 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
66.30.248.240
In Reply to: Acoustic polarity, again posted by KlausR. on May 6, 2007 at 02:36:41:
It might only make half your record collection sound better, for free.Move along folks, nothing to see here.
Follow Ups:
Only if one has speakers where all drivers are connected in the same polarity. 80% of the speakers aren't, so why bother?
Klaus
Because 20% of the speakers are . Including mine, for over 30 years. So, half of my record collection sounds better, for free.Why do you bother . . . with audio?
20% is a figure based on the 40 speakers I looked at. I'm sure that if you looked at all 350 or so speaker reviews/measurements in Stereophile you'd get a figure much lower than that. Simply because there's not many speaker designers who care about that. Thiel and Dunlay were such designers.Klaus
all there is to your argument that we shouldn't bother about polarity? Because not everyone has speakers that will reveal it?That's really all you've got?
Remarkable.
you finally got my point, almost. I did not that thay YOU shouldn't bother, I said it's quite understandable that the recording industry doesn't bother, and that for decades. For the rest, there's still not enough solid evidence that polarity inversion is`audible on properly designed speakers. So if polarity inversion is audible with sub-performing speakers only, why should anyone should bother at all?
Klaus
Actually for years at least the Japanese did bother. As Clark Johnsen has pointed out, almost all vinyl and at least the early CD's from Japan alternated absolute polarity every track change, and reversed the order on the flip side or half way through the CD. Why? So every system would be able to hear atr least half of the recording in correct polarity?
Obviously somebody over there thought absolute polarity was significant as that took considerable work for something "that is almost imperceptible to most individuals."
Not the case.It's obviously also inaudible to sub-performing LISTENERS.
wrote Clark in his book. Before making such statements, maybe he, or you, should first determine whether you can hear polarity inversion under controlled conditions with adequate gear.
Klaus
"Adequate gear"Like yours, I presume :-)
If yours are low distortion and time-coherent, feel free to use yours.Klaus
I'll continue to listen (and switch the speaker cables when the polarity is obviously "wrong" to my ears).You continue NOT to listen and believe what you like. It's a dogma thing for you, as should be obvious by now.
Deal?
BTW, a primary reason why polarity is so easily detected on my speakers is that they are (duh) polarity-coherent AND there is no crossover between the mids and tweeters and only a minimal one between the woofer and the mids. Low distortion and time-coherence are only part of the story. As the tech rep responded when I queried Gallo about why I could tell the difference so readily, the Ref 3s are "extremely sensitive" to polarity, sez he. Would that it were not so. You think I LIKE being able to hear this?
. . . to those who don't even bother to listen. But I have yet to see convincing scientific evidence of that.
In regards to the recording industry:The pop recordings often mix polarities for the sound effect. Aphex Aural Exciters invert phase, add a bit of EQ in order that singers have that husky voice. Compare early Linda Ronstadt (Stone Poneys) to current material: same as Barbara Streisand and many other artists including Neil Diamond, Michael Crawford, etc. Phil Spector's 'Wall of Sound' is having the background instruments recorded out of phase to the singers in order to give that big, but vague, soundstage.
Check out The Commitments director's cut's comments. He states in no uncertain terms that the instruments were played back out of phase in order to accentuate the lead singer's voice. The LP soundtrack is very different sounding from the soundtrack on the movie itself.In the case of Classical music, polarities become a mixed bag with the advent of heavy multimiking. The mikes are very close to the wind instruments because they are very few in numbers. They are further away from the string instruments because of the larger ensemble. The difference in spacing coupled with the fact that the mikes are mixed in real time, leads to phase and polarity anomalies. You can't have an oboe miked 4 feet above the player and then mix the sound with the violins which may have the mike 15 to 2o feet away from the players.
Even in the older period where the mikes were minimal, polarity was an issue. Ever notice all the Decca recorded RCA's are inverted in respect to the American recordings? Most Decca, Philips, and DDG recordings are inverted, as well as US Columbia's. One cynic told me that was because Philips wanted to make sure no one would ever blow their speakers from too much volume.The fact of the matter that most European labels are seemingly inverted to, say, a Mercury, seems to point out problems in recording convention. This is quite evident if the Decca master tapes were sent to RCA and played back on American equipment.
Early mono Ampex reel to reel machines had pin 3 as the hot on their XLR's. I have RCA LP's, say like the Reiner Scheherazade, where the 1s/1s pressing is phase correct, but the 1s/7s has the 7s side inverted. This points to equipment issues as there is no reason why later masterings should have their polarity changed.
The answer is not to accept the status quo as Clark has been stating. Education and awareness can create change. But, if no one else cares, there will be no reform.
Stu
Here you are describing phase issues, not polarity. This is one of the reasons a heavily multimiked classical recording has always sounded bogus: the sour, phasey comb-filtered sound.Some companies (such as DG) are going back to their multitrack masters and applying time delay to the spot mikes to bring them in time with the stereo pair. (DG can do this as they have always made a "map" of every mike's location for every recording they made.) This makes a huge improvement. I wish we could take the technique to other companies' output!
That's interesting that DDG would apply time correction for their recordings. My understanding is that most companies mixed their recordings down to only a few channels on their masters. It is true on the digitally recorded masters utilizing hard drives, it is possible to preserve the multiple channels of information.
Since you are so familiar with the DDG process, you surely must have access to deeper information about their methodology? I would be very interested as I do not hear such a correction being used by, say, the Ondine (ex Philips) engineers.
I would also suspect that such methodology would not be applicable to the very early DDG recordings where the master tapes were essentially mixed in real time.You can all it what you want, but I tend to lump most time issues as being phase and/or polarity related. What may be initially a phase issue manifests itself as a polarity issue on playback. The real issue is the timing of the notes which reaches the listener's ear. When they are not consonant to what occurs in real life, they are simply wrong. Call them what you will, but the reality is that when played back on a stereo pair of speakers, the information is confused and does not represent the true tonality and articulation of the instruments (or vocals) being played. Any human does not hear sounds simultaneously from two or more spots up to 50 feet apart
If Decca could get the vast majority of these issues cured with their well known use of the Decca 'Tree' and still have the use of spotlight mikes, I can not see why modern 'engineers' can not do the same. The mike positions in sites like Kingsway Hall were very well documented.
I have yet to see a mixer which offers time adjustment for more than two channels. I see polarity switches on many, but very few real time mixers with digital time delays. I have not kept up with the latest gear, however, and I would welcome information to the contrary.
Stu
"That's interesting that DDG would apply time correction for their recordings."It's one of their reissue series the name of which is escaping me right now. But I've heard before-and-after on two or three of them, and the difference is not subtle. They do increase the prominence of the stereo pair as well, but the main thing you hear is the absence of comb-filtering.
"My understanding is that most companies mixed their recordings down to only a few channels on their masters."
Your understanding would be wrong. There are plenty of multitrack digital tape machines out there and have been for a long time. The last time I did a Pops recording I stopped counting after 38 microphones, and it was clear that the mixing and balances were to be done later.
"Since you are so familiar with the DDG process, you surely must have access to deeper information about their methodology?"
Well, going by what DG says, it's quite simple: they simply plot the distance between each spot mic and its instrument(s), then the distance between the stereo pair and that/those instrument(s), calculate the difference and apply time delay to the spot mics so the arrival time for each is the same. I know that Denon used to do this in real time at the recording session for their multitrack projects, but I'm unaware of any other company doing it.
"I would also suspect that such methodology would not be applicable to the very early DDG recordings where the master tapes were essentially mixed in real time."
Obviously.
"You can all it what you want, but I tend to lump most time issues as being phase and/or polarity related."
Obviously "time issues" will be either/or, but why not be precise? I'm not calling it "what I want," I'm calling it what it is .
"What may be initially a phase issue manifests itself as a polarity issue on playback."
No. Polarity implies the relationship between 0-degrees and -180 degrees, either/or. Phase refers to any number of conditions in-between. Since the thread is nominally about polarity (actually it's about KlausR's preference for "scientific evidence" to the evidence of his own ears, but never mind) it makes sense to call it by its rightful name.
"Call them what you will, but the reality is that when played back on a stereo pair of speakers, the information is confused and does not represent the true tonality and articulation of the instruments (or vocals) being played."
Correct, but you can have a recording that is being played back with the correct polarity yet still has jumbled phase issues due to multimiking. See why precision is important here? Most folks are confused enough about these things that we ought not add to the problem, and clarity of expression helps.
"Any human does not hear sounds simultaneously from two or more spots up to 50 feet apart."
No, but neither do humans hear sounds from 12 feet apart from a stereo pair of omnis a la Telarc, nor do we hear them from zero feet apart as with a coincident pair. Perhaps the closest we come to the reality of what we hear is from an ORTF pattern, but that method has other issues, primarily a lack of low bass response.
Neither do we hear sounds from an orchestra from 10-15 feet away--where stereo pairs normally can be found--but that's because microphones don't process out the excessive reverberation that human brains do at our seats in the hall. This gets complicated fast, and it's worthwhile not generalizing excessively.
"I have yet to see a mixer which offers time adjustment for more than two channels. I see polarity switches on many, but very few real time mixers with digital time delays."
I presume DG worked with separate units. And while the usual mic positions for many recordings may well be known, they cannot be known exactly unless the company was as anal as DG was when they made them!
The presence of polarity switches on mixers is largely to spare the engineers from having to track down the odd miswired mic cable in the middle of an expensive recording session. It also allows them to do things like mic kick drums or snare drums from both sides--a common practice in the studios--flipping one of the inputs' polarity.
From what I've understood from psychoacoustic literature, but then I didn't look and read specifically about that, is that resolution of human hearing is about 1 ms, which relates to the circumferential distance between the ears, so that arrival time differences smaller than that wouldn't be detectable.I'm not sure whether or not that is true for all frequencies. In any case, many speakers have group delays way above that value so from that point of view alone they can;t be considered as accurate.
Klaus
Well, obviously the industry does, both on recording and on playback side, maybe because there is not enough convincing evidence (normal music played through low-distortion, time-coherent speakers in domestic listening environment). I agree, keeping polarity the same throughout the recording procedure is being accurate to the original signal, it doesn't do any harm so why not have a polarity standard.But by the same token one could speaker manufacturers to build time-coherent speakers only. The difference is audible, albeit subtle, and time-coherence is the only way to accurately reproduce a signal, it doesn't do any harm, so why not?
Apparently no one cares, When Dr. Heyser was president of AES, that would have been the very moment for Clark to try and get a peer-reviewed paper on the subject published. Did he ever try?
Klaus
"When Dr. Heyser was president of AES, that would have been the very moment for Clark to try and get a peer-reviewed paper on the subject published. Did he ever try?"No he didn't, but he can see very clearly that Klaus's usual research acuity is operative here.
Dr. Heyser *was* elected president, but unfortunately he died before assuming office.
Good try on the attack, better luck next time.
clark
I'm generous, you get the point.Now tell us what the score was in Lipshitz' DBT where musical excerpts were used.
Klaus
Instead you demand of me that I go off to Canada and beg to look in Stanley's lab books.Ridiculous!
clark
In his AES paper on midrange phase distortion Lipshitz cites an Audio Amateur article (Muller, TAA vol. 11, p.64 (1980 Jan.)) where the results of that two-part DBT (test tone + musical excerpts) were describes (at least that's what I assume).Now the questions:
Do you have that article?
Are the number of runs and and the scores for both parts of that experiment mentioned?
If your answer to both previous questions is yes, will you tell us that score? If not, why not?
Klaus
and usually it works with those folks. So, as usual, I have to do it myself, find the answers that is.I'll keep you posted.
Klaus
Remarkable. A speaker without the jumbled phase issues of many commercial speakers is improperly designed? Sub-performing? Ya learn something new every day.By the way, there is plenty of scientific peer reviewed evidence for the audibility of polarity. You just haven't done your homework. Why should anyone pay attention to you?
Feel free not to!!!Yes, there's plenty of scientific peer reviewed evidence for the audibility of polarity, I know that, but artificial test signals like asymmetric clicks are hardly what one is listening to. What I'd like to see is plenty of scientific peer reviewed evidence for the audibility of polarity where normal music material is used played through low distortion, time-coherent speakers in a domestic listening environment. If there is any and you know where to find it, feel free to post a list of bibliographic data.
Klaus
Music waveforms are asymmetric, too. But one does not need a scientific test; polarity is plainly audible on music of all kinds, with all sorts of purist or multimiked recordings."If there is any and you know where to find it, feel free to post a list of bibliographic data."
See above. Otherwise, do your own homework.
Meanwhile, if your speakers do not allow you to hear polarity, why not just relax and enjoy your music? Much more fun than telling others what they can or cannot hear. Or making wild leaps of logic about there being "insufficient evidence."
...the science there? I mean, to determine what's "solid evidence"? What are the criteria? Where are the papers in peer-reviewed journals that establish a basis for "solid"? You're a "scientist", so tell us where to look.Or does it just come down (again) to your opinion masquerading as authority?
clark
nt
z
distortion amps or preamps, nor for any 'high performance' audio gear? If you are not interested in maximizing the performance of your speakers, then why bother with any other component?The situation reminds me of an incident I had with a local music educator. While a very good player himself, his band was extraordinarily both out of tune, and imprecise in its performance. When I commented about that, his reply was that no matter how hard he tried, some one would always be 'wrong' and if most of the players were playing imprecisely, the single 'off' player would be more obvious. His solution was to have them all play imprecisely, pandering to the lowest common denominator.
If you won't care, or don't care, take your choice, then a Bose Acoustimass system ought to be more than good enough. Still, I have applaud the fact that you did the research and have come to the realization of the extent of the 'faulty technology' being pandered to the public. We demand so much more from an electronics designer than from a speaker designer, but reviewers do not point out very obvious faults in speaker designs. The 'numbers' of many speaker designs, if applied to electronics, would have that designer laughed out of the business.
And yet, we turn a deaf ear to these issues. No wonder audio has been in a quagmire, and those know of and do not complain about factors like polarity are themselves responsible for the very mess we are in.
ranting again,
Stu
s
Is that good enough?
Tell us?
Klaus
HiYep, no physical speaker looks like that (essentially perfect) and the minor pre/post ring on the impulse is a trait of a FIR filter, often used for this kind of correction.
My guess would be you have fixed the response down to about 15 – 20 Hz?What speakers are these and what dsp are you using if I could ask?
Also, how did you measure the speakers to derive the correction?I am working on an active version of the SH-50 which has FIR filters and am looking forward to hearing what they sound like fixed that way and tri-amped.
You can do a lot with DSP and I am about as far as I can go with physical placement and drivers parameters given what the horn has to do.
Best,Tom Danley
They use FIR filters, response is flat ± 1.5 dB down to 30 Hz, -3 db at 27 Hz, on demand they can go lower with decreased max. SPL.More here
Klaus
HiThis would appear to be a DSP corrected loudspeaker yes?
Tom
z
Did you see that from the graphs?Klaus
c
nt
z
Over a year ago, I was stating what Klaus has written, but you argued that since recordings were 50/50 in their polarity (not something I said or has been true in my experience), that basically, polarity of recordings and speakers were not an issue and nothing needed to be done.Acceptance of the status quo was what you preached, and the most that you advocated was simply an addition of a polarity switch on a preamp. When I pointed out that many speakers have drivers in mixed polarities, you seemed to be oblivious of that fact. I honestly thought you were simply pandering to the interests of the mainstream speaker manufacturers,
What has brought about this change in view?
Stu
| ||||||||||||
| ||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
