|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
173.215.206.130
So you read the title and are thinking "oh jeez, another anti objectivist rant that will probably lead to some bull***t excuse for avoiding bias controls.
Not this time.
This is about often infered meta-analysis of the alleged body of research on all things that "supposedly" make no difference in audio (high end snake oil). How often do we hear something akin to "the studies prove that such and such does not make a difference." or "There is not one case of such and such ever tesing for a positive result under bias controls?" I see these assertions made on a regular basis in one form or another. But what is the body of collected evidence that anyone can use to draw such conclusions? We don't have any peer reviewed published data base on amplifier sound or CDP sound or cable sound etc etc. So what is it that these objectivists who make or infer these beliefs about the "collective database" are using? How do they know there are no tests that have wrought a positive result with one thing or another what are they doing to collect a database and make a meta-analysis of it?
All of this was spawned by two threads on other forums. on Rec Audio High End there was a report of a positive ID of differences among DACs in an ABX DBT
http://groups.google.la/group/rec.audio.high-end/browse_thread/thread/17c82ae8c89c4106#
and on Hydrogenaudio there was a report of a positive result in comparing 44.1 Kh sampling rate to the same signal up sampled to 88.2 Kh.
http://www.hydrogenaudio.org/forums/index.php?showtopic=82264
In the thread on RAHE I raised the point of cherry picking when some objectivists make assertions about what the "body of evidence" shows us about the sound or lack there of of any particular point of interest often debated between objectivists and subjectivists. I found this terific article that pretty much makes my point about the bias of the meta-analysis we get from so many objectivists who make claims about evidence.
http://skepticwiki.org/index.php/Systematic_Review
" For in order to apply meta-analysis to a set of papers, it is first necessary to select the papers; and for the results to be worth anything, it is necessary to choose the papers in such a way as to avoid both junk science and cherry picking.
To overcome this problem, scientists have developed a technique known as systematic review, in which clearly stated and pre-determined criteria are used to select the papers for review. This has two advantages: first, it eliminates opportunities for conscious or unconscious bias to creep in; secondly, a clear set of criteria can be scrutinized to see if they are sufficient for the task."
"What is important is that you select your system in advance, and don't tinker with it by throwing in other papers that you think might be useful but which were not turned up by your system. For such practices will tend to introduce bias: it is likely that your favorite papers are those that confirm the opinions you already have. And even if you were, in fact, perfectly free even from unconscious bias, which you are not, then even so, no-one else would be able to accept your results as an objective, systematic search of the literature. By having a set of objective criteria for your search, which you decide in advance and publish along with your meta-analysis, you can convince others that you were objective in selecting your data; also, if it happens that there is something wrong with your method of selecting the papers, then by having a definite protocol that you've published, others can scrutinize it for flaws."
"it is important that you determine your criteria for weeding out flawed papers in advance. You might think that it would be a good idea to scrutinize each paper for specific methodological errors, rather than working from a check-list. However, there is a snag with that. It has been shown that people, including scientists, are much better at spotting methodological flaws in papers the results of which don't fit their preconceptions. This would result in unconscious cherry-picking if you looked for unique flaws in each paper. Or, again, if you looked at the results of the papers first, and then decided how long-term was long-term, then you might, consciously or unconsciously, decide on a limit that included some paper that produced results particularly congenial to your way of thinking."
In both threads we see extensive analysis and criticism of protocols, methodologies and statistical analysis. How often do we see the same objectivists taking such a critical look at tests with results that jive with their expectations? I asked someone to provide just one example of the same level of scrutiny applied to tests that wrought a null and none was provided.
If we take the famous amplifier shootout reported in Stereo Review back in 1985 we find all kinds of problems in protocols, methodologies and statistical analysis. they were in short really bad tests. what I have yet to find with this particular highest of profile ABX DBTs is any of the same in depth analysis and criticisms of the protocols and statistical analysis from objectivists that we see in the two threads cited above in which tests with positive results have been reported. in fact I have yet to find *any* examples of any of these objectivists ever making such in depth analysis of protocols and statistical analysis against *any* ABX DBTs in audio that resulted in a null.
I think this is a classic case of bias affecting the meta-analysis of an undefined body of evidence in audio. IOW me thinks that many objectivists are indeed cherry picking. ironic that many of those who wave the science flag and are so concerned about bias effects are actually being quite unscientific and quite affected by their own biases.
Long winded, I know. Sorry for that.
Follow Ups:
It's just an absurd philosophy, and no good philosophers take it seriously.
Well said.
Why any self respecting subjectivist/relativist audiophile would give a hoot about objectivist arguments/attacks is beyond me. We should ransack useful tools and leave them blabbering on thier own.
Remarkably enough, both the Stanford Encyclopedia of Philosophy and the Internet Encyclopedia of Philosophy have articles on Ayn Rand:
http://plato.stanford.edu/entries/ayn-rand/
http://www.iep.utm.edu/rand/
Neither has convinced me that there should be articles on Ayn Rand in philosophical dictionaries.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Are you related, per chance, to the 14th century French philosopher Jean Buridan? :-)
I was amused by Burdan's Principle after reading Carver Mead's book on VLSI design and Leslie Lamport's paper.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
"It's just an absurd philosophy, and no good philosophers take it seriously."
Oh Well... Who is John Galt?
R.
With more information and/or a link thereto, your post would be worth of more serious attention. :-)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Because that's what your post amounts to. Some inmates have made quite a career of attacking things they attribute to a group of people they call objectivists.
One big problem is that you simply present assertions you have alleged are made repeatedly, but provide no data, no criteria. You have your opinion, of course, but that's all it is. Where's your meta-analysis?
You then give a couple of URLs to discussions you seem to think of as examples. You show no sign of understanding the discussions very well and expect the rest of us to come to the same conclusions about them that you have, which is rather silly of you. Where's your own meta-analysis?
I read the HO thread down to where Arnie Krueger makes what seem to me to some perfectly technical objections to the procedure. Whatever his biases are, the objections he made about different sampling rates seem rational, and he knows much more about the subject than I do and I dare say than you do. He offered a rational explanation of why the results came out positive. Either refute his objections and show the test is valid, or, failing that, the only thing to do is run a better test.
If you don't like null results in some test, the only thing to do is either show the difference is in fact audible (JA did this recently here. JA sometimes whines about what ABX advocates say but he can also attempt some rational arguments.) or do a blind test that should be more sensitive. Whining about his alleged biases doesn't make the test any better.
As for the skeptics site, Tony Lauck has quite properly pointed out that if one doesn't know the field, one may not know what advance criteria should be set for a meta-analysis of the literature.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Am I psychic? Nah, you are just that predictable.I would argue your points.... if you had made any.
What points are you *trying* to make?
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Just the one I made in my OP
Something you posted to me in Critics really hit home. I've known it but it's significance never really registered. The two sides have been arguing the same issues since before 1985 (as evidenced by your link to a Brad Meyer letter to Stereophile concerning blind tests). Let's say it's been since 1980. That's 30 years. What has been accomplished?
Is there any argument I could make that would convince you of my position? Is there any argument you could make to convince me? Very unlikely, particularly since it's been going on 30 years with neither side budging.
My question - not just to you but to everyone on both sides - is, what's the point? Why not just be secure in one's own position?
It's been fun and occasionally funny, but spinning wheels is a pretty unproductive use of our time.
believe it or not I have become an advocate of blind auditioning. It was "objectivists" that lead me to read the scientific literature on bias effects. I think of all people JJ has learned a lot from subjectivists and has come to understand the merits of euphonic colorations through rigorous testing rather than just dismissing such preferences so often held by subjectivists out of hand.
Sure we can just throw our hands up and say we disagree. But we learn nothing from one another that way. I actually think these things are worth investigating and debating.
The conflict between my perceptions and what always seemed like a very reasonable perspective of objectivism never really sat well with me when I got into high end audio. In all other fields I would probably be considered an objectivist and skeptic. I think I have actually learned a lot in trying to sort out these conflicts.
This post was just one of many observations I have made in sorting out where each side tends to go wrong generally speaking. The only way you have a standing fued for 30 years is if both sides are making gross errors and both sides are refusing to consider the other side's points. Some say just walk away. I say jump in and learn.
Obviously such critical examination of one's personal beliefs is scary at times. look how much Pat gets shaken up by posts like this one. He will probably make 20 or more posts on this thread before it is done all in an attempt to obfuscate because the idea that his guys might actually be doing something wrong scares the crap out of him. I feel sorry for Pat actually. He's going to be in the mud arguing for god knows how much of his life time and never learn a thing from all of it.
I have learned a lot from objectivists in trying to resolve (at least in my own mind) the conflicts between the two camps. But I have also learned a lot *about* them as a group along the way including the major flaws in their belief systems.Oh and the same for subjectivists. Lots flaws found there as well. Lots.
Of course the biggest lesson was and still is 90% of the debate is ego based. No one can learn anything from the other side until they clear that hurdle.
Now I'll sit back and watch Pat obfuscate for another 10 or 15 more posts.
nt
One could say the same about all sorts of things. Controversies over astrology go back many centuries. Creationism has been argued for one and a half centuries, at least.
Meanwhile, people spend all sorts of money on them, just as many others spend lots of money on expensive interconnects, speaker cables, power cords, brilliant pebbles, intelligent chips, and so on. Some just want to see what the evidence actually supports.
Meanwhile, the advances in audio have been made by people who care about sound, who measure, develop and test hypotheses, and who look into psychoacoustics.
"Is there any argument I could make that would convince you of my position? Is there any argument you could make to convince me? Very unlikely, particularly since it's been going on 30 years with neither side budging."
30 years isn't all that long in history. It's not so much a matter of arguments but evidence. For example, you show me products whose performance differs by the amounts shown in the ABX matching criteria, and I'll accept some can detect the differences.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Wow Pat, really? On this thread of all threads?
"For example, you show me products whose performance differs by the amounts shown in the ABX matching criteria, and I'll accept some can detect the differences."
I showed you two of them in my OP.
Let the flawed meta-analysis begin again on the thread about flawed meta-analysis of many objectivists. I am enjoying the irony of it. call it a guilty pleasure. Please Pat, entertain me some more. tell us what you don't like about cited ABX DBTs that wrought undesired results. Please, unwittingly engage in cherry picking right before our eyes on a thread about cherry picking.
Pat D
"For example, you show me products whose performance differs by the amounts shown in the ABX matching criteria, and I'll accept some can detect the differences."AS
"I showed you two of them in my OP."Really? What were the differences in the frequency responses?
The point of my remark was that if the FR differences were large enough, blind testing would not be needed to convince me they sounded different. We would already know the differences are audible based on previous blind tests.
What are undesired results in DBTs? DBTs results are data. They are what they are. Data needs to be interpreted and sometimes interpretations are made that are too strong. In one respect, I agree with Tony Lauck, in that I don't primarily look for null results, I look for positive ones that can stand scrutiny.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Edits: 07/23/10
"The point of my remark was that if the FR differences were large enough, blind testing would not be needed to convince me they sounded different. We would already know the differences are audible based on previous blind tests."
Did you write that?
You are so funny.
Translation: if others tell me that the differences are large enough, then I'll say they are large enough. :)
rw
before you guys look at it! I already linked it, too.
If the frequency response differences are greater than shown on the curves, then the differences should be audible for someone.
If the frequency response differences are under amounts indicated on those curves, then they may or may not be audible, so I would want a blind test to confirm any claims they are. Of course, there may be other things which would indicate an audible difference, noise, for instance, but then I would want to see that shown, too, before I accept the claim.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Here's a hint: it isn't the "substance" of your point....
rw
... I'm sure you know that. You don't want the process to stop - you're just pretending to. Otherwise, what would you post about in Prop Head - that... what's that called... "technical" stuff???
.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
I mean, we all know what passes for your logic, your hearing, and your ability to learn - but that's a new low.
"I think this is a classic case of bias affecting the meta-analysis of an undefined body of evidence in audio. IOW me thinks that many objectivists are indeed cherry picking. ironic that many of those who wave the science flag and are so concerned about bias effects are actually being quite unscientific and quite affected by their own biases."
Even if I agree to some degree with your comments, understanding of course the broad generalizations mixed in with the pigeon holes, I fail to see any importance or ramification to them at all.
Is there some position you're are attempting to forward contrary to the "objectivists" you so obviously have intended to slander.
"Even if I agree to some degree with your comments, understanding of course the broad generalizations mixed in with the pigeon holes, I fail to see any importance or ramification to them at all."
I am not responsible for your failures.
"Is there some position you're are attempting to forward contrary to the "objectivists" you so obviously have intended to slander."
Um, the one you quoted...."slander?" Isn't truth the ultimate defense against accusations of slander? I rest my case on the evidence. If, *if* you actually understand my argument and disagree with it certainly you can provide some evidence in the form of objectivists scrutinizing an ABX DBT that had a null result. Ideally you would find something from the usual suspects. I don't say that *all* objectivists are guilty of cherry picking. If you want me to name names I can if it will help.
"I am not responsible for your failures."
Indeed you are not. Like I said I pretty much agree with your comments but at best it's only one side of a multi-faceted story.
Without your POV as well as the other facets of the story the comments are hardly worthy of interest much less excitement.
Maybe this is somekind of epiphany for you but believe me people have chosen what facts to use to defend their position as long as there has been rational thought.
"I don't say that *all* objectivists are guilty of cherry picking. "
Yes I noted your careful use of the term "many" in your OP.
All I can say is this is just the typical vomit posing as critical thinking on an audio website.
Due to your overly broad generalization by the misuse of the term "objectivist" your bias preceeds any rational comment you could have made. You're not fooling anyone except those who want to be fooled.
puke.
Isn't that what you pretty much always end up saying? Time for the pony to learn a new trick.You are becoming as predictable as Pat. Not a good thing.
Edits: 07/22/10
...is a movement toward some sort of claims being made in the context of almost anything other than 'sighted' "listening tests."
Most objectivists would start the party with any claim being based on a listener not needing to know exactly what gear he was hearing before being able to identify its characteristics.
Seemingly not too hard, but seemingly impossible.
Maybe reviewers can start faking it to please the objectivists and make the reviewer seem more golden eared.
;D
How many "order of magnitude night and day several veils lifted" sonic differences seem to be struck down by the dreaded 'blind listening deafness?'
All?
... is that results of ANY test, blind or sighted, pertain to only ONE listener that was subject of the test, and only ONE system, that test was performed on.
Of course, there are couple of pre-requisites:
- understanding that audio systems sound different (shouldn't be as difficult as it actually is for some here);
- understanding that hearing differs significantly among humans, especially in the high frequency range, which contributes most to discernable sonic differences.
As soon as they're able to wrap their heads around that concept, their wishes regarding "movement toward ..." will immediately become moot point to them - as they already are now to everybody else. Any kind of movement can be performed ONLY by the same person that wishes for that movement - as simple as that.
P.S. I hope not to see any mention of IRRELEVANT medical DBT in responses to this post, if there are any.
It seems they want more when they don't like the results
It's pretty basic in science. Why should anyone accept some anomalous result without replication?AS
"It seems they want more when they don't like the results"Sounds like many subjectivists--except they really don't want any real tests at all, let alone more tests.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Edits: 07/21/10
either you understand why it is cherry picking to only scrutinize tests that give positive results while giving all nulls a free pass or you don't understand it. The explination is clearly spelled out in my OP.
If you want to wave the science flag you sure can't make this argument which is a classic logical fallacy.
http://www.theskepticsguide.org/resources/logicalfallacies.aspx
"Tu quoque
Literally, you too. This is an attempt to justify wrong action because someone else also does it. "My evidence may be invalid, but so is yours."
If you want to wave the science flag you can't cherry pick nor can you accept others cherry picking.
At least the subjectivists who actually don't want tests aren't hypocritically waving the science flag.
"either you understand why it is cherry picking to only scrutinize tests that give positive results while giving all nulls a free pass or you don't understand it. The explination is clearly spelled out in my OP"
Actually, your OP is anything but clear.
Even as you state it here, it is simply a personal attack, whether correct or not. It has absolutely nothing to do with whether the criticisms made of various tests are valid or not.
As well, if a DBT test was competently performed, it provides data, even if you don't like the statistical analysis or some of the conclusions drawn.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
"Actually, your OP is anything but clear."
I thought the article I cited on meta-analysis was very claer. I suggest you go back and read that article I cited on meta-analysis and then tell me what you did and did not understand. We can go from there.
"Even as you state it here, it is simply a personal attack, whether correct or not."
No Pat, my assertion of misuse of meta-analysis on the part of many objectivists is not a personal attack. it is a well argued critique of a pathology I see in many objectivists in audio and I supported it with some clear cut examples.
"It has absolutely nothing to do with whether the criticisms made of various tests are valid or not."
Red herring. My point was about the approach to meta-analysis of the individuals not the specific analysis of the specific tests being scrutinized in the two cited threads. Again, I suggest you read the article on meta-analysis I cited. If you don't understand the points raised in that article or how they directly apply as I am asserting they apply I'll try to explain it to you.
You tried to support your thesis by referring to internet discussions of a particular experiment. However, those discussions brought up rational criticisms of the way the experiment was done. Your examples are red herrings, since they do not seem to illustrate whatever point you want to make.
Moreover, you did not in fact analyze those discussions, any more than you supported your assertions about how some of those whom you call objectivists proceed.
It is possible you may be right about some objectivists, but you really have done nothing to show this. You just assert your belief on the matter.
If you wanted to discuss meta-analysis, then the article is a good starting point. You would have been better to stick with that rather than engage in unsupported attacks. However, as Tony Lauck has pointed out, the article on meta-analysis is somewhat flawed. For example, just picking out criteria in advance as to what studies to include is not sufficient, as one must know enough about the field to pick relevant criteria.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
That being meta-analysis. Nothing in your post addresses meta-analysis. review that and get back to me.
"Your examples are red herrings, since they do not seem to illustrate whatever point you want to make."
So you don't know what point I am trying to make yet you claim my examples don't illustrate this point that is unknown to you? I hope even you can see the irony in that.
"Moreover, you did not in fact analyze those discussions, any more than you supported your assertions about how some of those whom you call objectivists proceed."
Well gosh Pat, if you understood the article on meta-analysis you would understand that there was no reason to analyze those discussions. the mere fact that they can only be found with undesired positive results is the point that is so well described in the article on meta-analysis. Again I suggest you read that article and get back to me when you understand it. Here is the part that basically explains it all. "Again, it is important that you determine your criteria for weeding out flawed papers in advance. You might think that it would be a good idea to scrutinize each paper for specific methodological errors, rather than working from a check-list. However, there is a snag with that. It has been shown that people, including scientists, are much better at spotting methodological flaws in papers the results of which don't fit their preconceptions. This would result in unconscious cherry-picking if you looked for unique flaws in each paper." now if you really can't understand this very well written excerpt from a very well writen article on meta-analysis then...well it's just over your head. sorry.
You then pick something out of it and tell us that most of what was in your original post is irrelevant.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Anything to avoid dealing with my points eh?
I find it entertaining to be skeptical of the skeptics. The cited web page offers a number of good targets. :-)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
I think objectivists can't wrap their heads around exagerrated claims made under sighted conditions that subsequently are rendered inaudible due to blind listening deafness.
A decent attempt at making a repeatable claim would be an awesome start for the delicate subjectivists who can hear whole new layers of detail so long as they know they have Valhalla cables hooked up, but can't tell it apart from Rat Shack magnetic wire when the chips are down.
As for the objectivists who demand instantaneous DBT - screw them, too. Take your time, look for subtle sonic cues at different listening points in a performance...enjoy...then see if those night and day differences are real! Subjectivists need to challenge themselves a little - I think much of where they spend time listening is "I'm so special land" and too many objectivists toss rocks over from "nobody can hear that-ville."
I'm going to frame this.
Did you read the OP or the threads I cited? No sighted tests involved in this discussion.
OCD objectivists, adamant subjectivists....I think in Hi Fi there is ample middle ground that remains unexplored, so your rant renminded me of that.
Sorry if you thought my post drifting.
A bit but no worries. I'm almost looking forward to seeing what bizarre tangent Pat will try to take this thread to
The last time I heard a night and day difference was when I replaced my Sony rack system with an "audiophile approved" one... which didn't cost that much more. That was in the late 1980's. Nothing I've auditioned since then was "night and day". I think your point is well taken. Too much rhapsodizing, not enough real testing. Under blind testing, even the biggest differences faded for me. They didn't disappear in many cases, but they were not as noticeable as when I auditioned sighted. Most of the time the diffs were not worth the additional $$ to me.
That said, I've heard systems that were considerably better than anything I've ever owned. Not night and day but maybe 6 a.m to 10:30 a.m? lol.
I agree.
I'm tired of 'scientists' wasting my tax dollars on their 'big bang' B/S.
Rick
nt
Could be...
I haven't been following this one as arguing about arguing leaves me cold but when I started on AA they were all the rage. JJ and others arguing about, about, who knows? Probably DBT's, subjective/objective, the weather in Malaysia...
Congratulations by the way, I guess you've gone as far right as a person can get! I think the John Birch electronically monitors for that so you will likely get a reward of some sort in the mail soon. Maybe a small statuette or a certificate suitable for framing.
Nothing else like being a marginal inmate at the Audiophile Asylum!
Rick
One can't replicate past events, such as Caesar crossing the Rubicon or the Big Bang, but one can re-examine the data available or try to replicate the experiments or do other experiments which bear on the issue.
For example, Banks and Krajicek were not satisfied with the design and set up of Stereophile's blind test of two amplifiers. So they administered a smaller scale but much better set up blind test with different speakers and were able to confirm that the two amplifiers did sound different in the tests. Given that one was a tube amplifier with a high output impedance and the other was a solid state amplifier with a low output impedance, this was hardly surprising.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
The infamous ABX DBTs of amplifiers from Stereo Review back in 1985? Clearly one of the mosty cited set of ABX DBTs in the history of the great debate. that test wrought a null between a SS amp and a a 50 watt OTL on a very difficult load. You don't think that was an "anomolous" result? Why no mention of it Pat? I think I actually cauht you doing the very thing I wrote about in the OP. You are cherry picking. That is ironic.
Now let's see if you can find any evidence that any of tyhe objectivists who like to scrutinize positive results of ABX DBTs because of alleged "anamolous" results have scrutinized that most famous of ABX DBTs due to the "anamolous" results. there is plenty to scrutinize in those tests by the way.
Knock yourself out Pat. I look forward to your obfuscation of my points. You have 14 more posts to go before you hit 20 on this thread.
I don't even have a copy of it.Also, I know that some amplifiers do indeed sound different from some other amplifiers. DBTs have shown some high fidelity amplifiers sound different when driving complex loads such as found in many speakers. That has not prevented a number of people from saying that others believe every amplifier sounds the same--it's a big straw man thrown around in controversy. Some throw that idiocy at me from time to time.
In the June 1991 issue, Stereo Review published an article by E. Brad Meyer called "The Amp/Speaker Interface: Are you loudspeakers turning your amplifier into a tone control?" A tube and a solid state amplifier were tested with two different speakers. One speaker offered a difficult load to the amplifier, one did not (the loads are not specified). With pink noise, the tests showed to a very high degree of probability that the differences between the two amplifiers were audible using either speaker.
With the music selections, the tests showed audible differences between the two amplifiers to a high degree of probability using the speaker with a difficult load. Using the speaker presenting the easier load, the tests results did not disprove the null hypothesis using those music selections, though it's always possible that with other music or other listeners the differences would have been shown to be audible. In any case, they were shown to be audible with pink noise.
Stereophile and Soundstage show the frequency responses of the amplifiers they measure into a standardized dummy speaker load. Some of the mainstream magazines did so, too. Quite a while ago, I emailed Doug Schneider of Soundstage and suggested that they include a graph of the load offered by the dummy speaker load used by BHK labs in amplifier tests. He agreed and the graph showing impedance vs. frequency and electrical phase vs. frequency is now shown at the following URL as part of the explanation of how they measure amplifiers:
http://www.soundstagemagazine.com/measurements/test_amplifiers.htm
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Edits: 07/22/10 07/22/10
at least you are consistant.
LOL
You don't even seem to know what "to cite" involves!
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
1. This citation is in the OP
"If we take the famous amplifier shootout reported in Stereo Review back in 1985 we find all kinds of problems in protocols, methodologies and statistical analysis. they were in short really bad tests. what I have yet to find with this particular highest of profile ABX DBTs is any of the same in depth analysis and criticisms of the protocols and statistical analysis from objectivists that we see in the two threads cited above in which tests with positive results have been reported. in fact I have yet to find *any* examples of any of these objectivists ever making such in depth analysis of protocols and statistical analysis against *any* ABX DBTs in audio that resulted in a null."
This is what the word citation means as I used it in the previous post.
http://www.merriam-webster.com/dictionary/cite
"to bring forward or call to another's attention especially as an example, proof, or precedent"
have fun finding a new tangent of obfusecation on this one Pat. ::snicker:;
.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Is this what it has come to Pat? is this some sort of panic attack?
Where did it appear?
You appear to know practically nothing about it, and certainly do not have first hand knowledge of it.
Yes, I am playing with you. I'm sorry you don't seem to like it. If you had been nicer, I might have taken pity on you. But then, maybe you'll learn something.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
http://bruce.coppola.name/audio/Amp_Sound.pdf
The wrong date and the lack of title and authors made it more difficult for me to find, but I located it last night.
Bruce Coppola would count as a (minor) objectivist who relies on that article and the conclusions drawn from it.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
since upon return I can always count on finding you up to your usual pedantic posturing and boorish behavior.
Thanks for the good chuckle at your expense.
Hint: there is no article such as he identified it.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
by being the guy who rests his arguments on technicalities.
Don't pretend you don't know what he's talking about regarding the Stereophile amp tests...
Thank you for your kind remark. You may well be correct. For example:RY
"Don't pretend you don't know what he's talking about regarding the Stereophile amp tests..."Who is talking about Stereophile blind amp tests? Not me or AS.
Well, maybe you just didn't follow the . . . uhhh . . . discussion. But I'll give you the benefit of the doubt: perhaps you really meant to type Stereo Review.
See the importance of clearly identifying something, including an article, that one wants to talk about? It's pretty basic. I might help AS out if he asked nicely.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Edits: 07/24/10
you are up to 22 now and have hit some pretty amazing highs in obfuscation including the deliberate attempt to derail things over a misremembered date. BRAVO!!! you exceded my expectations. I'm done toying with you now that you passed my prediction. The rest is just icing.
You know, the ones mentioned in the OP and which you later said exceeded known audible thresholds shown on the ABX site? The information I asked you for?
http://www.audioasylum.com/forums/prophead/messages/5/58746.html
Since you did not know who wrote the article, nor the title, and got the date wrong, that made it somewhat more difficult for me to find it. Once I did find it, it was clear that you did not have the article in front of you since you had the date wrong, and also did not have the contents right. Since you did not know who wrote the article or what the title was, it is no wonder you have never seen it criticized, so your attempted meta-analysis in your OP was so much hooey.
If you want a link the 1987 article in SR by Clark and Masters, I have found one and it is below. Read away to your heart's content. As I surmised, the program material was not as good for the purpose as it might have been--even Stereophile later discovered thick stuff like choral music made for a more sensitive test. I am surprised they did not use pink noise, too.
Oh, BTW, this means I have actually found an objectivist who relied too much on a superseded test, but so far, you have not, though you maintained you knew this. But really, you and JA should get over an old test done 23 years ago.
I cited a later article in SR by E. Brad Meyer in which he showed an audible differences between a tube amp and a SS amp driving two different speakers, though with one of the speakers, the difference was only shown to be audible on pink noise. You and JA really should get over that old superseded test.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
that there have not been any sonic improvements with amplification stages in the past twenty-five years. Do you ever get out and sample what is available?
rw
Totally unrelated to the discussion.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Nothing changes at all in a quarter of a century with respect to electronics.
rw
I am not sure where you get that--certainly not from me. There are a number of amps nowadays with very low S/N rations, very low distortion, and also a number that can drive very low impedance loads. I dare say there are many more available now than there were 25 years ago. Under what conditions these make an audible difference is another question.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Amplification stages have come a long way in the past twenty five years since Stereo Review tested the question of audibility. An SP-11 was a good preamp, but no match for a REF5. The same can be said for amplifiers.
Under what conditions these make an audible difference is another question.
That would be playing music. Without any of the built in crutches required by Clark's ABX testing.
rw
A back a few posts, you break in with some really off the wall statement about something you attribute to me with no justification.
"I am truly very sorry if you believe that there have not been any sonic improvements with amplification stages in the past twenty-five years."
Where on earth did you get that?
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Did you actually read the test to which you linked? Do you really think there have been no sonic improvements at the high end over the past quarter century? If so, then I'm really sorry to hear that your exposure is limited. If not, then you will understand how stupid referring to such a dated test is.
rw
I caught Analog Scott out on two things.
One is an blind amplifier test published in Stereo Review which he kept saying was published in 1985. I quickly found out that it was published in 1987, which meant that he was discussing an article which he did not have in front of him. We both managed to find the text on line and he then admitted the point. I did not defend the results in any way, shape or form.
You are correct that the article in question has been superseded, but not for the reasons you suggest. In Stereo Review for June 1991, there is an article by E. Brad Meyer, "The Amp/Speaker Interface: Are your loudspeakers turning your amplifier into a tone control?" In this article, an unspecified solid state amp and an unspecified tube amp were tested driving two unspecified speakers, one with a difficult load an the other with an easier one. Some FR measurements were shown. The blind tests revealed that the differences between both amps was audible on pink noise with both speakers. With two music selections, however, the differences were shown to be audible with the amps driving the difficult speaker load, but with the easier load, they failed to prove the difference between the two speakers was audible. Obviously, the program material used can affect the outcome of such tests, and it is possible that with other music, maybe the differences between the two amps would show positive results when the amps drove the easier load. In any case, the pink noise test was sufficient to show there were audible differences between those two amps when driving those two speakers.
Since I have seen the results of E. Brad Meyer's test and other tests, I do not ordinarily question whether a tube amp sounds different from a solid state amp since it is reasonably probable they do. I think E. Brad Meyer deserves credit for publishing why many tube amplifiers do not sound the same as many SS amplifiers. It is simply their high output impedances. Audio magazine began showing the responses of tested amplifiers into a dummy or simulated speaker load, and Stereophile started to do so a few years later.
To get to the second thing. In his OP, AS mentioned a couple of threads on other sites which he thought illustrated his view that at least some of those he calls "objectivists" make a biased selection of published test results and don't criticize the ones that achieve null results. He may or may not have a minor point, but the threads for which he provided URLs contained quite rational criticisms of a positive DBT and a systematic survey of the literature was not needed. When I pointed this out, AS decided the threads he used to try to illustrate his thesis were irrelevant!
I remarked in a post that if the differences in the frequency responses of two products differed by more than the limits shown in the ABX Matching Criteria on the ABX site were exceeded, a blind test would not be needed to convince me the differences were audible under some conditions because a threshold has been previously established. However, AS jumped in and asserted that he had already shown two examples of this in his OP, so I simply asked him what those differences were. He has not replied. So I caught him out again.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
> > One is an blind amplifier test published in Stereo Review which he kept saying was published in 1985. I quickly found out that it was published in 1987, which meant that he was discussing an article which he did not have in front of him.> >
When did I ever say I had it in front of me?
> > We both managed to find the text on line and he then admitted the point.> >
"admitted the point?" What point? There is no point there Pat. The article is real, It clearly describes highly flawed methodologies and clearly published very unlikely results. And we have no evidence of any objectivists scrutinizing this very high profile test and dismissing it.
> > I did not defend the results in any way, shape or form.> >
Who said you did? I suggest you reread my OP and see if you can figure out the problem here as it relates to meta-analysis commonly applied by so many objectivists when it comes to DBTs. With the 1985 Stereo Review article we have one of the highest profile examples of ABX DBTs in the history of the great debate. We also have a null which is a a dubious result given the presence of a 50 watt OTL.
The problem I am pointing to in the meta-analysis of so many objectivists is that I can find a multitude of examples of objectivists scrutinizing and then dismissing DBTs that wrought a positive result. One can not find the same criticism of said tests that wrought a null from objectivists. This pattern clearly fits the description i provided of poor meta-analysis. I simply cited the Stereo review article because it was so high profile that it should have had a slew of objectivists scrutinizing it and dismissing it had the objectivists been doing propper meta-analysis.
Where is the scrutiny of this test by the same objectivists who are scrutinizing DBTs that wrought positive results? Cherry picking......
OK now what was the other thing you "caught me out on?"
I'll just sit back and watch you obfuscate again.
Well, when you want to discuss an article, it behooves you to know who wrote it, what the title is, and what it said. You showed no sign of knowing anything of the sort. As for the flawed methodology and unlikely result, you haven't shown either one, though I happen to agree it could have been more sensitive. But since you did not know the particulars of the article, you could not possibly show that the 1987 SR amp test is high profile, and that it has not been criticized.
I cited a 1991 article in Stereo Review by E. Brad Meyer which showed audible differences existed between a SS amp and a tube amp, how audible depending on the speaker loads. Moreover, he showed why they sounded different.
The second thing? It is as I stated: You tried to illustrate that some objectivists, unnamed and unquantified, questioned the results of a positive DBT for no good reasons. However, your illustration fell flat because they did in fact come up with rational reasons why the test results should be questioned--in other words, it seems likely that the positive results simply reflected the test set up.
"To get to the second thing. In his OP, AS mentioned a couple of threads on other sites which he thought illustrated his view that at least some of those he calls "objectivists" make a biased selection of published test results and don't criticize the ones that achieve null results. He may or may not have a minor point, but the threads for which he provided URLs contained quite rational criticisms of a positive DBT and a systematic survey of the literature was not needed. When I pointed this out, AS decided the threads he used to try to illustrate his thesis were irrelevant!"
This reduces your allegation about cherry picking to mere assertion without evidence.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
"Well, when you want to discuss an article, it behooves you to know who wrote it, what the title is, and what it said."
Really one has to know the title? No Pat, all one needs to know in this case were the relevant facts. I'll list them for you.
1. It was ABX DBTs
2. The results as reported were a null
3. Given the presence of a 50 watt OTL amp with the load presented by the speakers used this is a very unlikely result.
4. The methodologies were a mess.
5. One can not find any scrutiny of this test anywhere online by any objectivists despite being one of the highest profile tests of it's kind.
I knew those 5 points from memory and rereading the article confirms those points were accurately remembered. And that is all I needed to make my point in the OP.
"You showed no sign of knowing anything of the sort."
Then you can't read the signs Pat.
" As for the flawed methodology and unlikely result, you haven't shown either one,"
Really? What aren't you getting about the problem with a 50 watt OTL not giving a positive result with the speakers used?
> > though I happen to agree it could have been more sensitive.> >
Clearly if it missed the obvious differences one would hear with a 50 watt Futterman and a cheap SS amp from the 80s.
" But since you did not know the particulars of the article, you could not possibly show that the 1987 SR amp test is high profile, and that it has not been criticized."
That is some pretty piss poor logic there Pat. again I would point you to the five key point above. they are all that matter in this case.
"I cited a 1991 article in Stereo Review by E. Brad Meyer which showed audible differences existed between a SS amp and a tube amp, how audible depending on the speaker loads. Moreover, he showed why they sounded different."
And this does what other than support my assertion that the results of the 1987 ABX DBTs wrought unlikely results?
> > The second thing? It is as I stated: You tried to illustrate that some objectivists, unnamed and unquantified, questioned the results of a positive DBT for no good reasons.> >
No I did not try to do that Pat. Again I suggest you read the article I cited on meta-analysis since this explains my actual point rather than the one you seem to be imagining here. I never commented on *the content* of these two threads that scrutinized the ABX DBTs that wrought positive results.
You maintain a number of things about the 1987 test but have not established any of them. You haven't shown why we would expect the Futterman amp to sound different--though from what I have seen somewhere, it should sound different under some common circumtstances.
Also, we have only your word that is a high profile article.
META-ANALYSIS?
Since I have shown that Stereo Review published an article in 1991 by E. Brad Meyer showing that a tube and a SS amplifier were audibly different with a speaker load. I should also point out that Stereophile did a very poorly set up mass blind test between a tube and a SS amplifier with barely positive results which readers showed were somewhat ambiguous, but Banks and Krajicek ran a smaller but much better set up blind test using the same two amplifiers and achieved a much less ambiguous positive result. Since the results reported in the 1987 SR article have been superseded, one wonders why you think it is so important. You give no reason whatever why subsequent testers would want to reexamine that old article except for historical purposes.
METHODOLOGY
You complain that:
"4. The methodologies were a mess."
But you do nothing to establish that. The results could have simply statistical fluke for all you have told us.
Since you maintain that "the methodologies were a mess," you should be able to tell us what they should have done better, but you aren't saying. Until you do, your complaints about the methodologies are simply your unsupported opinion.
-Why do you think the result of part of the test comparing the tube amp to the SS amps achieved a null result?
-What should the testers have done to make the test more sensitive?
I'm only trying to teach you something about how to support your positions.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
When you have a counter argument to any of my points let me know.
And that's what I have shown. If you want to make your points, it's up to you to argue for them.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
and that is what I have shown. And these personal attacks prove you have no argument that i have no argument.
Score one for the argument that "This Will Never Be Resolved". ;)
i just wonder if he got the jokes.
Is there some form of measurement for the jokes? Has it been determined that the jokes are "get-able"? Evidence required. ;)
That made me laugh at 5 in the morning. But it was under sighted conditions so I don't know if I really laughed or imagined it
I keep wondering if people in South Central L.A take the time to DBT gunfire or if they just take cover.
There is an end. One can invoke the Kleene Star operator.
For more on this subject read the following:
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
...wishing for 10 beers and then all non-biased thought went straight out the window. :)
Thanks - this was quite good. lol
Godel Escher Bach is one of my favorite books, but it requires a lot of determination to make it all the way through, especially if you follow all of the serious parts in detail and work out the puzzles.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
...which he kept saying was published in 1985. I quickly found out that it was published in 1987
You really got him there.
...a blind test would not be needed to convince me the differences were audible under some conditions because a threshold has been previously established.
Yeah, your wording was an absolute hoot.
rw
and he STILL doesn't get it.
Pat has brought the joy back to Propheads for me.
I am glad that you enjoy the simple pleasures. :-)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
since having to put the system into storage while my wife and I look for a home in DC, I've had to learn to enjoy the absurdities of the asylum just a little bit more...
you rely on typos to make a point??
You have fallen so far, in such a short time. I hope you don't even try to communicate with friends. Part of the social contract that underlies general social conventions is an ability to look past minor and/or irrelevant failings in order to participate.
Please Pat, I followed the conversation all along (yes, I can read, even when you can't see or know that I am following along. Cool thing about the 'net, eh?). What I have noticed (and I notice this a lot with you) is that you have managed to push the conversation so far away from its intended point(s) that it is just another episode of The Pat Show: Teflon Debater.
And I blame you for causing me to blow coffee out of my nose this morning.
The intended point? GMAB.
1. In the OP, AS asserted that some objectivists relied on articles without seeing if they could be criticized. He asserted some did but AS offered no evidence. He gave a couple of examples, which I showed did not establish his point, since the threads included quite rational criticisms of a DBT run at McGill U.
2. AS showed the URL of an article on meta-analysis, one that is a sort of a good beginning but which has some severe limitations as pointed out by Tony Lauck. AS tried to deny he had any other purpose, in effect denying no. 1 and most of his OP. But it is clear a main purpose was to attack (some) objectivists.
Of course, AS also kept talking about "undesired" results of DBTs, which is the fallacy of poisoning the well, rather than unexpected results. Meanwhile, he persisted in making unjustified and irrelevant personal attacks on me.
So, in fact, I systematically analyzed his OP. Moreover, I did, in fact, find an objectivist who did rely on the superseded article. AS did not, I did.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
especially about this one:
"Meanwhile, he persisted in making unjustified and irrelevant personal attacks on me."
It's you against the world Pat. Don't you ever forget it.
"It's you against the world Pat. Don't you ever forget it. "
You, AS, E-stat, and carcass93 are hardly "the world!" GMAB
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
It's a world of people discussing differences in sound of the preamps and amplifiers on Amp and Pre asylum. Take a note - differences, not whether differences exist.
It's a world of people discussing differences in sound of transports and DACs in Digital, cables in Cables, computer settings in PC Audio - and so on, and so forth. In general, it's a world of audiophiles, who strive for best sound possible, discussing audiophile matters, and chuckling at few lost anti-audiophiles preaching stagnant mediocrity.
In a forum that's supposed to be audiophile, no less.
It's also nice to be able to recognize them.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
> It's also nice to be able to recognize them. <
Definitely! But what's preferable to illusions is reality, particularly the reality that I'm told has to be illusion. It's a good way to confirm that we don't already know everything there is to know.
kind of pathetic.
Have you ever heard the expression (and its many variations) "its us against the world?" Because once again you missed the point of the barb.
.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
:)
The British magazine Hi-Fi Choice has been doing blind listening evaluation of products for 30 years now.
Donald North
From what I can tell from a few reviews in Hi-Fi Choice, their blind tests amount to a panel listening to a system including the components and answering a questionnaire. They don't seem to establish their panel members can actually distinguish the components and give no statistics on the number of trials and how many right and wrong answers.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
To me their testing is representative of people actually listening to the components and commenting on their experiences/observations. I've been reading the magazine for nearly 20 years and for an individual component there seems to be some consensus on fidelity but they also note when someone experiences it differently.
Donald North
"To me their testing is representative of people actually listening to the components and commenting on their experiences/observations."
Yes, I agree. That is precisely why I have a problem with it.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Seems to me we have a Hobson's choice here, between an arranged marriage based on the fact that the prospective husband has 2 cows and 1/4 acre, and eloping with a biker dude.
We are talking about people who offer professional reviews of audio equipment! This is not about consumers making up their minds as to what they prefer.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
I think that blind testing is excellent in that circumstance, since it removes a well-documented source of confirmation bias. Whereas what would be the purpose of ABX testing? It seems to me that in most cases it would be impractical, obscure audible differences of the sort that become apparent with extended listening or occur only with certain combinations of program material, and serve no useful purpose. Only if there was real controversy as to whether one could hear something would I bother with ABX testing, and even then, I wouldn't make the mistake of supposing that the results are completely reliable, anymore than I would make that mistake about a subjective evaluation.
Why impractical? Other reviewers have done DBTs in the same situation in which they did their sighted listening evaluations.
In this circumstances, DBTs are a method of proof (loosely speaking) that audible differences exist, rather than a method of discovery. Why would one do a DBT if audible differences were not thought to exist?
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
But I don't think that proof of audible differences is required in most cases. Really, when you think about it, the number of components for which there are credible allegations that there are no audible differences is very small. Mostly fringe stuff, like dots. Maybe in some cases cables and power cords. In those cases, I think ABX tests might help prove that differences do exist, although their utility in proving that they don't is as far as I can tell pretty marginal.
Another use for ABX tests might be in determining just what is audible, as in the Clark amplifier tests. It would certainly be useful to know that the audible differences in amplifiers are due to impedance matching and power. I know that I'd love to see too if demonstrated audible differences in cables could be attributed to shielding and RLC, as I suspect, or whether it's oxygen free copper, dielectrics, etc. as some claim.
In addition to blind and ABX tests, I'd like to see tests like distortion listening. Compare what's at the output of an amp with a real world load with what's at the input. Record the difference. Try superimposing it on the original signal to see if it has an audible effect and see whether we can hear the distortion spectrum of a good amplifier with real world transducers.
"Why would one do a DBT if audible differences were not thought to exist?"
Here are two reasons:
1. It is possible to note differences in bind testing that are not consciously audible. The perception comes as a bias in the statistical distribution, so lots of samples will be needed.
2. When one does both blind and non blind tests one learns about one's own perceptions and mind. This is useful knowledge, even if purely personal.
Some of us are interested in direct knowledge. We don't want a bunch of priests or other "authorities" telling us what to believe, especially if they come bringing magic boxes or graphs.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
"Some of us are interested in direct knowledge. We don't want a bunch of priests or other "authorities" telling us what to believe, especially if they come bringing magic boxes or graphs."Along these lines I don't want middle aged + old men telling me what I can't hear because their mind and/or acuity is limited. I had one such encounter 16 years ago on my first job interview, just to witness 5 years later this "authority" has lesser perceptual ability himself.
Donald North
Edits: 07/26/10
You are quite correct in general, but I was thinking of more informal tests done by reviewers or even consumers who wished to do them.
"1. It is possible to note differences in bind testing that are not consciously audible. The perception comes as a bias in the statistical distribution, so lots of samples will be needed."
jj has pointed this out here, that one may not consciously be aware of a difference but the results may show that one actually can detect the difference.
I'm not sure how this would apply to tests by reviewers or consumers who wished to do blind tests, but it is quite relevant to research.
"2. When one does both blind and non blind tests one learns about one's own perceptions and mind. This is useful knowledge, even if purely personal."
I do wish more reviewers would learn about their limitations.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
"one may not consciously be aware of a difference but the results may show that one actually can detect the difference.
I'm not sure how this would apply to tests by reviewers or consumers who wished to do blind tests, but it is quite relevant to research."
The significance of this is deep, because it shows that the relationship between consciousness and perception is complex. If one can "hear" things as proven by blind statistical tests that one can not "describe" consciously then one should not be too quick to dismiss the reverse possibility, that one can "describe" differences that one can not "prove" by blind statistical tests.
"What is dogma?
Dogma is an idea with a rigid boundary line, which won’t allow you to go beyond the periphery of that boundary line. Thus dogma goes against the fundamental spirit of the human mind. The human mind won’t tolerate anything rigid. It wants movement – not only movement, but accelerated movement. Humans want unbarred psychic progress, intellectual progress, without obstacles."
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
"The human mind won’t tolerate anything rigid. It wants movement – not only movement, but accelerated movement. Humans want unbarred psychic progress, intellectual progress, without obstacles."
"
Ahhh not all of them Tony, not all of them. ;-)
"Ahhh not all of them Tony, not all of them. ;-)"
The mind can be blocked by negative emotions. Fear will do it, for sure.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
"What is important is that you select your system in advance"
If you are trying to avoid fooling yourself perhaps this advice has meaning, but if you are trying to do a meta-analysis that will persuade others it is not sufficient to select the system in advance. It is also necessary to provide a convincing argument that your system was actually selected in advance . You can say that was your procedure, but in this day of questionable scientific integrity you will have to convince your readers that was what you actually did.
I don't see how you can provide any evidence that this was your procedure. Indeed, if one is dishonest it would be entirely possible to do a preliminary survey of the literature and come up with a set of criteria that gives the desired results, remember these criteria and then trash all the preliminary work. At this point one would begin the "real" work and just happen to start with "developing" the pre-arranged criteria.
As you can see, I am highly suspect of meta-analysis. They are subject to being gamed and the complex statistics are easily abused. If there is a well defined and reputable collection of studies, then there won't be such a problem. However, I don't believe this is the case in the audio field.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
What you are describing is basically a cheat.
As it stands there is a lot of evidence IMO of cherry picking among many objectivists in their informal "meta-analysis." I suspect if we start calling them on this some of them will do exactly what you describe and essentially try to cheat.
What I described is indeed a cheat, if done consciously. But it is also possible to do it unconsciously. If one is more or less familiar with the literature and one has preconceived notions of "truth" one may well have an unconscious tendency to invent criteria that rules in one's favorite research and rules out one's disliked research without even realizing what one is doing.
One can avoid this problem (but not be able to prove that one has avoided it) by coming up with the criteria before reading any reports on the subject. Unfortunately, this is unlikely to be very effective, because if one has no familiarity with previous work in the field one is unlikely to even be aware of the problems one should be looking for. One is trying to be an "unbiased expert" but there is no such thing as an unbiased expert.
Hence the debate goes on and on....
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
I can agree with that.
Cheers,
Presto
nothing wrong with completely agreeing with *yourself.* At least you are not feeling conflicted
Did any of the objetivists even understand the point you wre trying to make A.S.? Seems like all they did was attempt to defend their postion and beliefs!
Thetubeguy1954 (Tom Scata)
SETriodes Forum -- Central Florida Audio Society -- Fullrange Drivers
==============================================================
"The man that hath no music in himself nor is not moved with concord of
sweet sounds is fit for treasons, stratagems and spoils."
- William Shakespeare
I think Arny did. But so far none have admitted that they may be cherry picking.Well, no, one person admitted to cherry picking but it seems he didn't realize that he had made such an admission.
The possibilities are:
1. the skeptic says there are no tests showing any evidence
2. the skeptic says the test that is cited was incorrectly conducted, dishonest, or otherwise irrelevant
3. the skeptic says that the test was a statistical fluke
4. the skeptic says that the set of tests were preselected and hence the meta analysis is invalid.
5. ...
999. the skeptics eventually die off and progress ensues.
You can see these possibilities playing out in the threads that were linked above.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
You link to Kuhn's The Structure of Scientific Revolutions on Amazon. Kuhn has been very influential, but, of course, has his critics. For example, some have pointed out that rather than some revolutionary change in paradigm initiating discoveries, some discoveries made through normal science initiate revolutions. The discovery of the double helix structure of DNA is an example.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
It would seem to be a valid point. But on the other hand, as Schumann remarked, "Genius creates kingdoms, the smaller states of which are again divided by a higher hand among talents, that these may organize details which the former, in its thousand-fold activity, would be unable to perfect." Pretty good statement of Kuhn's philosophy 100 years before he came up with it, no? In any case, this particular criticism of Kuhn seems to me to belong in one of Schumann's smaller states.
Thanks for the link. I was not aware of the site, which looks extremely interesting.
I am aware of the debate concerning Kuhn, but I cited his book for the benefit of those who might not be familiar with it, particularly with regard to falsifiability.
In your reference it says, "Kuhn's account argues that resisting falsification is precisely what every disciplinary matrix in science does. Even disciplines that could not claim to be dominated by a settled paradigm but were beset by competing schools with different fundamental ideas could appeal to Kuhn's description of the pre-paradigm state of a science in its infancy. Consequently Kuhn's analysis was popular among those seeking legitimacy as science (and consequently kudos and funding) for their new disciplines."
This is precisely my point. Many of the audio skeptics operate just as described. This is one reason why I would never attempt to collect Randi's $1,000,000 were I to have an experiment that was apparently suitable. I have no doubt that some excuse would be found for either rejecting the experiment up front, sabotaging it so it would fail, or if it apparently succeeded finding an excuse for not paying.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
The Stanford Encyclopedia of Philosophy is indeed a good site.
Another one is The Internet Encyclopedia of Philosophy, linked below.
I don't think all the articles in them are of equal quality, but then that is par of the course with encyclopedias.
__
"Always be sincere, whether you mean it or not."---Flanders & Swann
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: