Home |
Search |
Today's Posts |
#241
|
|||
|
|||
Arny Krueger wrote:[i] "John Atkinson" wrote in message ups.com... Arny Krueger wrote: "John Atkinson" wrote Arny Krueger wrote: quoted me in message roups.com. .. "It should be noted for the record, Mr. Krueger, that in the tests that were performed at the 1988 AES Convention, the number of trials was set at 5 by your associate David Clark. I have already pointed out that this is an inadequate number of trials, as even a listener scoring 5/5 identifications correct would not reach the 95% confidence level. Obviously, just one of Atkinson's anti-scientific misapprehensions... I fear you don't seem to have comprehended what I wrote about the 1988 Clark AES test, Mr. Krueger. My correctly pointing out that scoring 5 correct out of 5 trials, as Michael Fremer did, doesn't reach the 95% confidence level is factually correct. Period. and lacking proof that anybody proclaimed it as "proving" that no audible differences existed. Not only has this series of tests been widely cited in the years since it took place as "proving" the amplifiers didn't sound different from one another, it was said so by David Clark and others in the AES workshop that followed the tests. I was present at that workshop, as was Michael Fremer, and the event was discussed both in Stereophile [and] in the JAES. However, as you were not present at the test or the workshop, Mr. Krueger, and had no involvement in the organization of either, it is diffcult to see what factual basis exists to support your denials. What this quote does prove is that my friend David Clark, who is not lacking in persuasive powers, failed to get that wiley old fox named John Atkinson to do a statistically significant number of trials, which would be required for this to even be a test. As explained in the magazine and journal coverage at that time, far from trying to persuade listeners to perform more than 5 trials, it was Clark himself who limited the number of trials any one listener could do at one time to 5. Yes it was possible for someone to perform a second set of 5 trials. However, this would have involved standing in line for an hour or more for a second time, and people do have other matters to attend to at an AES Convention. I guess can take this as proof that to this day, Atkinson can't look at a set of 5 results and determine that it is too small of a sample for finding a conclusive result. I have repeatedly staed that a test of 5 trials is too small for a positive result to reach the statistical confidence level to ve considered conclusive. I have also stated that this was, I believe, arranged deliberately by the organizer of the test so that no one listener's results could be used to produce anything but a null result. When _all_ the listeners' results were lumped together, then the number of trials is large enough to be considered statistically, of course. But then the inability of the test's designer to limit the variables led to the expected null result. Which still does not "prove" the amplifiers sounded the same, just as you correctly point out on your website, Mr. Krueger. So what was it John, were you blinded by the pretense of Science so badly that you couldn't see that this was actually a demonstration? No, Mr. Krueger, it was a_test_, as Michael Fremer and I have repeatedly informed you. It was described as a test by David Clark and by the AES. The only one who insists it wasn't is you and you have no evidence for that opinion. It was obviously a demonstration! The fact that Atkinson represents this demonstration as a test speaks to his poor understanding of scientific testing methodologies. Clark himself described it as a test, _not_ a demonstration. Just goes to show that if David Clark told John Atkinson that a pile of crap was cake, Atkinson would eat it. It appears your argument devolves to an accusation leveled at your associate David Clark that he was lying to both to the AES and to the listeners like myself who agreed to take part in his test. Perhaps you'd better make that point to Mr. Clark's face. But let's postulate that you are correct: that this test of David Clark's was so poorly designed and performed that the results should be discarded, as you insist. In that case, then, should't you admit that all the commentators who have cited this test as "proving" there were no audible differences between the amplifiers were _wrong_ to have made that claim? These events were what they were, which is that they were demonstrations. No, Mr. Krueger, they were a test, as described by David Clark. You propose that Clark was lying to me about this, which I (and probably he) reject. They are inadequte as the only evidence that would prove a far-reaching conclusion such as "all good amplifiers sound the same" My point exactly, Mr. Krueger. Going further, if you dsiqualify this 1988 test, then shouldn't you also then disqualify _all_ the tests that produced null results but failed to meet your 10 Minimum Requirements as reproduced on pcabx.com? At this time, this would seem to be a reasonable thing to do. Exactly my point, Mr. Krueger. It is indeed reasonable to reject the results of poorly designed, poorly performed blind tests as "proving" any specific result. As you shortly admit, Nousaine's collection of null results (which go well beyond the three or four examples cited in this post) is just a collection of circumstantial evidence. Despite your admission, Mr. Krueger, Nousaine has strenuously argued that the null results of all these tests _do_ so prove that amplifiers cannot be distinguished by ear. Was he lying also? Or does he just lack your own scientific insight? Note that Nousaine's paper never sucessfully made it through the AES referee process for publication in the journal. If you say so, Mr. Krueger. I have no knowledge of Mr. Nousaine's tribulations at the hands of the AES referees. The Nousaine/Zipser tests, for example? Those were about a failure by Zipser to prove Zipser's hypothesis that the amps sounded mind-blowingly different. But the Zipser test clearly failed to meet the criteria for a properly designed blind test. As you said above, Mr. Krueger, it is therefore "reasonable" to dismiss the results of that test as meaningless. The most damning evaluation of those tests was posted by Zipser on RAO. Then Zippy had his dramatic change of heart... The late Steve Zipser's opinion is hardly relevant as he neither designed the test nor did he proctor it. You and others have proclaimed that this test "proved" Steve could not distinguish the amplifiers by ear. Yet now you agree that the test was so poorly designed and performed that the results are meaningless. You can't have it both ways, Mr. Krueger. If you do so, then by your own logic, there no longer exists any of the blind tests that Tom Nousaine, for example, has loudly and longly proclaimed as being strong circumstantial evidence for the absence of audible amplifier differences. This claim rather vastly understates the total amount of evidence that Nousaine has presented. No, I have read Mr. Nousaine's writings on this subject. His circumstantial evidence is merely a list of anecdotes about blind tests that each, individually, can be dismissed on the grounds that they do not meet your own requirements for a test to produce meaningful results. So given that in one post you retreat into declaring David Clark a liar and Tom Nousaine an incompetent tester in order to support your arguments, Mr. Krueger, I fail to see what else needs to be said. :-) John Atkinson Editor, Stereophile |
#242
|
|||
|
|||
wrote: Howard Ferstler wrote: Let's be realistic, most serious high-end audio buffs are true believers who have no particular interest at all in doing careful comparing. Let's consider Howard's idea of "careful comparing": "I just kept pushing the button and making wild stabs at what I thought I heard. After a while, I did not bother to listen at all. I just kept pressing the same choice over and over." Don't forget that in the first round of trials, before Howard Ferstler decided to randomize his scoring as described above, proper statistical analysis revealed that he did indeed hear a difference between the amplifiers under test. However, he rejected that result because it didn't not conform to his preconceived notions, surely the mark of a "true believer." :-) John Atkinson Editor, Stereophile |
#243
|
|||
|
|||
"John Atkinson" wrote in message oups.com... Arny Krueger wrote: [i] I guess can take this as proof that to this day, Atkinson can't look at a set of 5 results and determine that it is too small of a sample for finding a conclusive result. I have repeatedly stated that a test of 5 trials is too small for a positive result to reach the statistical confidence level to ve considered conclusive. So then John, why do you keep using the word test in conjunction with events composed of 5 trials or less? I have also stated that this was, I believe, arranged deliberately by the organizer of the test so that no one listener's results could be used to produce anything but a null result. If it ain't a test then the results are of no major consequence, right? When _all_ the listeners' results were lumped together, then the number of trials is large enough to be considered statistically, of course. Subject to the reasonable constraint that the consolidation of results must have been planned in detail before any trials were run. Where is the statement by Clark that pre-defined this critical variable? But then the inability of the test's designer to limit the variables led to the expected null result. I see this gratuitous and cryptic statement as a shabby attempt to invent a claim that null results were the desired outcome of the event. Which still does not "prove" the amplifiers sounded the same, just as you correctly point out on your website, Mr. Krueger. Let's stipulate that I have devoted a number of years of my life and a substantial part of a popular web site to doing proper listening tests that would show, if possible that there are audible differences between so-called good power amplifiers. So what was it John, were you blinded by the pretense of Science so badly that you couldn't see that this was actually a demonstration? No, Mr. Krueger, it was a_test_, as Michael Fremer and I have repeatedly informed you. The facts that you have presented seem to paint a vastly different picture. The event was contrived so that it would be difficult for any participant to obtain statistical signficant results. You yourself John just complained about a natural 5 trial limit that was built into the event. You just wrote: "I have repeatedly stated that a test of 5 trials is too small for a positive result to reach the statistical confidence level to ve considered conclusive." It was described as a test by David Clark and by the AES. If that is true then they did so in error or subject to some other variables that haven't been sufficiently discussed yet. The only one who insists it wasn't is you and you have no evidence for that opinion. For better or worse I have statements by John Atkinson as my evidence. It was obviously a demonstration! The fact that Atkinson represents this demonstration as a test speaks to his poor understanding of scientific testing methodologies. Clark himself described it as a test, _not_ a demonstration. Just goes to show that if David Clark told John Atkinson that a pile of crap was cake, Atkinson would eat it. It appears your argument devolves to an accusation leveled at your associate David Clark that he was lying to both to the AES and to the listeners like myself who agreed to take part in his test. My arguement is that if they did so, then they did so in error or subject to some other variables that haven't been sufficiently discussed yet. Perhaps you'd better make that point to Mr. Clark's face. I certainly did so right in Mr. Nousaine's face, in your presence John. Neither Nousaine nor Clark nor myself have any secrets from each other in this matter. But let's postulate that you are correct: that this test of David Clark's was so poorly designed and performed that the results should be discarded, as you insist. In that case, then, should't you admit that all the commentators who have cited this test as "proving" there were no audible differences between the amplifiers were _wrong_ to have made that claim? These events were what they were, which is that they were demonstrations. No, Mr. Krueger, they were a test, as described by David Clark. You propose that Clark was lying to me about this, which I (and probably he) reject. What's unclear about "...if they did so, then they did so in error or subject to some other variables that haven't been sufficiently discussed yet"? They are inadequte as the only evidence that would prove a far-reaching conclusion such as "all good amplifiers sound the same" My point exactly, Mr. Krueger. Then we agree about this. Noisily perhaps, but some kind of agreement nevertheless. Going further, if you disqualify this 1988 test, then shouldn't you also then disqualify _all_ the tests that produced null results but failed to meet your 10 Minimum Requirements as reproduced on pcabx.com? At this time, this would seem to be a reasonable thing to do. Exactly my point, Mr. Krueger. It is indeed reasonable to reject the results of poorly designed, poorly performed blind tests as "proving" any specific result. Then we agree about this. Noisily perhaps, but some kind of agreement nevertheless. Of course we are discussing something that happened about 20 years ago. This was before the publication of recommendation BS 1116, before publication of the details of ABC/hr, before JJ's coder tests, and before the opening of the www.pcabx.com web site. To summarize, the so-called objectivists listed above have made a lot of progress, while leading self-proclaimed subjectivists are still basing all their current work on 27 years old poorly-done blind and sighted evaluations. Despite all the scientific progress that has been made, it remains true that there is no evidence for the idea that good power amps sound the same, despite present and future claims in Stereophile. As you shortly admit, Nousaine's collection of null results (which go well beyond the three or four examples cited in this post) is just a collection of circumstantial evidence. Despite your admission, Mr. Krueger, Nousaine has strenuously argued that the null results of all these tests _do_ so prove that amplifiers cannot be distinguished by ear. Was he lying also? Or does he just lack your own scientific insight? What you seemingly don't want to admit John is that you still have your head in the sand, and there still is no evidence that good power amps sound different as generally used. The Nousaine/Zipser tests, for example? Those were about a failure by Zipser to prove Zipser's hypothesis that the amps sounded mind-blowingly different. But the Zipser test clearly failed to meet the criteria for a properly designed blind test. As you said above, Mr. Krueger, it is therefore "reasonable" to dismiss the results of that test as meaningless. In the the light of current science the Zipser tests themselves proved nothing. However, Zipser responded to the situation in a highly dishonorable fashion, one that while different is at least as dishonorable as Stereophile's current equipment testing policy. The most damning evaluation of those tests was posted by Zipser on RAO. Then Zippy had his dramatic change of heart... The late Steve Zipser's opinion is hardly relevant as he neither designed the test nor did he proctor it. False claim, as Zippy had control over many important test parameters and had the freedom to agree or disagree with the rest. You and others have proclaimed that this test "proved" Steve could not distinguish the amplifiers by ear. Prove it. It is quite clear that despite Zippy's claims of mind-blowing audible differences between the amps, he failed to support the hypothesis that they sounded different. Yet now you agree that the test was so poorly designed and performed that the results are meaningless. That was due to intransigence on Zippy's part. You can't have it both ways, Mr. Krueger. I decline to take responsibility for Zipser's behavioir. If you do so, then by your own logic, there no longer exists any of the blind tests that Tom Nousaine, for example, has loudly and longly proclaimed as being strong circumstantial evidence for the absence of audible amplifier differences. This claim rather vastly understates the total amount of evidence that Nousaine has presented. No, I have read Mr. Nousaine's writings on this subject. His circumstantial evidence is merely a list of anecdotes about blind tests that each, individually, can be dismissed on the grounds that they do not meet your own requirements for a test to produce meaningful results. Mr. Nousaine's paper said quite clearly that the anecdotes should be considered as a group. This casts serious doubts on your remaining analysis of it, John. So given that in one post you retreat into declaring David Clark a liar and Tom Nousaine an incompetent tester in order to support your arguments, Mr. Krueger, I fail to see what else needs to be said. :-) A joke, of course. |
#244
|
|||
|
|||
Arny Krueger wrote:
"John Atkinson" wrote in message oups.com... It appears your argument devolves to an accusation leveled at your associate David Clark that he was lying to both to the AES and to the listeners like myself who agreed to take part in his test. Perhaps you'd better make that point to Mr. Clark's face. I certainly did so right in Mr. Nousaine's face, in your presence John. One final point, as Mr. Krueger seems unwilling to discuss this subject in any logical manner: No, Mr. Krueger, you did _not_ make this point "in Mr. Nousaine's face in [my] presence," Mr. Krueger. The only conversation you and I have had with Tom Nousaine present was at HE2005 when I informed you of the failure of the laptop I had procured for you to give a PowerPoint presentation. You made no mention of the 1988 AES tests or David Clark. Yes, subsequent to the debate, you were screeching "demonstration, demonstration, demonstration" in Michael Fremer's face when he was describing the 1988 Clark test to you, but you certainly did not refer to David Clark as a liar in that exchange, unless you did so when I was out of earshot. Neither Nousaine nor Clark nor myself have any secrets from each other in this matter. In that were the case, Mr. Krueger, then it is hard to grasp what you mean when you wrote in response to my _true_ statement that the 1988 AES amplifier listening test was described as a test by David Clark and by the AES. If that is true then they did so in error or subject to some other variables that haven't been sufficiently discussed yet. First you say that Clark doesn't keep secrets from you but then that Clark didn't keep you fully informed of the details of his 1988 test. Oh well... John Atkinson Editor, Stereophile |
#245
|
|||
|
|||
"John Atkinson" wrote in message oups.com... John invalidated his reply to me by not properly recognizing the appearances and meaning of the following phrase, which appeared several times in my post: "...then they did so in error or subject to some other variables that haven't been sufficiently discussed yet" John quoted this phrase once near the end of his reply, but this phrase appeared in the post of mine he purported to quote from, a number of times before that, including once before he started quoting my post. My entire reply is based on the reader understanding this phrase, so Atkinson's gratuitous excising of it and then largely ignoring it makes his reply invalid. |
#246
|
|||
|
|||
Arny Krueger wrote: John invalidated his reply to me by not properly recognizing the appearances and meaning of the following phrase, which appeared several times in my post: "...then they did so in error or subject to some other variables that haven't been sufficiently discussed yet" :-) John Atkinson Editor, Stereophile |
#247
|
|||
|
|||
In article ,
"Arny Krueger" wrote: "John Atkinson" wrote in message oups.com... John invalidated his reply to me by not properly recognizing the appearances and meaning of the following phrase, which appeared several times in my post: Well, you're not hiding behind a typo. "...then they did so in error or subject to some other variables that haven't been sufficiently discussed yet" John quoted this phrase once near the end of his reply, but this phrase appeared in the post of mine he purported to quote from, a number of times before that, including once before he started quoting my post. My entire reply is based on the reader understanding this phrase, so Atkinson's gratuitous excising of it and then largely ignoring it makes his reply invalid. If you're misunderstood, maybe you should try to be clearer. You appear to be saying in this thread that Clark's "demonstration" was self-evidently bogus and should have been rejected at sight and also saying that because it falls short of being a good test it wasn't a test at all. Stephen |
#248
|
|||
|
|||
Arny Krueger wrote: (snip) To summarize, the so-called objectivists listed above have made a lot of progress, while leading self-proclaimed subjectivists are still basing all their current work on 27 years old poorly-done blind and sighted evaluations. Despite all the scientific progress that has been made, it remains true that there is no evidence for the idea that good power amps sound the same, despite present and future claims in Stereophile. (snip) What you seemingly don't want to admit John is that you still have your head in the sand, and there still is no evidence that good power amps sound different as generally used. No evidence for same... no evidence for different. Plenty of evidence that Arny is retarded, but John still can't quite secure his position in this debate. Sad... very sad for both sides. ScottW |
#249
|
|||
|
|||
John Atkinson said:
Don't forget that in the first round of trials, before Howard Ferstler decided to randomize his scoring as described above, proper statistical analysis revealed that he did indeed hear a difference between the amplifiers under test. However, he rejected that result because it didn't not conform to his preconceived notions, surely the mark of a "true believer." :-) Agreed.™ Harold's cause is informed by religious fervor. Clerkie's professed love of "science" is fine as far as it goes. When it comes to subjecting himself to the rigors of "testing", however, clearly he's perfectly willing to substitute his faith for the much-lauded "facts". Fortunately for his own life and liberty, Harold has, to date, stopped short of calling for the destruction of the E.H.E.E. temples of iniquity. As he draws ever closer to his final sundown, I wouldn't be surprised if he gets even more radical in his ranting. |
#250
|
|||
|
|||
"MINe 109" wrote in message ... If you're misunderstood, maybe you should try to be clearer. You appear to be saying in this thread that Clark's "demonstration" was self-evidently bogus and should have been rejected at sight Come on Stephen, not even you are *that* reading-challenged, right? and also saying that because it falls short of being a good test it wasn't a test at all. Ditto. In your words Stephen, what I said is that Clark's AES demo was self-evidently not a true scientific test. |
#251
|
|||
|
|||
In article ,
"Arny Krueger" wrote: "MINe 109" wrote in message ... If you're misunderstood, maybe you should try to be clearer. You appear to be saying in this thread that Clark's "demonstration" was self-evidently bogus and should have been rejected at sight Come on Stephen, not even you are *that* reading-challenged, right? and also saying that because it falls short of being a good test it wasn't a test at all. Ditto. In your words Stephen, what I said is that Clark's AES demo was self-evidently not a true scientific test. "What this quote does prove is that my friend David Clark, who is not lacking in persuasive powers, failed to get that wiley old fox named John Atkinson to do a statistically significant number of trials, which would be required for this to even be a test. It was obviously a demonstration! The fact that Atkinson represents this demonstration as a test speaks to his poor understanding of scientific testing methodologies." Obvious, self-evident, potato, po-tah-to. If it's not a test, the results should be discarded (or "rejected") immediately, yes? Stephen |
#252
|
|||
|
|||
"MINe 109" wrote in message ... In article , "Arny Krueger" wrote: "MINe 109" wrote in message ... If you're misunderstood, maybe you should try to be clearer. You appear to be saying in this thread that Clark's "demonstration" was self-evidently bogus and should have been rejected at sight Come on Stephen, not even you are *that* reading-challenged, right? and also saying that because it falls short of being a good test it wasn't a test at all. Ditto. In your words Stephen, what I said is that Clark's AES demo was self-evidently not a true scientific test. "What this quote does prove is that my friend David Clark, who is not lacking in persuasive powers, failed to get that wiley old fox named John Atkinson to do a statistically significant number of trials, which would be required for this to even be a test. It was obviously a demonstration! The fact that Atkinson represents this demonstration as a test speaks to his poor understanding of scientific testing methodologies." Obvious, self-evident, potato, po-tah-to. Exactly.; If it's not a test, the results should be discarded (or "rejected") immediately, yes? Not necessarily , since the results could still be part of an interesting, even instructive anecdote. |
#253
|
|||
|
|||
MINe 109 wrote: If it's not a test, the results should be discarded (or "rejected") immediately, yes? According to the Book of Howard, Stephen, there is a complicated protocol to be followed in deciding whether or not to reject the results of a blind test. First, the question must be asked: did the test produce null results? If yes, then the the test should _never_ be discarded. Instead, some explanation should be offered for not doing so, such as saying "Not necessarily, since the results could still be part of an interesting, even instructive anecdote." Remember, achieving a null result, according to the apocrypha of the Book of Howard validates the test methodology no matter how flawed. :-) If the results were _not_ null, ie, appeared to reveal that the listeners could identify the devices under test by ear, then the test should be rejected forthwith without any equivocation. First, you claim that the statistics were invalid. If they are valid, then you claim that the program was inappropriately revealing. If a number of musical examples were used, suggesting that the devices were identifiable under many circumstances, then you claim that the test was not meant for publication but was merely a magazine reader's letter. If no-one was impressed by that argument, then you attack the integrity of the test's organizer, perhaps drawing an analog with "Clever Hans," the horse that could read its trainer's body language. If that doesn't work, you declare that the positive results "prove" that the test can't have been blind. If no-one is convinced by any of these arguments, then you merely declare that the test never happened :-) Or you change the subject by claiming that the tester attacked your religious preferences. Or your wife. John Atkinson Editor, Stereophile Stephen |
#254
|
|||
|
|||
In article .com,
"John Atkinson" wrote: Or you change the subject by claiming that the tester attacked your religious preferences. Or your wife. And if you say it three times, it's true: "Demonstration, demonstration, demonstration!" Stephen |
#255
|
|||
|
|||
"Howard Ferstler" wrote in message ... Unless the Adcom has some gross design problems (unlikely) or is defective or broken (also unlikely), it should sound as good as any other good amp up to its overload point. Don't let these idiots give you a bum steer about amplifier sound. It's funny how one person's failure to hear differences between amplifiers implies that other people are idiots. It might be a good thing if you bothered to at least do some level-matched DBT work with amps before you assume that your preconceptions are valid. The comparisons are not all that hard to do. That's right, it's not. So, it might be a good thing if you took your swelled head out of your ass and stop making idiotic assumptions. |
#256
|
|||
|
|||
"Robert Morein" wrote in message ... Howard, if I were to admit you were 25% right, I would be doing the audiophile community a disservice. Like Arny, you have been polarized by the absurdity of "snake oil", to the extent that you are really blind to the many subleties that actually do exist. Together with Arny, you constitute the French Terror of audio. You willingly put to the torch any idea that doesn't fit into your scheme of things. You may kill afew rats, but it isn't worth the many fine houses you destroy in the process. Well put. |
#257
|
|||
|
|||
jeffc wrote:
"Howard Ferstler" wrote in message ... Unless the Adcom has some gross design problems (unlikely) or is defective or broken (also unlikely), it should sound as good as any other good amp up to its overload point. Don't let these idiots give you a bum steer about amplifier sound. It's funny how one person's failure to hear differences between amplifiers implies that other people are idiots. It might be a good thing if you bothered to at least do some level-matched DBT work with amps before you assume that your preconceptions are valid. The comparisons are not all that hard to do. That's right, it's not. So, it might be a good thing if you took your swelled head out of your ass and stop making idiotic assumptions. Whatever. Howard Ferstler |
#258
|
|||
|
|||
John Atkinson wrote:
wrote: Howard Ferstler wrote: Let's be realistic, most serious high-end audio buffs are true believers who have no particular interest at all in doing careful comparing. Let's consider Howard's idea of "careful comparing": "I just kept pushing the button and making wild stabs at what I thought I heard. After a while, I did not bother to listen at all. I just kept pressing the same choice over and over." Don't forget that in the first round of trials, before Howard Ferstler decided to randomize his scoring as described above, proper statistical analysis revealed that he did indeed hear a difference between the amplifiers under test. Yeah. That really had me going. I suddenly realized that unlike you and some of your fellow Stereophile writers, I am a true golden ear. Howard Ferstler |
#259
|
|||
|
|||
Robert Morein wrote:
"Howard Ferstler" wrote in message ... Robert Morein wrote: [snip] Let's be realistic, most serious high-end audio buffs are true believers who have no particular interest at all in doing careful comparing. They have a faith to defend. Howard Ferstler Howard, if I were to admit you were 25% right, I would be doing the audiophile community a disservice. Like Arny, you have been polarized by the absurdity of "snake oil", to the extent that you are really blind to the many subleties that actually do exist. Whatever. Howard Ferstler |
#261
|
|||
|
|||
Harold "Professional Writer" Ferstler waxes prolix. That's right, it's not. So, it might be a good thing if you took your swelled head out of your ass and stop making idiotic assumptions. Whatever. Glad you agree with his assessment. |
#262
|
|||
|
|||
Brother Horace the Everlasting Naysayer said: I suddenly realized that unlike you and some of your fellow Stereophile writers, I am a true golden ear. Is that why you rammed the icepicks into your ears? |
#264
|
|||
|
|||
John Atkinson wrote:
MINe 109 wrote: If it's not a test, the results should be discarded (or "rejected") immediately, yes? According to the Book of Howard, Stephen, there is a complicated protocol to be followed in deciding whether or not to reject the results of a blind test. First, the question must be asked: did the test produce null results? If yes, then the the test should _never_ be discarded. Instead, some explanation should be offered for not doing so, such as saying "Not necessarily, since the results could still be part of an interesting, even instructive anecdote." John, anybody who has done a level-matched DBT involving decently built amps (or wires, or CD players) will have to be immediately struck by the fact that, even if they end up maybe (just maybe) hearing slight differences in terms of the statistical outcome, the devices being compared will still sound a hell of a lot more alike than what they may have thought they sounded like prior to doing the DBT. They do not need any statistical analyzing to realize that fact. When doing a DBT with good components, the single most profound thing that occurs is that the participant will realize that the units sound maddeningly alike. My problem with you and your people (and your magazine's approach) is that they typically evaluate amplifiers (or wires, or players) in an open-ended manner, and come to conclusions about the sound that show one of the units to be monumentally different from (and therefore either monumentally superior to or inferior to) the other unit. For you, your contributors, and your readers, this seems to show that any DBT protocol is flawed, simply because those differences just HAVE to be there. Well, I disagree. To be truthful, I am more convinced than ever that you are with me on this and that the only reason you behave the way you do is that you have a reading clientele that demands that differences be profound. You have my deepest sympathies. Howard Ferstler |
#265
|
|||
|
|||
John Atkinson wrote:
Arny Krueger wrote: Those were about a failure by Zipser to prove Zipser's hypothesis that the amps sounded mind-blowingly different. But the Zipser test clearly failed to meet the criteria for a properly designed blind test. As you said above, Mr. Krueger, it is therefore "reasonable" to dismiss the results of that test as meaningless. So what? Zipser had claimed that he would be able to EASILY hear differences. He could not, even when comparing his big reference monoblocks to a mainstream Yamaha integrated unit. Even if Zipser had eventually been able to spot subtle differences (possibly related to background hiss), the fact is that those big, high-end-related differences he believed existed did not exist at all. Howard Ferstler |
#266
|
|||
|
|||
MINe 109 wrote:
In article , Howard Ferstler wrote: wrote: Howard Ferstler wrote: Robert Morein wrote: "Howard Ferstler" wrote in message ... Let's be realistic, most serious high-end audio buffs are true believers who have no particular interest at all in doing careful comparing. Lets consider Howard's idea of "careful comparing." I just kept pushing the button and making wild stabs at what I thought I heard. After a while, I did not bother to listen at all. I just kept pressing the same choice over and over." Sure. What else was there to do when the amps sounded the same? You would probably do the same thing, but then you would go on to claim that the switch box was masking the differences. That's a clear admission of "test anxiety" on your part. Stephen Anxiety would exist if I had believed that amps sounded different and the test was proving my preconceptions wrong. No doubt when guys like you get involved in such demonstrations the anxiety levels go through the roof. No, I simply realized all over again something that I had known for years: the amps sounded the same and doing a lengthy and precise series of comparisons would be a waste of my time. On the other hand, a series like that would be just the ticket for a believer like you. Howard Ferstler |
#267
|
|||
|
|||
In article ,
Howard Ferstler wrote: MINe 109 wrote: In article , Howard Ferstler wrote: wrote: Howard Ferstler wrote: Robert Morein wrote: "Howard Ferstler" wrote in message ... Let's be realistic, most serious high-end audio buffs are true believers who have no particular interest at all in doing careful comparing. Lets consider Howard's idea of "careful comparing." I just kept pushing the button and making wild stabs at what I thought I heard. After a while, I did not bother to listen at all. I just kept pressing the same choice over and over." Sure. What else was there to do when the amps sounded the same? You would probably do the same thing, but then you would go on to claim that the switch box was masking the differences. That's a clear admission of "test anxiety" on your part. Anxiety would exist if I had believed that amps sounded different and the test was proving my preconceptions wrong. Anxiety exists when a test subject can't concentrate and stops listening in order to get the whole thing over with. No doubt when guys like you get involved in such demonstrations the anxiety levels go through the roof. This anxiety might show itself in unexpected ways. No, I simply realized all over again something that I had known for years: the amps sounded the same and doing a lengthy and precise series of comparisons would be a waste of my time. That last part's true whether one accepts the premise or not. On the other hand, a series like that would be just the ticket for a believer like you. Who, me? $250 amp guy? Stephen |
#268
|
|||
|
|||
Howard Ferstler wrote: John Atkinson wrote: wrote: Howard Ferstler wrote: Let's be realistic, most serious high-end audio buffs are true believers who have no particular interest at all in doing careful comparing. Let's consider Howard's idea of "careful comparing": "I just kept pushing the button and making wild stabs at what I thought I heard. After a while, I did not bother to listen at all. I just kept pressing the same choice over and over." Don't forget that in the first round of trials, before Howard Ferstler decided to randomize his scoring as described above, proper statistical analysis revealed that he did indeed hear a difference between the amplifiers under test. Yeah. That really had me going. I suddenly realized that unlike you and some of your fellow Stereophile writers, I am a true golden ear. No, you are hearing impared. You are a fraud and a plagiarist though. Scott Wheeler |
#269
|
|||
|
|||
"Howard Ferstler" wrote
Sure. What else was there to do when the amps sounded the same? You would probably do the same thing, but then you would go on to claim that the switch box was masking the differences. Are you saying it's not possible that a musical reproduction systems has a subconscious effect on a person? |
#270
|
|||
|
|||
In article .com, "John
Atkinson" wrote (in reply to Arny Krueger): I fear you don't seem to have comprehended what I wrote about the 1988 Clark AES test, Mr. Krueger. My correctly pointing out that scoring 5 correct out of 5 trials, as Michael Fremer did, doesn't reach the 95% confidence level is factually correct. Period. Wrong. Getting 5 correct in 5 trials is a significant result at the .05 level (actually p = .03125) if the test is one-sided, i.e., a test with null hypothesis that the subject is guessing with Pr(correct score) = .5 versus the alternative that Pr(correct) .5 . However, if the test is two-sided (so the alternative is Pr(correct) != ..5), then p = .0625 and the result is not significant at level .05. Usual practice for ABX-style tests has been one-sided tests. In either case, the issue in a statistical test is "significance' rather than "confidence". When someone uses 95% confidence as if it were the same as 5% significance it is obvious that he does not understand what he is talking about. JC |
#271
|
|||
|
|||
MINe 109 wrote:
In article , Howard Ferstler wrote: Anxiety would exist if I had believed that amps sounded different and the test was proving my preconceptions wrong. Anxiety exists when a test subject can't concentrate and stops listening in order to get the whole thing over with. Particularly when they cannot hear differences from the get go. No doubt when guys like you get involved in such demonstrations the anxiety levels go through the roof. This anxiety might show itself in unexpected ways. Those who are sure of themselves would not have anxiety at the beginning. That anxiety would only show up when they began to realize that they could not hear differences. By then, however, it would be too late, because they also could not hear differences when they were confident. On the other hand, a series like that would be just the ticket for a believer like you. Who, me? $250 amp guy? OK, by your answer I am going to assume that you agree with me about the sound of overpriced amps. Howard Ferstler |
#272
|
|||
|
|||
In article ,
Howard Ferstler wrote: MINe 109 wrote: In article , Howard Ferstler wrote: Anxiety would exist if I had believed that amps sounded different and the test was proving my preconceptions wrong. Anxiety exists when a test subject can't concentrate and stops listening in order to get the whole thing over with. Particularly when they cannot hear differences from the get go. Frustrating, yes? No doubt when guys like you get involved in such demonstrations the anxiety levels go through the roof. This anxiety might show itself in unexpected ways. Those who are sure of themselves would not have anxiety at the beginning. That anxiety would only show up when they began to realize that they could not hear differences. By then, however, it would be too late, because they also could not hear differences when they were confident. You persist in assuming anxiety is rational and limited to audiophiles. Your impatience with the test, combined with your a priori assumptions, is a form of anxiety. On the other hand, a series like that would be just the ticket for a believer like you. Who, me? $250 amp guy? OK, by your answer I am going to assume that you agree with me about the sound of overpriced amps. I've never heard an overpriced amp. Stephen |
#273
|
|||
|
|||
Brother Horace the Prayerful chanted: Anxiety exists when a test subject can't concentrate and stops listening in order to get the whole thing over with. Particularly when they[sic] cannot hear differences from the get go. Emended version, as taken directly from "Ferslerianism for the Masses and Misses," Vanity Press, 1993: "Particularly since I cannot hear differences from the get go." -- H. Ferstler, verbatim, ad nauseum, ad infinitum You can't even quote yourself accurately without plagiarizing. |
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
Upgrading My Adcom Preamp & Amp | High End Audio | |||
FS: ADCOM GTP-400 Preamp/Tuner | Marketplace | |||
FS: ADCOM GTP-400 Preamp/Tuner | Marketplace | |||
Adcom GPT-450 pre-amp/turner Humming sound | Tech | |||
Adcom CD player, help needed | Tech |