals (Revision of AES26-1995)

Abstract
This document standardizes the polarity of the signals at the various interface points between different items of equipment, in particular from the acoustical, electrical, mechanical, digital, and magnetic aspects. Each item of equipment complies separately with the polarity requirements for the input and output signals.

Can we close this thread now, please... it's going round in circles.

[dcollins]

Originally Posted by Lagerfeldt

Can we close this thread now, please... it's going round in circles.

I wonder if Clark will be adding contributions from this very thread into the second edition of his book?


DC

[Lagerfeldt]

No, I remembered to add a © to all my posts...

To answer your question seriously though: no, I don't see that happening.

[clarkjohnsen]

Originally Posted by dcollins

Haven't we been over all this material before? The standard is no polarity inversion. Full stop.

That's an assertion unbacked by the evidence. Or are you saying, ain't no such thing as polarity inversion? That would be... amusing...

The media is immaterial.

Immaterial to the whichever polarity is inscribed, yes. But the whole point was, ALL the various media are subject to this variation.

As a physicist, certainly you must see that a 50% (null) result is trying to tell you something..............

Begging your pardon, but you've shifted grounds. "A 50% (null) result" between two pieces of gear, say, would mean there's no detectable difference; when two binary events are equally split, the 50% means there's no predictability, not no detectable difference.

clark

[clarkjohnsen]

Originally Posted by glassmaster

Actually Clark, there IS a standard, and it encompasses the ENTIRE electroacoustic signal chain.

Is that right? LPs too? I don't think so! But read on.

AES26-2001 (r2006): AES recommended practice for professional audio -- Conservation of the polarity of audio signals (Revision of AES26-1995)

Abstract
This document standardizes the polarity of the signals at the various interface points between different items of equipment, in particular from the acoustical, electrical, mechanical, digital, and magnetic aspects. Each item of equipment complies separately with the polarity requirements for the input and output signals.

All well and good, and I have been long familiar with this document. In fact I tried to force some changes on it at the time of writing, but they weren't having it. Throughout the document the word "polarity" is used to mean electrical polarity. This can readily be seen in the sentence, "Since the introduction of multichannel techniques, and in particular stereophony, [polarity preservation] has become essential to permit the satisfactory combination of signals from different sources."

Our topic here is acoustic polarity. The "standard" is uselsss in this respect, as it never manages even to define the concept, much less standardize it. I made this point earlier with Mr. Katz, when he asserted that he can measure a burst output from his CD player (or whatever it was, I forget) and know that its polarity is positive. Yes, but, what is the connection to an acoustic transient? He had no answer, as there is none.

Gentlemen, I repeat repeat repeat: No means, no standards exist to ensure that an acoustic transient that hits the microphone as a compression wave, will necessarily output as a positive-going voltage from a CD, LP or tape on any CD player, phono cartridge or tape playback unit.

It was that connection that I strived to get the AES committee to include, as otherwise the document would be pretty useless. Which it is.

Finally, even if it did perform as the standard it is alleged by glassmaster to be, it would have been to date as much honored in the breach as in the observance, to paraphrase Hamlet.

clark

[clarkjohnsen]

Originally Posted by Adam Dempsey

The whole thing gives me ridiculous ponderings about the planetary standard... the Coriolis effect and polarity... if it's a hemispherical thing (ignoring, perhaps, universal electromagnetic laws)... cyclones/hurricanes/low pressure systems/water down a plug hole all naturally spin clockwise here, as opposed to northern hemisphere.
Perhaps less "work" and greater design efficiency would be involved if records & CDs behaved the same...

Wouldn't that be grand? I've spent over two decades trying to solve this problem. Mostly what I get is poo-poohed and ridiculed. And the higher up a fellow is placed in the, ah, recording chain -- the more adverse his reaction. Odd.

clark

PS Wasn't that hemispherical thing on the Coriolis effect disproved?

[clarkjohnsen]

Originally Posted by Lagerfeldt

Can we close this thread now, please... it's going round in circles.

Oh, another Coriolis effect?

But I must beg to differ. The AES "standard" you quote/quote does not apply, as shown in another response. So long as people here think the problem is solved, it will never be solved.

clark

[clarkjohnsen]

Originally Posted by dcollins

I wonder if Clark will be adding contributions from this very thread into the second edition of his book?

More likely I'd add remarks from the polarity threads over on Steve Hoffman's forum. (You've heard of Steve Hoffman?) There one can find a minimum of sarcasm and a maximum of comprehension, for Steve himself quite understands absolute polarity -- as do many if not most of his contributors.

clark

[dcollins]

Originally Posted by clarkjohnsen

That's an assertion unbacked by the evidence. Or are you saying, ain't no such thing as polarity inversion? That would be... amusing...

This thread has had enough comedy.

It's not like recording, mixing, and mastering is somehow flipping a coin and generating records with random polarity from song to song.

The electrical standard of preserving polarity throughout the chain is clear. The demonstration of this all the way to the acoustical impulse in air, equally so.


DC

[MattGray]

Interesting topic, I think it's been established that there is a standard for absolute electrical polarity which is being maintained by most engineers who are conscientious & thorough in setting up their playback & processing equipment. We can test & set this up right so why wouldn't we? It's also safe to say that for the consumer playback system it could be quite a mixed bag. I also agree that for the most part there is no standard being maintained effectively for acoustic polarity direction that could end up being recorded or released on any available medium.

Today for example i have an instrumental track that was produced entirely within Reason 4. Should be correct absolute polarity being an all digitally created recording & mixed in the box right? I found that the kick was a bit wimpy & weird sounding on the Duntechs, looking at the waveform it's front edge transient was inversed, but the snare was positive. Flipping the absolute phase restored the impact & solidity of the kick but then the snare lost some impact.... So now I have to make a decision do I favour the polarity of the snare or the kick? Further more whichever I choose to favour will only be played back the way I prefer on correctly calibrated monitoring systems. Will it really matter which way I choose?

Perhaps as mastering engineers we can choose to favour one way or another, but it's more likely that if there is an obvious polarity error in a mix that it will be tied to one element in the mix & not the entire mix itself. Correcting the obvious fault by flipping the whole mix might fix it but then it will likely create other problems with other elements in the mix. I think the mix engineer has much more power over maintaining or correcting these issues with the individual tracks. Mastering engineers, not so much (unless you have separate stem files that isolate the problem element).

Matt

[Bob Olhsson]

Virtually all microphone manufacturers specify how their microphones will respond electrically to a positive change in air pressure and any system meeting the polarity standards will reproduce the very same change in air pressure from the loudspeaker, assuming it is capable of doing so.

[dcollins]

Originally Posted by Bob Olhsson

Virtually all microphone manufacturers specify how their microphones will respond electrically to a positive change in air pressure and any system meeting the polarity standards will reproduce the very same change in air pressure from the loudspeaker, assuming it is capable of doing so.

Well, there's page one of the new book for you. I'm thinking big fonts and lots of illustrations.


DC

[soulviasound]

What I do when I discover absolute phase error ? I found a simple solution. Give client a mono master... That's right! I was just kidding!

If it's a serrios absolute phase error I'm sure that my ears will detect from the first listenings, or after some pause. If not, I don't pay attention. Keep my focus on less compromission and how can I obtain best result with less processing. At least, I do music mastering, not PCB mastering.

Actually, long time ago I recive a mix with huuuuge phase error. I mean huuuge! When I push the play button, I tought that I get a M/S mix. I wonder myself if I miss something with the new "mixdown trends". In fact, the "mixerman" had a problem with one chanell of the D/A convertor.

[Lagerfeldt]

Since my thread has descended into funny logic a long time ago, I remembered this letter

[Nordenstam]

^^^^^^

Now that's funny!



@ Clark: apologies for the rudeness earlier in the thread.

[Bob Yordan]

Wow, what a thread.

I visited a friend of mine, last week, that builds speakers (Intelligent Sound - Petter).

He switched between a pair of Tannoys & his phase compensated
speakers and the difference was major.

With his speakers I could walk around in the room, hearing the sound
correctly but with the Tannoys even in the sweet spot I could hear
phasing issues. I guess the brain will get more tired earlier during mixing
sessions if there are phase issues involved.

I will soon be a happy owner a pair of IS8 Pro, for detailed and clear mixing.

[clarkjohnsen]

Originally Posted by MattGray

Interesting topic, I think it's been established that there is a standard for absolute electrical polarity which is being maintained by most engineers who are conscientious & thorough in setting up their playback & processing equipment.

You have that right, mate. Only, no need to say "absolute".

It's also safe to say that for the consumer playback system it could be quite a mixed bag. I also agree that for the most part there is no standard being maintained effectively for acoustic polarity direction that could end up being recorded or released on any available medium.

Congratulations! You have grasped the situation. Why is that so hard for so many others?

You don't have to answer that!

Today for example i have an instrumental track that was produced entirely within Reason 4. Should be correct absolute polarity being an all digitally created recording & mixed in the box right?

I must caution you however against use of both "correct" and "absolute"; it clouds the issue.

I found that the kick was a bit wimpy & weird sounding on the Duntechs, looking at the waveform it's front edge transient was inversed, but the snare was positive.

How did that happen?

Flipping the absolute phase...

No, sir; flipping the POLARITY. Simple as that.

And by so doing you re-established the Absolute Polarity of the music/acoustic field.

restored the impact & solidity of the kick but then the snare lost some impact.... So now I have to make a decision do I favour the polarity of the snare or the kick?

Or do you decide that it scarcely matters, as down the line somewhere it may well get flipped again?

Further more whichever I choose to favour will only be played back the way I prefer on correctly calibrated monitoring systems. Will it really matter which way I choose?

Oh. I forgot: you get it!

Perhaps as mastering engineers we can choose to favour one way or another, but it's more likely that if there is an obvious polarity error in a mix that it will be tied to one element in the mix & not the entire mix itself. Correcting the obvious fault by flipping the whole mix might fix it but then it will likely create other problems with other elements in the mix. I think the mix engineer has much more power over maintaining or correcting these issues with the individual tracks. Mastering engineers, not so much (unless you have separate stem files that isolate the problem element).

The best the mastering engineer can do is keep all stems in alignment. After the work leaves his hands then, the listener will have the music all right or all wrong, but at least one simple flip will rectify the whole lot.

clark

[clarkjohnsen]

Originally Posted by dcollins

........ The electrical standard of preserving polarity throughout the chain is clear. The demonstration of this all the way to the acoustical impulse in air, equally so.

MattGray and I (inter alia) beg to differ.

Were you correct, polarity switches on home gear would be unnecessary. But every astute listener knows, that switch must be flipped!

clark

[clarkjohnsen]

Originally Posted by Bob Olhsson

Virtually all microphone manufacturers specify how their microphones will respond electrically to a positive change in air pressure and any system meeting the polarity standards will reproduce the very same change in air pressure from the loudspeaker, assuming it is capable of doing so.

Are you saying then that there is no variation of polarity on LP, CD and tape?

clark