|'); } // End -->|
In Reply to: Re: Agreement and disagreement posted by David Aiken on March 4, 2006 at 00:51:09:
"but we normally don't have the artists tell us what they intended us to hear"
But we do have the artists art in the form that they agreed was what they wanted to express to us (CD's, DVD's). Do we not owe it to the artist NOT to reinterprete their art by manipulating it simply because we find it personally more satisfying. That's like recoloring Picasso's Blue Period, because we don't like blue.
To me this whole discussion gets lost the moment one implies "preference" or the like. Thats because there is no and can never be an answer to "preference" in "reproduction". There can only be "preference" when responding to the art, and reproduction is not the art. It may be an art in achiveing it, but it, in and of itself, is not the art or even part of it.
I am very careful in all my work NOT to imply "preference" in our testing, only "perception". We can perceive reproduction differences without ever implying or judging a preference. When there is no "perceptable" difference between the reproduction and the art, then, as an audio designer, my job is done, and the "preference" judgement can be made solely where it belongs - of the art and not the playback system.
If this is "boring" to you, then you and I are seeking different things.
I know that many will take strong exception to this point of view, but believe me, it was not arrived at lightely. I am a passionate listener who over the 45+ years of listening and designing has come to recognize what the importance aspects are. I attempt to describe them here and through my work.
Well, I never said that it was right to reinterpret. I actually said it was wrong to do so. Go back and read that point, and please refrain from trying to accuse me of saying things I most definitely did not.
I did raise issues with knowing the artist's intentions and there are issues. Performances don't always work out as intended, surprises happen, and sometimes the surprises are what makes the performance. Should you reproduce such surprises accurately when the artist tells you they were unintended? Should you try to deliver what the artist intended or what they achieved? I do side with reproducing what was achieved.
And it is worth noting that not all recordings do meet with the artist's approval. Discs have been released that the artists objected most strongly to. You simply can't assume that the recording documents the artist's intentions.
I spoke strongly in favour of accuracy as an equipment design goal and I meant that. I also have yet to find any component that does that. I said I was uncertain that delivering such a goal would satisfy the listener, though I also said I was prepared to find that I was wrong on that point if we ever really reproduced the wave form. It's impossible to say how I would react to hearing something no one has ever heard, but as I said, I think listening to recordings is a different experience to listening to a live performance. We can and do attend in different ways, so why shouldn't we find that what would genuinely satisfy us when listening to recordings is something different to what satisfies us when we listen to live music. And as I said, I've listened to both live and recorded and I've played publicly as an amateur artist. I do have a bit of experience as a performer as well as a listener.
As far as accurately reproducing the waveform goes, which waveform do you want to reproduce for a live recording? The 5th row or the 10th row? Is it re-interpretation to arrange your reproduction so that what you hear is very close to what you heard at the actual concert, even though the mics were located elsewhere?
What I am certain of is that I do think that there are issues about accuracy and faithfulness to intentions which aren't necessarily as clear cut as you try to make them seem; and also that trying to achieve a goal that our equipment currently does not support and always falling short does not seem like a recipe for contented listening to me.
"Well, I never said that it was right to reinterpret. I actually said it was wrong to do so. Go back and read that point, and please refrain from trying to accuse me of saying things I most definitely did not."
Don't be so touchy, I was not accusing you of anything - simply asking the question. I was responding as much to the general list of comments as to yours specifically. You have to choose one to respond to and yours was the most sensible. So don't take offense.
"As far as accurately reproducing the waveform goes, which waveform do you want to reproduce for a live recording? The 5th row or the 10th row? Is it re-interpretation to arrange your reproduction so that what you hear is very close to what you heard at the actual concert, even though the mics were located elsewhere?"
I am not a waveform accurate advocate because its not perceptually important. I advocate playback accuracy of the recording as mixed by the engineers, which is two channel. (And I know the argument about the monitor loudspeakers and their problems. Once all loudspeakers are accurate then this won't be a problem anymore. ) If the release does not "meet with the artist's approval" then I am sorry for the artist who will release inferior work - I can't help that. Your last question is only applicable to a rather unique circumstance - its hardly ever applicable (since classical music accounts for such a small % of sales), but its still what was mixed and approved that counts.
I will say this, I have a good friend who has released many CD's. He asks me what I think of them and I point out all the flaws. He never even heard these flaws. All he knows is the performance (which was perfect) not the recording (which was not). But he is the last word on the release. He now has me listen to everything that he releases on my system before he releases them. So no, the situation is not perfect, but modifying the playback will, at best, improve one situation and at worst screw up everything. So I don't see how we at the playback end can ever hope to correct problems at the recording end.
"What I am certain of is that I do think that there are issues about accuracy and faithfulness to intentions which aren't necessarily as clear cut as you try to make them seem; "
I think that you are oversimplifying what I am saying as I am hardly saying that the task is easy to impliment, only that the goal seems pretty clear to me. Yes one can make it cloudy with a lot of specific examples, but when one looks at the big picture I think that the direction is still pretty clear and I think that you would agree with that direction.
"and also that trying to achieve a goal that our equipment currently does not support and always falling short does not seem like a recipe for contented listening to me."
What goal does our equipment currently not support? I find that I am very contented with my listening and I find that people like Duke, who I believe shares my point of view, are also quite contented with thier listening. I'm unclear on why you think that my goal leads to discontent.
At any rate, don't take offense, I think that we are mostly on the same page with perhaps a few disagreements on what the goal is. I think that you will find that while the issues you raise make the situation less clear, that striving for accuracy in reproduction independent of preference still ends up being the best choice overall.
I do think fidelity to the recording as it was mixed is a much more workable goal, but there are still issues.
Most studios work with a nearfield setup for mastering and most listeners seem to prefer far field. Both produce quite different experiences but I don't regard one as being intrinsically more right than the other. Some people are definitely more partisan than I.
And I'm quite content with my listening too, if that means something.
But I'm content with something less than total accuracy, as are you. None of us knows what total accuracy sounds like because it's currently unachievable, and small differences in reproductive quality can cause significant changes in our perceptions. I suspect that we all try to achieve accuracy by ensuring that our system's strengths lie in the areas that are most important to us while minimising weaknesses and ensuring that none of them are significant enough to cause us problems. Different people do like different presentations—I have a set of low sensitivity mini-monitors in a near field setup and one of my friends has high sensitivity vintage Altec Lansings in a far field setup. We both listen to live as well as recorded music and both find our respective systems accurate. My feeling is that we each get off on different characteristics of musical sound and we optimise playback for our respective preferences. In other words, we're selective about what we're accurate to.
I see no reason for believing that achieving total accuracy is going to change that. Total accuracy can be determined by measurement, but it may not reflect the 'mix' I hear when I listen to either live or recorded music.
And yes, I may be "making things cloudy" in your view but I do feel a need to withold judgement on how satisfying such a result would be. I totally agree that equipment should be accurate but people do derive pleasure from using things in differrent ways, and the goal of this hobby is pleasure. I'm prepared to bet that delivering totally accurate equipment would not result in all that much of a reduction in the range of equipment manufactured and sold, and some of today's gear is more accurate than others and there are people who buy at both ends of that range. I can't see that changing and I'm not sure that I would necessarily approve if everybody settled on the same thing, no matter how accurate it was. Contrast can be instructive and enjoyable also.
I can hardly disagree with you, but I also have some further points.
"Most studios work with a nearfield setup for mastering and most listeners seem to prefer far field. Both produce quite different experiences but I don't regard one as being intrinsically more right than the other. "
I posted this same topic as a question here some months ago but did not get much response. This is a critical point, one that I have not wholly resolved in my own mind and one that does have a strong effect on what we perceive.
Which leads me into the area that you are uncomfortable with and the one that I am most comfortable with. You seem to be unsure that true accuracy will achieve what you are seeking - fair enough, perhaps you haven't experienced it, perhaps it can never really be achieved, whatever. But what I have found (personal subjective opinion here) is that I am more satisfied with the sound the more accurate it is when judged by an absolutely objective criteria. This could be coincidence, or wishful thinking, but I ask you, does that seem likely. Doesn't it just seem intuitively like it should happen this way. What else in our universe works in any other way? How could it be that the key to good sound is locked away from us never to reveal itself through science? Maybe its the scientist in me, but this just doesn't seem logical. Most professional believe that measurements are absolutely reliable predictors of mean opinions of quality.
Now I am not saying that objective measurements tell all - hardly. Many objective measurements that I see done are pointless to useless, and they definately don't tell the whole story. To measure axial frequency response, for example, is, to me, a wholly unsatisfactory measurement. Its one measurement in a field of many and not even the most important one. And yet so many think that this is accuracy - hardly.
"I see no reason for believing that achieving total accuracy is going to change that. Total accuracy can be determined by measurement, but it may not reflect the 'mix' I hear when I listen to either live or recorded music."
And this, of course, is where we disagree, because in my experince the better something measures - when done right - the better it sounds. Your experience may be different, but I feel bad for you if it does. Because I have a clear path to follow and metrics by which to measure my progress which are both meaningfull and accurate. This allows me to move forward unambiguously without the trial and error so often associated with audio progress. I have designed an untold number of audio systems and if it weren't for the ability to accurately quantify their performance I would not be able to move forward with any speed or certainty.
The "trial and error so often associated with audio progress" that I spoke of above, is, of course, what many people are in this hobby for. I don't criticize them for that, but I do object to being criticized for seeking an objective way to proceed, since, as a professional, my livelihood depends on it. I understand, perhaps better than most, that this puts me at odds with many here, but still, some seem to appreciate what I can offer in the way of an objective point of view, so I am willing to offer them my expertise.
Nice talking with you.
"Which leads me into the area that you are uncomfortable with and the one that I am most comfortable with. You seem to be unsure that true accuracy will achieve what you are seeking - fair enough, perhaps you haven't experienced it, perhaps it can never really be achieved, whatever. But what I have found (personal subjective opinion here) is that I am more satisfied with the sound the more accurate it is when judged by an absolutely objective criteria. This could be coincidence, or wishful thinking, but I ask you, does that seem likely. Doesn't it just seem intuitively like it should happen this way. What else in our universe works in any other way? How could it be that the key to good sound is locked away from us never to reveal itself through science? Maybe its the scientist in me, but this just doesn't seem logical. Most professional believe that measurements are absolutely reliable predictors of mean opinions of quality."
So far improved accuracy has seemed to work for me too, however I remain hesitant about total accuracy.
Part of my reason for that is that I think there's a question about accuracy to what. I know people with wildly different systems and they all claim to be seeking accuracy, but different aspects of the "sound of music" appeal to them and their systems do sound different and individual. Do I assume many of us are fooling ourselves about what is accurate, including people who do listen to a reasonable amount of live music? More importantly, are they choosing the sound they get because of personal preference or because aspects of that sound strike them as more accurate in relation to what they focus on when they listen to live music? That's not a trivial question and I'm unaware of any research to try and answer that question.
The 2 of us can drive separately from A to B in identical cars along the same route. Weather conditions and time of day are the same. Do we perceive the same scenery and do we have the same driving experience? We'll probably attend to some different aspects of the scenery and give a different description of some aspects of the route so I think it's fair to say that our perceptions are different. Our driving styles will definitely mediate our driving experience. Whose report of the trip is going to be more accurate, assuming both of us report honestly? It's a meaningless question. We've both answered accurately but the trip was different in some ways for each of us.
Our perceptions are complex and they do differ from person to person. What we perceive when we listen to music isn't simply the product of the acoustic waveform in the room. There are other contributors to our perception and they are personal. Does everyone's perception correlate to the same degree with their raw sensation? I doubt it, but I also don't know how we could go about establishing it.
And, if one person's "accuracy" now is another's "re-interpretation", why should that change when we have the ability to reproduce things with 100% accuracy to the source when we don't all perceive the source in the same way?
It's possible that having total accuracy would satisfy many of us, and it may make it easier for many of us to focus on those aspects of the music that most interest us, even though there is some variation from person to person there, but I'm uncertain that it will satisfy all of us. It probably would satisfy those of us who fall within 1 standard deviation of the mean for what we respond to, however that is measured, but I wonder whether it's less likely to satisfy those at the extreme ends of the bell curve for this. There's a wide range of variation in musical tastes (classical vs pop vs jazz vs…) and many of those forms have very different sonic traits. I would expect there to be similar variations in what we all perceive as accurate, irrespective of measurements.
I always remain cautious about views that leave no room for individual differences. They tend not to work for everyone though some views do seem to work for more people than do others. There is less variation among us on some things than on others. Until we can achieve total accuracy and see how a wide range of people respond, I will choose to remain cautious.
No system will ever please everyone - lets face it. And your argument assumes that someone can define "accuracy" in a subjective manner, which I disagree with.
"I know people with wildly different systems and they all claim to be seeking accuracy"
Yes, of course, but lets face it, most people do view "accuracy" as "preference". I find that people say they want accuracy, but thats not what they buy nor what they prefer. No, to me accuracy can only be measured through objective means and then this leaves little doubt as to the reality of the situation.
"And, if one person's "accuracy" now is another's "re-interpretation", why should that change when we have the ability to reproduce things with 100% accuracy to the source when we don't all perceive the source in the same way?"
Because this question can only be meaningful if "accuracy" is ambiguous, which it is if it is defined subjectively. But when defined objectively then the above example is not possible. It either is accurate or its not, no ambiguity.
"It's possible that having total accuracy would satisfy many of us, and it may make it easier for many of us to focus on those aspects of the music that most interest us, even though there is some variation from person to person there, but I'm uncertain that it will satisfy all of us."
Of course it won't because you are talking about "preference" not "accuracy". Its very possible, and has in fact happened to me, that an accurate sound system makes a recording sound worse. That truely has to be possible right? So why should we believe that accuracy would always lead to preference? I don't believe that, not for a moment.
But lets just be clear on our terminology. If one is seeking preference then they cannot claim that it is necessarily high fidelity or accuracy. They are not and never can be the same thing. If someone wants to seek preference then go for it, but lets just be clear that what they seek is a personal thing not to be confused with high fidelity. Thats the root of most of my arguments; that people want to define accuracy as preference and this simply is not the case.
I can say unambiguously that I seek "accuracy", that this is also my personal "preference" is a seperate issue. And I am completely comfortable with someone saying that "accuracy" is not what they "prefer". But I am uncomfortable with someone saying that they prefer accuracy while the objective measurements clearly show otherwise. This is a misuse of the term.
I think our problem is that you're defining accuracy in terms of the sound produced, which is great when it comes to assessing components.
People judge accuracy on the basis of their sensations which is a different thing.
I'm unwilling to jump to the conclusion that "most people do view "accuracy" as "preference"" and I'd be very interested if you know of any studies to support that. I tend to believe most people are honest and if they say that something sounds life-like or accurate to them, that's what they mean. They're not confusing what they like with what they think is accurate. I know people who, if asked, can tell you where they think their system departs from accuracy and where they erred in favour of a preference for something rather than a different deviation. They are aware of the difference between accuracy and preference but, if they can't get accuracy they do rely on preference when choosing their unavoidable inaccuracy. I could be wrong but these are intelligent people who are honest in other areas so I'm reluctant to distrust them over this.
In fact, I can quickly think of a simple reason for someone's perceptions disagreeing with objective measurement of the sound. Let's say we have a music lover—someone whose assessments of accuracy you previously have been happy with—who has recently started to develop noise related hearing loss and neither of you are aware of this. As you are no doubt aware, this loss starts as a loss of sensitivity in the mid-range. They haven't yet developed it badly enough to bother them sufficiently to get their hearing tested, but their hearing sensitivity is measurably down in that region. They report that a system with a slight boost in that area is accurate whereas you, without the damage, hear the boost and claim it isn't. Both of you happen to be right in this case.
This isn't a "one size fits all" world, and anyone who deviates from what you regard as the right assessment of accuracy is not necessarily wrong or falsely reporting a preference.
To me accuracy cannot be a subjective judgement - we disagree here, Ok.
Your example is false for two reasons. By far the most common hearing loss is in the high frequency, not the mid frequency, but never mind that. This person still has a hearing loss when they go to a concert and so they will still find the acurate system accurate on comparison. What your saying is that they may want the sound system to correct their hearing loss. Then its no longer accurate, and they should buy a hearing aid instead, so they don't screw up the sound for everybody else.
I specified noise induced hearing loss. It tends to show up first in the vocal range frequencies, not the higher frequencies. I used to work in occupational health and safety and was required to study noise problems. I suggest you read up a little on it yourself.
It's a rude shift on your part to ignore the condition I referred to and try and imply that my example was wrong because other conditions don't manifest in the same way.
Yes, a person with a hearing loss can compare things to a live performance but, as you are well aware, live performances can sound noticeably different depending on where you sit. We build an "image" of what sounds accurate over time and, if our hearing deteriorates, that image will change also but it will change sometime later than the hearing deterioration. It will not occur along with it. That's why people get their hearing tested because "things don't sound right". If there has been a loss, their standard for what sounds right hasn't yet changed or they wouldn't think things didn't sound right. We adjust over time, and adjustment occurs only after an interval. There are periods where the adjustment has not started to be made, or is incomplete, but during which the deterioration does affect a person's hearing and their judgements about their perceptions are influenced by their former hearing experience rather than their current experience.
You say " it seems to be dwindling down to some very specific exceptions to the rules" and attempt to dismiss my points but you were the one to state "the rules" in such a categorical way as to imply there were no exceptions, and to keep arguing against any suggestion that there were. In this particular case there happen to be exceptions and your categorical statement simply isn't true for everyone as I pointed out.
You could at least have the courtesy to admit, when making some of your frequent categorical statements, that there are exceptions to them from time to time.
Enjoy your trip to Brazil.
This post is made possible by the generous support of people like you and our sponsors: