 
|
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 |
In Reply to: Bad choice of words but *YOU* know what I meant... posted by David Aiken on September 11, 2003 at 12:27:39:
I don't think that recording staff is playing with polarity since they don't know what gear the end use is using. Some amps invert polarity, others don't. Some speakers have their drivers wired with different polarities, others don't.Vibration tweaks : I have difficulties to believe that vibration has an effect whatsoever on electronic components. Has this been proven in blind tests ? I further believe, please don't feel personally agressed, that tweaks are inherently subject to self-deception.
As you say, perhaps other factors than polaryit alone are involved when detecting polarity inversion. Methodological testing would appear approriate.
Klaus
Follow Ups:
"vibration has an effect whatsoever on electronic components. Has this been proven in blind tests ? I further believe . . . "I believe, I believe, I don't believe . . .
Klaus, this isn't theology. Like some religious zealots, if something doesn't comport with your belief system you do your best to deny its existance.
If I had a nickel for every time one of my "beliefs" was shattered by the evidence of my own ears--and others whose ears I respect--I'd be rich.
How about putting aside what you believe--just for experiment's sake--and try messing around with stuff? Fool around with cones or aurios or squishy pucks or something. Lord knows you won't fall prey to self-deception with your attitude. (Except that it works both ways. Careful!)
And you just might have more fun.
I believe I'll have another drink.
regards,
Mark
He said "I further believe, ..., that tweaks are inherently subject to self-deception".I love that old one which implies we hear what we want to, when anyone who has done any tweaking for any length of time knows damn well that we hear what we don't want to hear or something different and quite unexpected at least as often as we hear what we would like to hear. That criticism simply ignores the facts but then, if they can't mount a convincing argument against the claim some people simply resort to ignoring the facts in order to attack the perceptions they don't want to address.
It also ignores the fact that we trust our hearing and other perceptions all day long in other situations where we are just as prone to perceiving what we would like to but we don't. Our perceptions are, by and large, reliable and we are justified in trusting them. We simply couldn't interact successfully with the world if that were not the case, so what is so special about audio that it totally destroys the credibility of our senses? Yes, we can and do make mistakes but nowhere near as often as some would like us to believe. We usually discover our mistakes in time also and are prepared to admit them, which is carefully ignored.
It also ignores the fact that those who tell us that we can't be hearing what we hear because we are fooling itself and there is no difference in the sound said exactly the same thing initially about things like IMD and jitter and ...(you name it)..., all because they didn't have an accepted theory. Then along came the theory and the tests and all of a sudden the thing that absolutely couldn't exist and was just our imagination is alive and well and a routine part of THEIR knowledge. Makes you wonder whether those who know it all never make mistakes because they know it all or because they have no memory of their last mistake.
You really can't do anything about that mindset so why bother trying?
David Aiken
Listener bias has been shown to be a factor that has to be taken into acount and that since it affects the results it has to be reomved from the experiment in order to have sound as only parameter to judge on. Bias in the sense that the subjects know thye identity of the gear or that they simply know where the gear is placed (for source location experiments).Facts : if you are telling me that you hear a phenomenon whatsoever or that a particular tweak changes sth in sound or image I have no problems in believing that.
Theory : manufacturers of accessories want to sell their stuff, so often THEY come along with some theory explaining in length why this and that is a problem, because audible, and that the item in question solves it. Often the theory does not withstand close inspection and NEVER proof is given for the audibility : listen yourself and you'll hear !
You say that I tell you that you hear what you want to hear. Now you may tell me that I don't hear (polarity for that matter) because I'm not convinced. We will never know who is right until evidence other than of anecdotal nature is provided.
Sorry for bothering yo all with a different mindset.
Klaus
You were the one who stated that "tweaks are inherently subject to self-deception" . I assume you meant all tweaks because you didn't say some tweaks, and you classified them as 'inherently subject'. As I said, what makes them any different to any other instance where we're forced to rely on our senses and we have something we want to believe? Why do I not mistake every person, or even most people, turning into my street for my wife when I'm waiting for her to arrive home? Why do I not mistake people's voices for the person I'm waiting for when someone behind me says "David", after all it's a common name and is said often, but I'm rarely confused or mistaken by what I hear in that sort of case. Why are tweaks problematic in ways that other things we use our senses to determine, and in which we have an interest in the results, are not problematic?The simple fact is that your generalised statement is neither accurate or scientific. It relies on the fact that some tweaks don't do what is claimed as a basis for the generalisation that all don't. That isn't a sensible or scientific position. If you want to argue about one particular tweak, then fine. Do so by addressing the issues and the science but not by making generalised statements that don't address the issues. If you want to make a generalised claim about 'most tweaks', then produce some evidence that more than 50% of tweaks (that's the standard definition of 'most') either work or don't work or whatever your claim is.
In other words, be honest and scientific. Don't make sweeping generalisations and do provide some evidence for your specific assertions. Then we can have a rational and even scientific discussion and may even find some common ground, even if we agree to differ on a specific issue at the end.
And don't assume that because the explanation given for an effect is wrong, the effect does not exist. Science is riddled with a history of effects being explained by theories that were eventually proved to be wrong. Take a look at medicine and the theories that were expounded for the causes of disease before we discovered bacteria. Viruses were discovered much later, in the 20th century, and people had difficulty accepting a non-bacterial cause for illness. Look at some of the early explanations of electricity, magnetism, gravity, and many other phenomena. They're still refining our theories of these things and there is no doubt that some of the things we now believe will be proven wrong, either in part or fully. There are things we don't yet have explanations for.
There is no reason to believe that currently accepted theory in audio-related science is complete and absolutely correct. Just because someone gives a totally erroneous, illogical and unscientific reason for why a change they believe they have observed occurs does not mean that the change didn't occur. I agree that more than a few of the explanations given for why some tweaks work seem to be totally unfounded, but that doesn't automatically mean that the tweak does not work. Getting the explanation wrong doesn't necessarily and automatically mean that the observation was wrong.
On the other hand, the fact that the state of our current knowledge isn't perfect does not mean that anyone can claim that anything goes. Still, science starts with observation and if a large number of people are making the same observation then there is the possibility that there is something in the observation regardless of what current theory says. It's much easier for one person to make a mistake than for 10 to make the same mistake, and it's easier for 10 to make a mistake than for hundreds or even thousands in the case of some of the claims being made for differences in audio. If the people making the observation are wrong, and that is a possibility regardless of how many make the claim, then it's worth while finding out why the error is being made. There are many more reasons for possible error than observers simply deluding themselves so it's well worth while considering them also. It's very easy to write off those whose views you don't agree with as deluded but the chances are that if they are wrong, it's for some other reason than the fact that they are deluding themselves and there is always the possibility that the person in error is you.
As far as I know, there has been no attempt to prove whether people making errors in observation in relation to tweaks, either specifically or in general, did so because they deluded themselves or because of any other possible reason for observational error, yet those who wish to discount tweaking at a general level seem always willing to assert that whatever effect being claimed is simply an observational error due to delusion. Without some effort to examine the claims and to determine te actual reason for error, such a claim is simply an untested and unsubstantiated opinion with no basis in fact.
For what it's worth, I do have professional qualifications in health and safety. Research I did while undertaking my studies was published in a peer-reviewed journal and while it was nothing major as far as research goes, at least it shows that I do have some understanding of scientific issues and the research process. I'm making no sweeping claims for all tweaks nor am I claiming every tweak works. What I am saying is that generalised statements like the one you made are perfect examples of what one cannot logically or scientifically claim.
There is no evidence of which I am aware that shows that there has been any attempt to determine whether people make mistakes more often than not in relation to claims about audio tweaks, or that whatever errors are made are due to people deluding themselves. If you have any such evidence, I'd be quite happy to read it. If you have no evidence and you are still prepared to make such a claim, then you are making just as fatal an error as any individual making a mistake about a tweak.
David Aiken
Aural memory has been shown to be "volatile". You can't rely on it when you want to judge subtle differences.Science 1965, vo.150, p.1625 : Bindra :Judgements of sameness and difference : experiments on decision time
JASA 1999, vol.106, p.2805 : Clément : Memory for pitch vs memory for loudness
AES preprint no. 4585 : Subjective audio testing methodology and human performance factors
You are referring to or senses we are using every day. Take two sheets of paper with different whites. If you don't hold them side by side you can't tell that they are different. If you take one first then go to the other end of the room to look at the 2nd you won't be able to tell. Of course you will notice that the 2nd sheet is also white, just like you will recognize a familiar voice behind you, but you won't be able to tell whether the white of the 2nd sheet is identical or different.
As I've said elsewhere in this thread, listener bias has been shown to affect the test results and has therefore to be removedPerception
1979, vol.8, p.323 : Warren : Spatial localization under conflict conditions : is there a single explanationAES preprint no.3893 : Toole :Hearing is believing vs Believing is hearing : Blind vs sighted listening tests and other interesting things
Which leads us inevitably to blind tests. As long as people judge on tweaks using sighted testing and before-after comparison (just think of the famous audio reviewers' memory, which works over months or years) I think that there are good reasons to question the results.
And even if a tweak can be shown to work, the (positive) result is valid only for that particular audio system, that particular person, that particular test environment. You can't generalize from a test, be the results positive or negative.
I did not try any tweaks whatsoever because, given the above reasons, I would question the results of my own tests.
Coming back to the polarity issue, 2 more points :
1. in order to be sure that it's not due to other factors that polarity inversion can be heard, like distortion or time coherence, these other factors should be checked in systematic tests.
2. polarity inversion in stereo : if right and left speakers are not pair-matched within limits below threshold of detecetion, say 0,5 dB, wouldn't you think that it is possible that the combination of factors as mentioned above and the bad pair-matching could be held responsible for the audibility of polarity inversion (changes in stereo image, focus etc.) rather than polarity inversion on its own.
What about group delay : Clark Johnsen did not report on that parameter. His minimum phase speakers (hence not time coherent) possibly have a group delay so imoportant that polarity inversion is made audible.
If you look into journals like JASA (Journal of the Acoustical Society of America) or Acustica you will find that there are not many papers relating to audio directly, audio standing for listening to music. Most audio related papers are to perceptional issues, leading to better understanding of our hearing. Some of these results will obviously find their way into say speaker design (like more thorough knowlegde of the precedence effect), but that's not the initial intention of the research.
Klaus
I've made more tests with my speakers in minimum phase mode, both with CD and LP. Today I'll try and re-activate the small passive system. More to come.
Klaus, if aural memory is too volatile to be trusted in a standard audiophile-type A/B test--that is, you play a selection with one gadget and then play it again with another--and can only be trusted with rapid (instantaneous) switching, please tell me this:How did you select the components for your system?
regards,
Mark
I had been convinced by some literature (Martin Colloms, Meridian, other) that active speakers are superior to a passive system. I was further looking for speakers having high directivity as to decrease the influence of the listening room. I could not find such speakers among the hihi/high-end stuff. I could find such speakers among the pro-stuff. I did not perform what audiophiles would classify as auditioning.BUT, my speakers have a 10-band digital equalizer, if I need differently sounding speakers I grab the remote and the user's manual and change.
Preamp : I am of the opinion that if component measures well it very likely is accurate. So I checked out specs of several preamps and, surprise, a pro-preamp won the competition. Since both speakers and preamp are pro-stuff with standardized inputs/outputs I did not worry too much about possible interface effects, this includes the wires as well (miniature microphone cable).
Turntable : I bought the one about which I could find the greatest amount of information.
Tonearm : you need one so I bought one, not too expensive, not too cheap.
Cartridge : technical literature (again) convinced me that the cartridge I bought is the best (IMO that is).CD : pro-stuff
tape : Nakamichi
MD : pro-stuff
Conclusion : 100% non-audiophile way of proceeding but the result is overwhelming.Klaus (I bet that you are not surprised at all)
.
I could consider buying "standard" components without auditioning as taking a risk, certainly when you are an audiophile. BUT, since my speakers allow to produce any desired sound and timbre, why worry ? I can adjust the sound to my needs if so I wish. No problem. sir.For instance, I just learned that my hearing has a small gap at 6 kHz. I simply go home and increase amplitude at 6 kHz by 5 dB !! At other frequencies hearing is better on one ear than on the other. My speakers allow to adjust for this as well.
Buying SOTA has some advantages, you know :-)
Klaus
Yes, I am referring to senses we use every day. I listen to music on my system using the same ears and brain I use elsewhere for all listening tasks I undertake.And yes, I do admit that I can make mistakes, and even have done so on occasion. On the other hand, I do prefer to try various things and trust my perceptions enough to make decisions on them. I choose between different speakers or amplifiers or sources based on listening tests, often conducted in different stores and not always with the same system because that is sometimes the only way I can do it. That introduces even more variables than just simply aural memory. I spend a fair amount of time and care when I have to do it that way because it's definitely not the way I would like to do things, but it does get me down to a short list and then I can sometimes arrange other options. I make bigger decisions in terms of both long term listening and cost and no one questions them the way they question what I claim to hear with a tweak. THAT makes me wonder.
In the end, I listen to music at home for pleasure and I rely on my ears to help me get a system that provides pleasure. I'm happy to experiment and some tweaks have added to my pleasure, more than a few have not. At the end of the day, I'm listening to my music and I get more pleasure accepting what I hear than spending my life worrying about whether I heard what I think I heard. I trust my ears to tell me when the sytem isn't working properly and when it's working well, and that relies on aural memory too to a degree. No one questionss that either or says I'm deluded when I make a mistake about whether the gear is working properly or not.
From my perspective there are 3 big problems with your position. The first is the automatic assumption of delusion you've pushed up until now as the cause of error, without consideration of any other sort of error including aural memory. If the problem is aural memory, then the listener has made a mistake. Delusion is a different process entirely and it doesn't do anyone any good to label mistakes due to failure of aural memory, or any of the other sorts of error that can arise, as delusion. That isn't saying that delusion doesn't occur, but it is saying that not every error is a result of delusion and calling it that automatically, as you did, is just simply wrong.
The second problem is your expectation of error at virtually every step of the way. It simply doesn't make sense to doubt your abilities in relation to tweaks and not question them elsewhere. Aural memory applies as much to the recognition of a familiar voice as it does to distinguishing between the sound of a system with and without a tweak yet you are no doubt quite happy to rely on it for voice recognition. On the other hand, if you want to just doubt your ability to rely on your hearing everywherre, go right ahead and double check everything you think you hear and then double check the double checking. I just think life's too short to allow that kind of doubt to get in the way of a pleasurable activity.
It's also worth commenting that there is research which indicates that not all aural memory is 'volatile'. If it were, then voice recognition would not be reliable and it is reasonably reliable as we all know. There is research I've heard about to back that up. There is no reason to believe that there aren't some aspects of our system's sound that are sufficiently familiar to us to be stable enough for comparison purposes and, since you mentioned blind testing, the best blind testing protocols allow the listener control over when to switch. Aural memory has to be stable enough to allow for whatever listening period the listener wants prior to reaching a decision or many blind tests would be invalidated because a failure of aural memory would render the results unreliable.
Finally, my third problem is simply the fact that I think there's something really wrong when a person who is spending a fair amount of time, effort and money with audio equipment in order to listen to music for enjoyment isn't prepared to trust their perceptions. That always makes me wonder whether a different hobby insn't indicated.
The problem is not that there is always a chance of error. The real problem is that the assumption that there is virtually no chance of being correct is defeatist. It results not only in people not trying tweaks, as you have indicated you don't, but also assuming that absolutely nothing in the area can be trusted. But if you really can't trust your ears on differences between tweaks for any of the reasons you raise, then you can't trust them on listening tests to identify differences between speakers or any other components. The baby gets thrown out with the bathwater.
David Aiken
I would classify self-deception not as error in itself but as a consquence of error/s. Error no. 1, aural memory. I would assume that there is a big difference between recognition of someone's voice and a subtle difference between signal A and signal B. I would therefore further assume that aurla memory is not reliable for small subtle differences. Maybe that this has been shown in psychoacoustic experiments on aural memory, I don'tknow. At least the literature on blind tests and aural memory I've read sofar do suggest the above.I can set the lower cutoff-frequency of my speakers to different values. If I set to 40 Hz instead of 30 Hz, you will notice the difference immediately. If I set to 30 Hz, you listen, you then do sth else for a minute or so, you then come back and listen, you very probably won't be able to tell whether it is now set to 30 or 40 Hz. The difference between both settings is audible on direct comparison but it is very subtle.
Error no.2, expectance effect. If you compare a Sony amp to an Accuphase you will, a priori, assume that the Accuphase is better. Same is valid for tweakers and their tweaks : you put effort and/or money into it and you'd like to have positive results, otherwise, why tweak ? Remove this expectance effect by testing blindfolded and things may be different.
Blind tests use short music samples and instantaneous switching (because of the volatility of aural memory, see also http://www.tnt.uni-hannover.de/project/mpeg/audio/public/w2006.pdf ). The only parameter to judge on is sound.
I do trust my aural perception but I don't trust it any more when aural and visual (or other) factors are mixed. Blind tests remove all these others factors. Since sighted listening IMO is basically flawed for decision making and blind testing is not possible, i choose another way to select gear. As you said, different mindsets.
Klaus
In normal use of language, we make a significant distinction between self deception and error.We are deceived when we are lied to or given misleading information. The term 'self deception' recognises that the person lying or providing the misleading information is none other than ourself, and there is an implication there that we are doing so because it suits us to believe something wrong. This is the 'hearing what you want to hear' accusation that is used so often against people claiming to hear certain things, not always the results of tweaks. It was said of people who claimed to hear differences between amplifiers which measured the same prior to the identification of IMD, and of people who claimed to hear differences in some digital components before jitter was identified.
On the other hand, when we say someone made an error or mistook something for something else, we recognise the fact that our senses are fallible and that we do make mistakes for quite understandable reasons, and we do not imply that they did so because it suited them better than seeing or hearing the real facts. We don't say that someone who made an error deceived themselves unless we want to say something quite different to the simple attribution of error.
Self deception is not simply a 'consequence of error' - at the very least it is something that predisposes one to making an error where one normaly would not and it often is the actual cause of error. Self deception implies motivation whereas simply making an error is an unmotivated act. That distincition is the reason the two descriptions of what went wrong are not interchangable in common usage.
If you want to talk about 'honest' errors such as those caused by aural memory, then fine, but self deception is not due to failures of aural memory - it is due to a desire to fool oneself. You chose the term 'self deception' and set the tone of this discussion by doing so. I repeat once again my assertion that you have absolutely no evidence to substantiate your claim that "tweaks are inherently prone to self deception". You have no reason to believe that most people who try a tweak either consciously or unconsciously want to believe that are hearing something different than what they are actually hearing. Yes, they may want to change the sound of their system to something they will enjoy more, but most of our actions in life are intended to improve our situation in some way and we don't go through life believing that everything we do works perfectly and improves our situation. Once again I have to ask what distinguishes tweaking from any other activity we engage in as human beings and why is it more prone to self deception than all those other activities?
As a final point, you said "I do trust my aural perception but I don't trust it any more when aural and visual (or other) factors are mixed". Fine, but it sounds like you actually trust it less when other factors are mixed. Yet, in normal life where we make many quite fine and usually reliable distinctions regarding what we hear, visual and other factors are mixed. We don't go around automatically closing our eyes and shutting off other input in order to make our hearing more reliable. Sighted listening is not normally flawed for decision making though it may be in some cases.
It seems to me that if the way we use something is by listening to it, then the way we judge it can only be by listening. If most of that listening is done sighted, then sighted listening tests are reasonable. Errors are always possible but if they were common we wouldn't trust our senses as much as we do because we would know from experience that they are highly reliable. We do trust our senses because errors do occur but aren't common enough to cause us to regard our senses as unreliable. I'd rather make an honest mistake by listening than decide on something I'm going to listen to by avoiding listening and relying on other information.
David Aiken
.
x
in particular aural memory. As long as I can't A/B a tweak side by side, rather than before-after, I'm afraid that messing around is useless for me.A German manufacturer demonstrates on his website the nasty effects of vibration of components' housings (Burmester in that case) and subsquently the beneficial effects of his shelf, which btw. is patented, not that this fact is of any importance.
To start with, if Burmester components are suffering from housing vibrations, it would be Burmester's job to get rid of those, especially when considering the price for his gear.
But apart from this, if you have close look at the data presented, you find minuscule peaks of the housing plates (in the micron range, if I remember correctly), when playing extremely loud signals (cannon shots on Telarc's 1812 CD). It's simply hard to believe that such housing movements and the effects they cause are not masked by the actual cannon blast.
There are more examples like the above, it just makes you wonder what manufacturers think people are willing to swallow.
As for my beliefs, I think that hifi is about accurate reproduction and if there are parameters which really interfere with accurate performance of a component they should actually be solved by design, not by aftermarket tweaks. If all those who spend time, money and effort in tweaking their system, why don't they simply get an equalizer instead.
Klaus
I've got time now and will go for the 2nd round of my polarity test.
Because it feels so good when I stop.
nt
Klaus,please define 'objective'/ly' for us, would you"
I would love that.
Oh, and my qualifications are in the scientific method and its history, and complex systems. Okay?
So your answer had better be good, and grounded in real science.
Timbo in Oz
I'm speaking 4 languages and this fact makes me subject to what is called "false friends", meaning that the same word has slightly different meanings in these different languages, which is the case for the term "objectively".I should perhaps better have used the term "non-emotional" instead.
Apart from that I do find some type of language used in this thread inappropriate, because it's too personal and does not address the issue itself. Such behaviour, however, seems to be human and can be found in discussion groups everywhere, to a greater or lesser degree.
Klaus
.
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 ]
