Home |
Search |
Today's Posts |
#41
|
|||
|
|||
wrote in message ...
"By contrast, RAHE seems to me to be a theological place where firm belief in DBT's is aceepted as a "truth" for all purposes, regardless of questions raised or a complete change in the nature of the item being measured....eg music instead of sound....and I might add despite any attempt by the faithful to ever verify the test's applicability." Science proceeds not by proving some thesis but by failing to disprove it. The thesis at hand is that a benchmark of listening alone testing now exists showing inability to distinguish amps, wire, etc. by listening alone even and especially by those holding the contrary view. That, as science, is what takes it outside the realm of belief system. Science to support the contrary thesis would have to show a constant and continuing failure, and thus disprove the thesis. It is bad science if it is the wrong test for it's purpose. In this case if it is unable to include the emotional and right-brain aspects of musical enjoyment in determining the difference. No matter how useful it may be elsewhere in the audio field for developing codecs, listening for specific audio cues, etc. "Sorry, I spent years in the food industry were we spent enormous amounts of time, energy, and intellectual capital on testing...much much more sophisticated than the testing you suggest. Simple blind comparative testing was use to answer simple questions..." And we have before us the simple question of can by listening alone can one bit of audio gear be distinguished from another, can a difference, any difference be shown. That is is simple as it gets. Any other complex attributes suggested to exist between gear is mute if the simple reality of simple difference, any difference can not be demonstrated by listening alone tests. Not simple at all. We are evaluating how well the equipment under tests reproduces a musical experience, measured against our life experience of how live music sounds to us. And that involves an involuntary emotional component that this is no evidence that quick-switch, comparative testing can include. " This isn't a question of science. It is a question of questionable testing proceedures used as a blind article of faith." That is atestable thesis, we will look forward to the results. Such testing is the bread and butter in all other areas of human testing. If it isn't valid for audio then it is not valid anywhere else either and many decades of research in humans will have to be tossed. Absolutely a false premise. So the evidence for such an extraordinary claim that audio is an exception will require some equally extraordinary demonstration. Except that it is not extraordinary because the assertion is based on a false premise. Astrology and esp etc. make that exact same claim. Testing can hand has been done on such thing with expected results. The same litany of excuses those folk propose for their failure to demonstrate any reality to their claims is exactly the same offered for those whose worldview is upset when listening alone testing doesn't support their worldview. |
#42
|
|||
|
|||
wrote in message ...
Harry Lavo wrote: By contrast, RAHE seems to me to be a theological place where firm belief in DBT's is aceepted as a "truth" for all purposes, regardless of questions raised or a complete change in the nature of the item being measured....eg music instead of sound....and I might add despite any attempt by the faithful to ever verify the test's applicability. A cult religion apparently also embraced by every accredited psychology department in the country. Talk about brainwashing. Sorry, Bob. in those departments it is only one type of test among many....and there are very many variations of DBT'ng. The kind promoted here...quick-switch, compartive DBT'ng...would not be used in many situations and has not been validated as providing the same results as blind, monadic testing among a large sample of audiophiles. Which by all the underlying assumptions would be a test much less likely to interfere with a true evaluation of musical accuracy, including unconscious emotional reaction. The differences arise as to how to answer questions about the origin of the perceptions we have when experiencing reproduction of sound events. It is equally valid to suggest that much of the perception resides in the brain and does not exist until after the signal has reached the ears and that perspective is one that commends itself well to testing using what are routine approaches in all other areas of human testing. Sorry, I spent years in the food industry were we spent enormous amounts of time, energy, and intellectual capital on testing...much much more sophisticated than the testing you suggest. Simple blind comparative testing was use to answer simple questions...is the salt level correct, is the sweetness correct, how do they affect preference, etc. The tests were blinded to eliminate differences in color or appearance, often because we were dealing with prototypes that were not fully developed. But when it came to evaluating the overall appeal of the product, with the decision on the line to go to test marketing (and therefore spend a lot of money and expose the product to the competition) we always used monadic testing among samples of no less than 300 people. Therefore a simple test might involve 600 people (test and control). A complex test could involve twice that many. The were blind (what we call "white box" but they were not comparative). They were used/eaten in the environment of normal use, and were rated after. Something similar would be a big step forward in audio evaluation vs. DBt. I can't speak for othe subjectivists, but I know I (and at least some others) simply want appropriate testing rather than inappropriate testing. Why in the world would we expect that the protocols of food testing would be appropriate for audio testing, or pharmaceutical testing, or... Because the underlying test techniques have strengths and weaknesses, and must be chosen for the task at hand. Obviously, there are implementation differences. |
#43
|
|||
|
|||
wrote in message ...
"I see sense in your words and i respect your words. But this is not so simple. Believe me, i see many things. I dont want be part of one or another extrem. This is problem here, you guys are EXTREM. I can accept some things, but not all. I am lucky that i dont need to buy amps and speakers. Everyone can make this if know something about electronic. And i try many things. Often i expect a lot, sometimes not, and results are not always like i expect. This is important! Theory about foolishnes is here very weak. Some amps really make change sound! For me this is so obviously that i really dont know what to say here and how to response. You have your believes i have mine. And then we have short-circuit. I dont want this. I try dont look further here, and maybe sometime all we find on same wavelenght in better frame of mind" It is easy to make an amp sound different and the electrical parameters by which to do so are well known. For those amps not trying to sound different but to only increase the signal as faithfully as it enters the amp we have become so successful as to have created a commodity market. In a large series of tests when the obvious and well known amp differences are controlled,ie. frequency response and staying within it's power supply design goals, by using listening alone they cann't be distinguished one from another. People who did accept that such amps can sound different and say they experience it on a common basis find that using listening alone they cann't distinguish the amps. That is the benchmark of data and experience from which we can have such a discussion by asking the question - why do the differences said to exist and be heard disappear when listening alone testing is done? Are you speaking of Tom Nousaine's tests. That's the only "large set" of dbt's of amps claiming no difference that I know of. And if I recall correc tly, all we have is Tom's anecdotal telling of the results of those tests. I don't recall they were ever published or even a white paper issued. Am I wrong? |
#44
|
|||
|
|||
"Stewart Pinkerton" wrote in message
... Possibly, or possibly there actually is something wrong with them. The existence of many good amplifiers does not preclude the existence of many bad ones! Yes, true! Of course! And this differences are often AUDIBLE. I believe that you think so too (but some persons here dont think so). This is only thing what is in this discussion important to me. I dont try here explain why is one amp better or not. That's simple. Such a loudspeaker is physically impossible - if it is to cover the full acoustic range. There are however many reasonably efficient and very high quality 'minimonitors' which can be combined with a good subwoofer to provide SOTA performance. Of course, but you understand me wrong! And i hope that you dont think that i give this yob to you. This long yob i grant to "scientific" persons what think that all CD-players and amps sound same. Impossible or not, it is more usefull. Dejan Petrovic |
#45
|
|||
|
|||
Harry Lavo wrote:
wrote in message ... Harry Lavo wrote: By contrast, RAHE seems to me to be a theological place where firm belief in DBT's is aceepted as a "truth" for all purposes, regardless of questions raised or a complete change in the nature of the item being measured....eg music instead of sound....and I might add despite any attempt by the faithful to ever verify the test's applicability. A cult religion apparently also embraced by every accredited psychology department in the country. Talk about brainwashing. Sorry, Bob. in those departments it is only one type of test among many Sorry Harry. Perceptual psychologists would never use anything other than a DBT for listening tests of partial loudness thresholds. And for the umpteenth time, please note that your post referred to DBTs with no qualifiers. If you want us to take you seriously, the very least you could do is be precise in your language (or else stop complaining that people are misinterpreting you). Now, it is true that psychologists-and audio professionals--use various forms of DBTs. It is NOT true that objectivists here insist on using only one form of DBT. I'll be glad to give credence to any form of DBT you can pull off. But I note that you still haven't pulled off any, and you continue to insist that the only ones worth doing are impossible to do. Very convenient. Some people here recommend quick-switching ABX tests over other forms of DBTs, and they do so for good reason. The scientific literature demonstrates that such listening tests are more sensitive than those that do not allow subjects to do proximate comparisons. A psychologist would also typically use a quick-switching test to measure thresholds--unless, of course, he were testing aural memory. So even if you were to amend your original statement to add your usual qualifiers, you would be wrong. bob |
#46
|
|||
|
|||
On 15 May 2005 22:52:57 GMT, "_Dejan_" wrote:
"Stewart Pinkerton" wrote in message ... Possibly, or possibly there actually is something wrong with them. The existence of many good amplifiers does not preclude the existence of many bad ones! Yes, true! Of course! And this differences are often AUDIBLE. Not among the good ones........... I believe that you think so too (but some persons here dont think so). This is only thing what is in this discussion important to me. I dont try here explain why is one amp better or not. That's simple. Such a loudspeaker is physically impossible - if it is to cover the full acoustic range. There are however many reasonably efficient and very high quality 'minimonitors' which can be combined with a good subwoofer to provide SOTA performance. Of course, but you understand me wrong! And i hope that you dont think that i give this yob to you. This long yob i grant to "scientific" persons what think that all CD-players and amps sound same. The good ones do. Impossible or not, it is more usefull. No, what was *really* useful was all the engineering effort that went into making sonically transparent CD players and amplifiers available to all. The 'high end' in 2005 is merely big boys toys and snobbery. -- Stewart Pinkerton | Music is Art - Audio is Engineering |
#47
|
|||
|
|||
wrote in message
... ............. People who did accept that such amps can sound different and say they experience it on a common basis find that using listening alone they cann't distinguish the amps. The worst thing is that this work probably in that way. Fantastic formula to cheat peoples. For me is enough hard only read this. Listen can be only nightmare. I listen music to enjoy, not for test. And i dont know all reasons why this metod dont work. I dont even know how this work. This is not my job, and i dont care about this. Your job is to find better solution (if you are so preoccupied with this) because this solution obviously dont work. And then you can talk about testing, not before. So now, go to work! Petrovic Dejan |
#48
|
|||
|
|||
"you have documentation? Lab reports?
Or do I have to repeat Dejan's remark: "I do not see science here, I only see talk about science." Which is the only thing the objectivists are good at: talk about "science" in a rather metaphysical way." Review the archives of this group to see discussion of the testing over several years that has been done. To make the question even more intresting, a standing offer of around $6000 for wire, $10000 for amps, $1000000 for some of the really fringe claims is available. In past cycles of this question, either here or in another list there was mention of similar offers in other countries for substantual sums, finland is one that comes to mind but there were others also. We would be happy to arrange to have you try for the money, and happy to add your datum point to the set of examples where folk could not demonstrate that a difference could be shown by listening alone blind testing. One discussed here in some detail before was an audio store owner who thought his then top of the line pass labs and an older yamaha in his store using his gear would be a cake walk, not so. |
#49
|
|||
|
|||
Just do the test, tap dancing will not make it's reality go away.
"It is bad science if it is the wrong test for it's purpose. In this case if it is unable to include the emotional and right-brain aspects of musical enjoyment in determining the difference. No matter how useful it may be elsewhere in the audio field for developing codecs, listening for specific audio cues, etc." Irrelevant strawman in the extreme. Have two amps, one said to have all of the dubious above, another judged by you not to have same. Do the blind test by listening alone. If you can not spot a difference, any difference, any such need no longer be considered valid. Will you soon be asking to demonstrate this ability? There are folk here who have repeatably offered to do it, you bring your own right brain. "Not simple at all. We are evaluating how well the equipment under tests reproduces a musical experience, measured against our life experience of how live music sounds to us. And that involves an involuntary emotional component that this is no evidence that quick-switch, comparative testing can include." A review of the archives would show innumberable times it has been pointed out that no such contraists are required, another strawman. You can listen as long as you want, on your gear, in your home, using your music; all that is required is that yyou not know which bit of As for the "life experience" and "musical experience", again irrelevant strawman used in an attempt to divert the issue. Take an amp of your choice said to create this "experience" and another cchosen by you said not to do so and by listening alone demonstrate a difference, any difference. If you cann't then we have our answer. This is not rocket science and the attempts to inject some mystical element into the question brings us smack back into the same arena of excuses used by astrology and esp etc. who try to obscure the basic situation wherein time after time they are shown not to have validity. Put the burden of any thing your imagination can muster said to reside in one amp and not in another. If you can not tell which amp is which by detecting difference then the question is answered wherin the source of such proported aspects pasted onto the amp reside. |
#50
|
|||
|
|||
wrote in message ...
Harry Lavo wrote: wrote in message ... Harry Lavo wrote: By contrast, RAHE seems to me to be a theological place where firm belief in DBT's is aceepted as a "truth" for all purposes, regardless of questions raised or a complete change in the nature of the item being measured....eg music instead of sound....and I might add despite any attempt by the faithful to ever verify the test's applicability. A cult religion apparently also embraced by every accredited psychology department in the country. Talk about brainwashing. Sorry, Bob. in those departments it is only one type of test among many Sorry Harry. Perceptual psychologists would never use anything other than a DBT for listening tests of partial loudness thresholds. And for the umpteenth time, please note that your post referred to DBTs with no qualifiers. If you want us to take you seriously, the very least you could do is be precise in your language (or else stop complaining that people are misinterpreting you). Now, it is true that psychologists-and audio professionals--use various forms of DBTs. It is NOT true that objectivists here insist on using only one form of DBT. I'll be glad to give credence to any form of DBT you can pull off. But I note that you still haven't pulled off any, and you continue to insist that the only ones worth doing are impossible to do. Very convenient. Some people here recommend quick-switching ABX tests over other forms of DBTs, and they do so for good reason. The scientific literature demonstrates that such listening tests are more sensitive than those that do not allow subjects to do proximate comparisons. A psychologist would also typically use a quick-switching test to measure thresholds--unless, of course, he were testing aural memory. So even if you were to amend your original statement to add your usual qualifiers, you would be wrong. You are right. I was referring to quick-switch, comparative DBT's a la the ABX promoted here for so long by Arny. And I can agree with you that some kind of blind test is appropriate to determine, not just differences, but also preferences and the reasons for those preferences. Unfortunately , this is not "measuring thresholds". This is evaluating audio and requires relaxed listening over periods of some length followed by a rating. It is a common test technique, but it is not a one-person test. It must be done by dozens, even hundreds, of people and then statistically compared to determine difference on rated quality and on specific audio characteristics. It is the statistical measure of difference that determine whether there is a difference/preference, but it also measures "what" the perceived difference/preference is in terms of audio characteristics. Most importantly, it its a test that allows the emotional response to music first, and then an evaluation of that response. Since such a test is impossible for one person to do, and since an ab test protocol for the home has not yet been developed which is simple enough to be practical and transparent, I believe for now sighted home listening, even with its possible error, is superior to using ABX. ABX or even direct, comparative AB testing are two cognitive and left-brain IMO to actual measure what they profess to measure. Perhaps that will change if we can get a non-quick-switch, semi-monadic, transparent home ab test protocol nailed down. |
#51
|
|||
|
|||
_Dejan_ wrote:
............. People who did accept that such amps can sound different and say they experience it on a common basis find that using listening alone they cann't distinguish the amps. The worst thing is that this work probably in that way. Fantastic formula to cheat peoples. For me is enough hard only read this. Listen can be only nightmare. I listen music to enjoy, not for test. And i dont know all reasons why this metod dont work. I dont even know how this work. This is not my job, and i dont care about this. Your job is to find better solution (if you are so preoccupied with this) because this solution obviously dont work. And then you can talk about testing, not before. So now, go to work! Hey Pet, How do you determine if some piece of gear is not up to your definition of good sound? You might enjoy listening to your crappy kitchen radio and be contented. (And continuing your logic, you would not have an idea of upgrading?) Do you compare different components and how? Or do you buy a new piece of gear just by recommendation? And what kind of process helps you evaluating if you feel a component is below your standards? -- ciao Ban Bordighera, Italy |
#52
|
|||
|
|||
_Dejan_ wrote:
And, you probably dont have problems with components and sound. You have less then 0.01% distorsions. Voila! Why you then reading this? Just dont tell me that you are soooo good man and you want help to others. You are allready happy with your sound and its time for something new. I see you like F1 so you can be another Schumacher. But this is not so easy like telling nonsenses. Or, maybe you trying to find speakers with 0.0001% distorsions. This is only thing what you need. Like i see in this group we have scientists, so you are maybe another one and you maybe have idea how to achieve this. I am very interested in this. You know much about CD-players, maybe about speakers too. Horns, TQWP, transmission line, bass reflex, closed box, variovent, open baffle, MKP, MKT, Elcos, air coils, ferrites, cooper foil coils, MF resistors..bla..bla....single drive speakers, two-way, three-way, or more??, co-axial speakers, tri-axial?...passive crossovers, active.. What is better for you? Pet, what in the world makes you think your opinion has any more weight than tose of others? You can list all the expressions you have heard or read about, that doesn't give you a PhD either. To me it seems more you are missing an academic title and want to proof that you have almost accumulated the same amount of knowledge if not more. I think this is foolish and also your neglected aversion against scientific approach. Who are you to decide if a certain test is valid or not. I think it much more foolish to pretend being an expert, but with all credentials missing. Or is it not so? you again dont hear differences because are distorsons always high? I can only gues what you think about peoples what cares about materials for boxes, even about varnish. This is hard to measure. Many things can be strange to me too, but i am openend mind, you are not, and this is not good. Sorry. And how do you know that the poster is *not* "open mind". Your open mindedness seems much smaller, because you do not even want to experiment, but believe you have already enough understanding. -- ciao Ban Bordighera, Italy |
#53
|
|||
|
|||
"Stewart Pinkerton" wrote in message
... Not among the good ones........... This is RELATIVELY true. And i say *often*, look again. BTW, this remind me to one song "Black/White World"(translation). But, i see colors. The good ones do. Again. Wich is "good one"?-I dont ask for answer. Bad/good? Black/white? Nothing more? In this moment i see ideological differences too. And i am happy with this - with differences i see colors. But i see logic in you answer too. In theory one amp must be better then another. If we now lay down all amps onto scale: from begin-bad one, to finish-good one. How are big differences betwen first and last?-biggest. Betwen one and next?-Veeeery Little - a lot amps is in the market. So your answer is logical. In the same time, price increasing of amps on this scale is for surely not linear. Far from that. With many peaks and drops. This is nothing new. Impossible or not, it is more usefull. No, what was *really* useful was all the engineering effort that went into making sonically transparent CD players and amplifiers available to all. The 'high end' in 2005 is merely big boys toys and snobbery. -- Stewart Pinkerton | Music is Art - Audio is Engineering I have not changed my mind. But good point. I am from poor country. I dont have money. All audio gear what i have i made myself. Except tweaked CD-Player wich is now broken. Even new laser is in this moment to expensive to me. Living in better world is my dream. In start i say "High-End", and Sullivan ask me to define this, but i resist. I dont give him answer - my error. To me is not neccesary that good component need to be expensive. And i dont say that are overpriced componets always better. But theory wich lead to conclussion that all amplifiers and CD-players have same sound is crap. This is reason why i dont want read to much about DBT, but last 3 days i spend my time to this like never before. Enough for me. I dont like the way how are results of DBT interpreted and published. Results of DBT can be only informations, NOT final answer. Audio DBT is more ideology then science to me, with many holes. If i change my mind, you will be know. Thanks for replays Cheers |
#54
|
|||
|
|||
When i start this thread i have no clue what that would be. In fact, i
dont like discussions, because of big probability of hard words, sparcles, etc. Ironically, *I* was guilt for this (i start this, and some bad emotions are from me). Maybe is to late, but now i want stop this like i was start: with apologyes. This dont mean that i think different then before (sorry for that), but this discussion is endless (like many others). My belief: True is somewhere in the middle (sorry again). Providentially, we all have ears and we can make answer for yourself. For shure, never again one single word from me because is this expirience awful for me, and probably for others. I come back to my cave. Thanks for answers, bye. Peace Petrovic Dejan |
#55
|
|||
|
|||
Harry Lavo wrote:
wrote in message ... Harry Lavo wrote: wrote in message ... Harry Lavo wrote: By contrast, RAHE seems to me to be a theological place where firm belief in DBT's is aceepted as a "truth" for all purposes, regardless of questions raised or a complete change in the nature of the item being measured....eg music instead of sound....and I might add despite any attempt by the faithful to ever verify the test's applicability. A cult religion apparently also embraced by every accredited psychology department in the country. Talk about brainwashing. Sorry, Bob. in those departments it is only one type of test among many Sorry Harry. Perceptual psychologists would never use anything other than a DBT for listening tests of partial loudness thresholds. And for the umpteenth time, please note that your post referred to DBTs with no qualifiers. If you want us to take you seriously, the very least you could do is be precise in your language (or else stop complaining that people are misinterpreting you). Now, it is true that psychologists-and audio professionals--use various forms of DBTs. It is NOT true that objectivists here insist on using only one form of DBT. I'll be glad to give credence to any form of DBT you can pull off. But I note that you still haven't pulled off any, and you continue to insist that the only ones worth doing are impossible to do. Very convenient. Some people here recommend quick-switching ABX tests over other forms of DBTs, and they do so for good reason. The scientific literature demonstrates that such listening tests are more sensitive than those that do not allow subjects to do proximate comparisons. A psychologist would also typically use a quick-switching test to measure thresholds--unless, of course, he were testing aural memory. So even if you were to amend your original statement to add your usual qualifiers, you would be wrong. You are right. I was referring to quick-switch, comparative DBT's a la the ABX promoted here for so long by Arny. And I can agree with you that some kind of blind test is appropriate to determine, not just differences, but also preferences and the reasons for those preferences. Whoa, there. DBTs can determine differences and preferences. Determining the reasons for those preferences is either trivial (ask the subjects why they preferred something), or extremely difficult (as in, I don't even know how I'd begin to tackle that one). Even Sean Olive doesn't claim to have determined WHY people prefer flat speakers with good dispersion. He only knows that those characteristics correlate with listener preferences. Unfortunately , this is not "measuring thresholds". Whoa, again. Determining difference is most definitely "measuring thresholds." By definition. bob |
#56
|
|||
|
|||
"Ban" wrote in message
... _Dejan_ wrote: And, you probably dont have problems with components and sound. You have less then 0.01% distorsions. Voila! Why you then reading this? Just dont tell me that you are soooo good man and you want help to others. You are allready happy with your sound and its time for something new. I see you like F1 so you can be another Schumacher. But this is not so easy like telling nonsenses. Or, maybe you trying to find speakers with 0.0001% distorsions. This is only thing what you need. Like i see in this group we have scientists, so you are maybe another one and you maybe have idea how to achieve this. I am very interested in this. You know much about CD-players, maybe about speakers too. Horns, TQWP, transmission line, bass reflex, closed box, variovent, open baffle, MKP, MKT, Elcos, air coils, ferrites, cooper foil coils, MF resistors..bla..bla....single drive speakers, two-way, three-way, or more??, co-axial speakers, tri-axial?...passive crossovers, active.. What is better for you? Pet, what in the world makes you think your opinion has any more weight than tose of others? You can list all the expressions you have heard or read about, that doesn't give you a PhD either. To me it seems more you are missing an academic title and want to proof that you have almost accumulated the same amount of knowledge if not more. I think this is foolish and also your neglected aversion against scientific approach. Who are you to decide if a certain test is valid or not. I think it much more foolish to pretend being an expert, but with all credentials missing. Or is it not so? you again dont hear differences because are distorsons always high? I can only gues what you think about peoples what cares about materials for boxes, even about varnish. This is hard to measure. Many things can be strange to me too, but i am openend mind, you are not, and this is not good. Sorry. And how do you know that the poster is *not* "open mind". Your open mindedness seems much smaller, because you do not even want to experiment, but believe you have already enough understanding. -- ciao Ban Bordighera, Italy Yes, i know, i tell him to hard words. Its answer is probably free of bad intention. And my reaction is like from idiot. This is not hard to confess to me. But in this group are some peoples wich have its believes, and THIS peoples talk free of of any another possibility. (in a way like "this is like that, story is over"). Please accept my appology for my behaviour |
#57
|
|||
|
|||
"Ban" wrote in message
... Hey Pet, How do you determine if some piece of gear is not up to your definition of good sound? You might enjoy listening to your crappy kitchen radio and be contented. (And continuing your logic, you would not have an idea of upgrading?) Do you compare different components and how? Or do you buy a new piece of gear just by recommendation? And what kind of process helps you evaluating if you feel a component is below your standards? -- ciao Ban Bordighera, Italy I am tired. ....listening, buying, testing, recommendation and crappy kitchen radio.... In short: Peoples here give me advice that is not important wich CD-player i need to buy because all CD-players have low distortion. I have no comment. If is this result of advanced scientific listening testing, then i dont want test nevermore. I know, i am Pet because i think so, but what i can? Because i am Pet. I am out Greetings |
#58
|
|||
|
|||
wrote in message ...
Just do the test, tap dancing will not make it's reality go away. "It is bad science if it is the wrong test for it's purpose. In this case if it is unable to include the emotional and right-brain aspects of musical enjoyment in determining the difference. No matter how useful it may be elsewhere in the audio field for developing codecs, listening for specific audio cues, etc." Irrelevant strawman in the extreme. Have two amps, one said to have all of the dubious above, another judged by you not to have same. Do the blind test by listening alone. If you can not spot a difference, any difference, any such need no longer be considered valid. Will you soon be asking to demonstrate this ability? There are folk here who have repeatably offered to do it, you bring your own right brain. "Not simple at all. We are evaluating how well the equipment under tests reproduces a musical experience, measured against our life experience of how live music sounds to us. And that involves an involuntary emotional component that this is no evidence that quick-switch, comparative testing can include." A review of the archives would show innumberable times it has been pointed out that no such contraists are required, another strawman. You can listen as long as you want, on your gear, in your home, using your music; all that is required is that yyou not know which bit of As for the "life experience" and "musical experience", again irrelevant strawman used in an attempt to divert the issue. Take an amp of your choice said to create this "experience" and another cchosen by you said not to do so and by listening alone demonstrate a difference, any difference. If you cann't then we have our answer. This is not rocket science and the attempts to inject some mystical element into the question brings us smack back into the same arena of excuses used by astrology and esp etc. who try to obscure the basic situation wherein time after time they are shown not to have validity. Put the burden of any thing your imagination can muster said to reside in one amp and not in another. If you can not tell which amp is which by detecting difference then the question is answered wherin the source of such proported aspects pasted onto the amp reside. I am reminded of what the mathematician at HE2005 said about engineers -- they are also too busy to get to the end to bother to examine the underlying premises. I simply and profoundly disagree, and have indicated why. You are not buying. I can't see that this discussion has to go any further. |
#59
|
|||
|
|||
On 18 May 2005 01:14:18 GMT, "_Dejan_" wrote:
"Stewart Pinkerton" wrote in message ... Not among the good ones........... This is RELATIVELY true. It's true to a level well below human perception. And that includes yours. And i say *often*, look again. BTW, this remind me to one song "Black/White World"(translation). But, i see colors. The good ones do. Again. Wich is "good one"?-I dont ask for answer. Bad/good? Black/white? Nothing more? In this moment i see ideological differences too. And i am happy with this - with differences i see colors. We all do - but sometimes there's no colour to be seen. But i see logic in you answer too. In theory one amp must be better then another. In practice also, they are all different - but not *audibly* so. If we now lay down all amps onto scale: from begin-bad one, to finish-good one. How are big differences betwen first and last?-biggest. Betwen one and next?-Veeeery Little - a lot amps is in the market. So your answer is logical. In the same time, price increasing of amps on this scale is for surely not linear. Far from that. With many peaks and drops. This is nothing new. Indeed so, which means that the majority of good mid-price amps all sound the same - and they are sonically transparent. It's only at the very bottom and the very top of the market that you find inferior products. Strange, but true. It's not an asymptote, it's a bathtub curve. Impossible or not, it is more usefull. No, what was *really* useful was all the engineering effort that went into making sonically transparent CD players and amplifiers available to all. The 'high end' in 2005 is merely big boys toys and snobbery. I have not changed my mind. But good point. I am from poor country. I dont have money. All audio gear what i have i made myself. Except tweaked CD-Player wich is now broken. Even new laser is in this moment to expensive to me. Living in better world is my dream. In start i say "High-End", and Sullivan ask me to define this, but i resist. I dont give him answer - my error. To me is not neccesary that good component need to be expensive. And i dont say that are overpriced componets always better. But theory wich lead to conclussion that all amplifiers and CD-players have same sound is crap. Indeed so - and you won't read anyone saying that. However, you *will* read people saying that all *well designed* amps and CD players sound the same - and no one has yet been able to show that this is not true. This is reason why i dont want read to much about DBT, but last 3 days i spend my time to this like never before. Enough for me. I dont like the way how are results of DBT interpreted and published. Results of DBT can be only informations, NOT final answer. Audio DBT is more ideology then science to me, with many holes. It is however the most sensitive test that we have, and sighted testing is easily shown to be useless for small differences. DBT is not 'ideology' at all, which is why it is used in so many branches of science. -- Stewart Pinkerton | Music is Art - Audio is Engineering |
#60
|
|||
|
|||
On 18 May 2005 00:58:06 GMT, "Harry Lavo" wrote:
You are right. I was referring to quick-switch, comparative DBT's a la the ABX promoted here for so long by Arny. And I can agree with you that some kind of blind test is appropriate to determine, not just differences, but also preferences and the reasons for those preferences. Unfortunately , this is not "measuring thresholds". This is evaluating audio and requires relaxed listening over periods of some length followed by a rating. It is a common test technique, but it is not a one-person test. It must be done by dozens, even hundreds, of people and then statistically compared to determine difference on rated quality and on specific audio characteristics. It is the statistical measure of difference that determine whether there is a difference/preference, but it also measures "what" the perceived difference/preference is in terms of audio characteristics. Most importantly, it its a test that allows the emotional response to music first, and then an evaluation of that response. Since such a test is impossible for one person to do, and since an ab test protocol for the home has not yet been developed which is simple enough to be practical and transparent, I believe for now sighted home listening, even with its possible error, is superior to using ABX. This is utter nonsense! Because you think that DBT requires hundreds of resulytts to verify its accuracy, you go on to suggest that it *follows* that a *known* flawed test method is superior? Sighted testing is *useless* for small differences, that can be proved by *anyone* in a matter of minutes. You seem to be a person of little respource or determination, as I have no problem whatever in setting up double-blind tests, with or without fast switching. If indeed you insist on long listening periods, then it's even easier, as simple cable-swapping is appropriate for such a leisurely and relaxed comparison. Just don't peek! Of course, such a scheme still won't give results which match your prejudices, so no doubt you'll come up with another 'problem'. If only you would put as much energy into actually doing the research, rather than picking minute flaws in existing and well-practiced techniques. Pluck first the beam from thine own eye............. ABX or even direct, comparative AB testing are two cognitive and left-brain IMO to actual measure what they profess to measure. Perhaps that will change if we can get a non-quick-switch, semi-monadic, transparent home ab test protocol nailed down. This ain't rocket science, Harry! Just match levels and make it double-blind, that's all you have to do. Hardly a challenge...... -- Stewart Pinkerton | Music is Art - Audio is Engineering |
#61
|
|||
|
|||
wrote in message ...
Harry Lavo wrote: wrote in message ... Harry Lavo wrote: snip Determining the reasons for those preferences is either trivial (ask the subjects why they preferred something), or extremely difficult (as in, I don't even know how I'd begin to tackle that one). Even Sean Olive doesn't claim to have determined WHY people prefer flat speakers with good dispersion. He only knows that those characteristics correlate with listener preferences. I agree, sometimes it is easy to surmise the reason, and sometimes it is very difficult. "Creativity" in test design is sometimes required to get insight. You do this (although you can blow it) by first getting a "quality" rating of some kind (usually a five point scale, which if statistically different between the two pieces of gear presumes a sensed differenced.). Then you ask for a series of carefully constructed subsidiary ratings for various aspects of the test experience. You can then apply statistical significance tests to determine which also prove "different" in support of the overall rating. Then if you are highly skilled at statistics, you can also do multi-dimensional scaling, which can further help illuminate the reasons for difference/preferences, in particular what the relative importance (or unimportance) of various factors were. Let me give you two possible audio examples, one soft and a bit far out, and one fairly defined and practical. Assume you are Sony having coming up with SACD and a new player. You are measuring it against your flagship player at roughtly the same price point. You use a test hybrid SACD with exactly the same mix done using the best available technology for both. The SACD wins.. If you had asked "how relaxed were you while listening ?" and found a significant difference tha highly correclated with overall preference, then you would know that somehow this unit was putting the listener more at ease than the other. You might have to do probing with focus groups to get further insight, and you would have to use your audio and engineering expertise to surmise what might be causing this, make sure it is intrinsic to the system, and perhaps decide to use it in your marketing. On the other hand, if the factor that correlated was "rank your impressions of the soundstaging" then you would know the unit was preferred mostly because of better soundstaging (and if you designed the scale or scales right, exactly how the better soundstage was percceived). Let me give you a real example from another field. In a past life I ran a small mid-westen mexican food restaurant chain of about 120 stores. We had a standard, square-cornered slump block store similar to the old Taco Bells. We designed a new store that had a much softer adobe look, with rounded corners and a less plastic interior. Our testing asked people to rate the stores on a whole host of atrributes (not comparative, just rate them) Then statistics were applied. It turned out men were less likely to frequent the new store than the old. But the market researchers went crazy, because they just couldn't explain it...all the standard measurements they thought important in a restaurant rated the same, and on a few measures the new store clearly excelled. But I had thrown in a group of questions relating to appropriateness of being a place to bring a date, when the would consider eating there, etc. The "date" question did correlate strongly with the opposite negative frequency intent...in other words it was a good place to take a date, a less good place to eat frequently. Using these as clues, follow up research showed that the typical young male found the stores somewhat "uncomfortably" feminine (what with the new rounded corners, softer materials and lighting, etc.) and was also perceived as (again uncomfortably) "more upscale" even though they knew and indeed their formal rating showed that the price was no diffrent than in the older stores. This led to a slight modification of the store design, easy to fix once you found out what was opertional. Unfortunately this is not "measuring thresholds". Whoa, again. Determining difference is most definitely "measuring thresholds." By definition. Depends somewhat on the definition. Typically it is measuring differences in perception and preference. It presumes awareness, but this may be an unconscious quality based directly on emotion as well as more conscious awareness as I attempted to show above. One of the advantages of doing monadic testing is that no time are you putting the test-taker in the position of "going comparative" with it's supposed elevation of left brain thinking. The young men rating these stores were not consciously examining their feelings of masculinity, nor rationally deterrmining their projected eating frequency. They just gave answers. It was the statistical comparison that determined the differences and significance of same which ultimately led to the underlying feelings masculinity and cultural appropriateness. That is why I think it is a much better test for something as ethereal as rating equipment in its ability to reproduce live sound. |
#62
|
|||
|
|||
"Stewart Pinkerton" wrote in message
... On 18 May 2005 00:58:06 GMT, "Harry Lavo" wrote: You are right. I was referring to quick-switch, comparative DBT's a la the ABX promoted here for so long by Arny. And I can agree with you that some kind of blind test is appropriate to determine, not just differences, but also preferences and the reasons for those preferences. Unfortunately , this is not "measuring thresholds". This is evaluating audio and requires relaxed listening over periods of some length followed by a rating. It is a common test technique, but it is not a one-person test. It must be done by dozens, even hundreds, of people and then statistically compared to determine difference on rated quality and on specific audio characteristics. It is the statistical measure of difference that determine whether there is a difference/preference, but it also measures "what" the perceived difference/preference is in terms of audio characteristics. Most importantly, it its a test that allows the emotional response to music first, and then an evaluation of that response. Since such a test is impossible for one person to do, and since an ab test protocol for the home has not yet been developed which is simple enough to be practical and transparent, I believe for now sighted home listening, even with its possible error, is superior to using ABX. This is utter nonsense! Because you think that DBT requires hundreds of resulytts to verify its accuracy, you go on to suggest that it *follows* that a *known* flawed test method is superior? Sighted testing is *useless* for small differences, that can be proved by *anyone* in a matter of minutes. You seem to be a person of little respource or determination, as I have no problem whatever in setting up double-blind tests, with or without fast switching. If indeed you insist on long listening periods, then it's even easier, as simple cable-swapping is appropriate for such a leisurely and relaxed comparison. Just don't peek! Of course, such a scheme still won't give results which match your prejudices, so no doubt you'll come up with another 'problem'. If only you would put as much energy into actually doing the research, rather than picking minute flaws in existing and well-practiced techniques. Pluck first the beam from thine own eye............. I'll ignore all the attacks on my competency and bias by this *most objective* of all usegroup players. ABX or even direct, comparative AB testing are two cognitive and left-brain IMO to actual measure what they profess to measure. Perhaps that will change if we can get a non-quick-switch, semi-monadic, transparent home ab test protocol nailed down. This ain't rocket science, Harry! Just match levels and make it double-blind, that's all you have to do. Hardly a challenge...... Fine, if you want to do it casually, amateurish, and ... *wrong*. Not so fine, if you are talking about doing it professionally so you can be sure the results are as accurate as they can be...but it would still be *wrong* if it measures the wrong thing because intrinsically it alters the mode of listening. I am trying to spell out (based on a research background in general and audio logic in particular) the kind of test needed to verify that the test doesn't get in the way of what it attempts to measure. Not hard to understand, just hard to do. But until it is done, the test you want us to use is unproven and suspect at best and in error at worst. What is the sense of doing a test if you don't know that it measures the thing it purports to measure? I'd rather spend my time and energy trying to help somebody (the group, Stereophile, somebody) do the work necessary to determine what type of test measures most accurately *what we want to measure .... proficiency in playing music with realism*. |
#63
|
|||
|
|||
"I am reminded of what the mathematician at HE2005 said about engineers --
they are also too busy to get to the end to bother to examine the underlying premises. I simply and profoundly disagree, and have indicated why. You are not buying. I can't see that this discussion has to go any further." The subjective enterprise cann't demonstrate even it's most basic assumptions. The above same cuts both ways in all points. It depends on what one places at the spot of underlying etc. and over looks in haste to support an end. For each and every aspect that is said to be critical that some amp audibly has and others don't then spotting same should be easy to pick, you just don't get to know which is active but in every other aspect you wish are free to identify the amp you think holds those aspects. It ain't rocket science, nor does being a math person prove an obstical; it's just a listening alone test. In haste to get to the end that the subjective enterprise is valid, the underlying undemonstrated assumption is made that the enterprise can not have such a critical and obvious flaw, the failure to demonstrate the validity of the enterprise must lie in the test which continiously fails to confirm. So too is the automatic and oft repeated haste of astrology and esp which too "know" they are right and are victums to being validated by testing but failing same are only speculation. When faced with "might be" and demonstrated every day in the whole of human testing why should we buy the pig in the poke? |
#64
|
|||
|
|||
Get two amps, one said to have any number of audible effects and another
said to clearly lack them. Listen as long as you like using the exact context and gear of your choice to either in any order or none at all to really nail down your conclusion. Then have someone choose one and add only a cloth over the connections which might reveal which amp is active and pronounce which it is, based squarely on what was so very obvious that caused you to pick the obviously contrasting amps in the first place and which you so clearly confirmed just minutes before. No comparson need be made. A random series, still with no comparison amps is done with each amp ending up with the same number of trial attempts. If the differences reside in the gear the results will jump right off the paper on which the results were recorded. No left brain or right brain or any other objections of the day, just a recreation of your experience; you just don't get to know which amp is active. All of the many objections are at this point conjecture and must in themselves be shown to be confounding elements before one need design a test to control for them. Just do the test. |
#65
|
|||
|
|||
"Harry Lavo" wrote
What is the sense of doing a test if you don't know that it measures the thing it purports to measure? I'd rather spend my time and energy trying to help somebody (the group, Stereophile, somebody) do the work necessary to determine what type of test measures most accurately *what we want to measure .... proficiency in playing music with realism*. What's the sense in doing that since we all hear and listen differently? |
#66
|
|||
|
|||
wrote in message ...
Get two amps, one said to have any number of audible effects and another said to clearly lack them. Listen as long as you like using the exact context and gear of your choice to either in any order or none at all to really nail down your conclusion. Then have someone choose one and add only a cloth over the connections which might reveal which amp is active and pronounce which it is, based squarely on what was so very obvious that caused you to pick the obviously contrasting amps in the first place and which you so clearly confirmed just minutes before. No comparson need be made. A random series, still with no comparison amps is done with each amp ending up with the same number of trial attempts. If the differences reside in the gear the results will jump right off the paper on which the results were recorded. No left brain or right brain or any other objections of the day, just a recreation of your experience; you just don't get to know which amp is active. All of the many objections are at this point conjecture and must in themselves be shown to be confounding elements before one need design a test to control for them. Just do the test. This is a comparative test, and if such testing "kills" the right brain/Type I response than it gives erroneous results, not matter how much listening preceeds it. Until that possibility is tested and put to rest (and there is reason to believe it may be valid) then such testing is not only foolish but potentially misleading. |
#67
|
|||
|
|||
"Norman M. Schwartz" wrote in message
... "Harry Lavo" wrote What is the sense of doing a test if you don't know that it measures the thing it purports to measure? I'd rather spend my time and energy trying to help somebody (the group, Stereophile, somebody) do the work necessary to determine what type of test measures most accurately *what we want to measure .... proficiency in playing music with realism*. What's the sense in doing that since we all hear and listen differently? The issue, Norm, is "how do we determine the underlying truth...is there a difference of isn't there?". We can only do that by using the best technique for the taks, the one least likely to get in the way of recording the full impact of the musical test on the testee. That is monadic testing...i.e. listen for an extended period, completely relaxed...then rate the event. Ideally don't even know what is being tested. Then rely on statistics to tell you whether or not the two devices under test are different or not. That is not a single person test...it requires a bunch of people...dozens at least, preferably several hundred...and they are needed as well to determine "truth" in the sense of being broader than one person. Then, once we have that measure, you have a standard to test single-person blind AB testing, ABX, ABC/hr, etc. against to see if they give the same results, measured over a bunch of people as well. And then, only if they do, can they be used as a definitive test for an individual person with any validity. That work has not been done...by Arny nor by anybody else in audio so far as I can tell. |
#68
|
|||
|
|||
wrote in message ...
Also, lest you think we are frauds, rest assured that everyone who posts here really is interested in the quality of audio reproduction. But some of us think the best way to achieve that is to concentrate on the pieces of the chain that really do make a difference in sound quality: the recording itself, the loudspeakers, and the placement of the speakers inthe room. I must say there is little discussion of the above 3 items that "really do make a difference". I agree that they are the 3 critical items (although the whole room treatment is critical, not just the placement of speakers), but 90% of the traffic is subjectivist v objectivist, over and over and over and over..... and 90% of the few threads on speaker and recording are subjectivist / sighted, too. I'd love to hear from someone or see references to actual distortion measurements of speakers, of various harmonics and intermod dist. .....and actual measurements of recordings' noise floor, musical dynamic range, evidence of clipping and compression, average-to-peak levels. |
#69
|
|||
|
|||
nowater wrote:
wrote in message ... Also, lest you think we are frauds, rest assured that everyone who posts here really is interested in the quality of audio reproduction. But some of us think the best way to achieve that is to concentrate on the pieces of the chain that really do make a difference in sound quality: the recording itself, the loudspeakers, and the placement of the speakers inthe room. I must say there is little discussion of the above 3 items that "really do make a difference". I agree that they are the 3 critical items (although the whole room treatment is critical, not just the placement of speakers), Agreed. I was trying to be succinct. but 90% of the traffic is subjectivist v objectivist, over and over and over and over..... Yeah. I wish the subjectivists would give it a rest. :-) and 90% of the few threads on speaker and recording are subjectivist / sighted, too. It's well-nigh impossible for amateurs to do blind evaluations of speakers. Unless you work for Harman or a Canadian manufacturer, sighted subjective evaluations are all you've got. I'd love to hear from someone or see references to actual distortion measurements of speakers, of various harmonics and intermod dist. Good sources for speaker measurements include: www.stereophile.com (Ignore the reviews themselves and go straight to John Atkinson's measurements.) www.soundstageav.com/avreviews_speakers.html (The ones labeled NRC have measurements.) www.soundandvisionmag.com (Tests by Tom Nousaine accompany speaker reviews.) www.theaudiocritic.com (Very little so far, and full site access costs $10.) ....and actual measurements of recordings' noise floor, musical dynamic range, evidence of clipping and compression, average-to-peak levels. I don't know of anybody who actually measures recordings. bob ____________ "Further carefully-conducted blind tests will be necessary if these conclusions are felt to be in error." --Stanley P. Lip****z |
#70
|
|||
|
|||
"nowater" wrote in message
... ....and actual measurements of recordings' noise floor, musical dynamic range, evidence of clipping and compression, average-to-peak levels. Actually not much of a problem with some available software for you to investigate what is on CD's or DVD's with your home PC. Many popular CD's only have average to peak levels of around 15 dB. Some don't have any more dynamic range than that. Noise floors are almost never more than 70 dB down from max level in the few dozen I have looked at this way. In general objectively, seems few if any recordings have needs exceeding the capability of your basic CD format. Actually few even exercise the CD format very much. Also can be instructive to play with compression, expansion and other alterations of recordings with CoolEdit (now Adobe Audition) or similar programs. Frequency response tilts are enlightening too. Results can be very surprising. Such as finding a little compression often seems to enhance the subjective dynamic range. While expansion leads to a lifeless low energy sound that is subjectively 'compressed'. Or that slight frequency cuts in the high frequencies sound like a boost of the low end or vice versa. Maybe someone can suggest a bit of software with these capabilities that is less costly than Adobe Audition has become. Dennis |
Reply |
|
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
please compare & comment (improbably similar?) | Pro Audio | |||
please compare & comment (improbably similar?) | Pro Audio | |||
comment on my proposal | Tech | |||
A Comment Britney et al. | Vacuum Tubes | |||
Comment on new planned setup | Car Audio |