James Randi vs. Anjou Pear - once and for all

Metro04, do assume that DBTs or beyond scientific criticism. Many would claim that an invalid measure as most commonly used, namely 30 second exposures and that setting the testing hypothesis as no difference violates the logic that one cannot accept an hypothesis, but one can reject an hypothesis. This is the reason we normally seek a "null hypothesis" that a variable thought to be causing a relationship has no effect. Rejecting it then lends support for the relationship you wish to support.

Carl109, if you were to take say 10 of the cables that audiophiles revere plus one bought from Radio Shack and test them as you propose and find no differences even with the most accurate measurement instruments we have, what would you conclude? I would think there are but two alternatives. Either there are no differences or the measurement lack necessary precision or focus on the wrong things.

While I do think I have to rely on what I hear, I agree with you that tests such as Randi proposes are invalid measures of what we want in realistic music reproduction.

This thread makes no sense. The failure of anyone to accept the Randi challenge shows nothing except the inability to accept the methodology of the test. The fact that both sides think the method determines the result to me at least, means that the method, not whether there are differences among cables is all that is in question. As such, nothing will ever be proven or disproven.

I must say that I cannot understand how someone who says all cables sound alike picks cables.

Muralman1, the Mikro-Omega top interconnects have been using very thin ribbons with a clear plastic surrounds for at least ten years. They sound very neutral. Omega Mikro firmly believes in as little insulation as possible. The were the originators of ribbon power cords. They also strongly believe in designing with awareness of the direction of the wire or ribbon draw.

Brizonbiovizier, no one needs to prove anything to someone else. We all make our own buying decisions. I my case I buy what I think sounds good. Why would you think that someone needs to prove anything to you?

Shadorne, I certainly have known people who buy equipment given its manufacturer's name or because of bells and whistles, but most I know are indifferent to this, at least as I judge the appearance of their gear.

I have no objections whatsoever to people comparing gear behind a curtain, but I think DBT's 30 sec. exposures are invalid.

Finally, very little medical research deals with placebos. That there is centers on evaluating new drugs. Most medical research is simple statistical analyses looking focusing on what contributes to those having a disease.

Brizonbiovizier, I buy on belief as do you. Beware that it is not just significance, it is statistical significance as in is it possible that our random sample came from a population where there was no relationship between variables. What you are talking about is that it is very improbable, not that it is statistically significant, much less meaningfully significant.

I agree with you that room treatment is often neglected, but I certainly would not include digital correction for the room in what needs to be done.

Why do you use expensive connectors on you cables? How do you know they make a difference? You use cables because they were provided and claim cables make no difference without any observations to backup that conviction. This is quite contrary to the scientific method. One seeks a theory that accounts for the observations rather than avoiding observations because they are contrary to the theory.

brizonbovizier, all I can say is that we differ greatly on what is science and statistical significance, and how we personally choose components. I think that my listening pleasure would be sharply reduced were I to be driven in making decisions on audio based on limited engineering concepts that you view as fundamental. I think EE has very limited understanding of why things work and uses a fundamentally unscientific "good enough" perspective on circuits, parts, wire, and even what is safe.

What you say about statistical significance is in error. As I said, you are talking about probabilities, not statical significance. To use significance is to make believe that you are talking about meaningful significance when you are not.

Brizonbiovizier, wrong on both counts. Yes, electronic components work, but that does not mean that we really know what variations might improve them. Certainly, to think that resistance, capacitance, and inductance are all that matters is simplistic. What about the geometry, what about the insulation, and what about the impact of RFI and EMF?

You are right about probabilities and statistics being linked at a fundamental level but it is still the case that statistical significance is the proper term, not significance as the single word believes the fact that what is statistically significant may be of little value. If you don't realize that statistical significance can be greatly enhanced by increasing the size of the sample, your understanding is deficient.

Statistical significance center entirely on the question of whether sampling error could mean that our random sample based findings could with a certain probability have come from a population where there was no relationship between the variables in our hypothesis. We are willing to risk type I error and reject our null hypothesis lending credence to our hypothesis if the probability of such an unfortunate sample is below typically below 5 in a 100 samples.

I have a PhD in both political science and psychology where I got most of my methods training both in experimental and sample based research. I have taught research methods and statistics for 40 years and published broadly in refereed journals, not that this really matters.

I still say that your dismissal of the possibility that wires sound different is inherently unscientific and unjustified by anything you mention. Furthermore, I cannot understand why any audiophile would not avail themselves of the opportunity to listen first before dismissing possible improvements.

Brizonbiovizier, I think you hurt what case you had by your condescension. I suggest you get off your high horse and read something about sampling theory and statistics. Phrases such as "much the merriment of true statisticians" suggest that you are quite insecure in your competencies and certainly not scientifically trained. You claim that no one has ever produced any rigorous evidence that has stood up to scientific scrutiny. Well I would claim that I doubt if you would know anything about this, that it is just rubbish.

Objectivists always claim the scientific high ground and often I have learned do so without justification. It certainly is not scientific to forego listening to find if your cherished "theory" continues to hold. I doubt if anyone reading this thread is any better off for doing so.

I like it, Leica man.

Brizonbiovizier, why don't you take the Randi challenge? What if you heard a difference? Of course, you could just randomly say different or not different and assure you fulfilled your prior conception. This is the fallacy of DBTs. The real question is whether most people can hear a difference in a test where people could not lie to fulfill the hypothesis. Were a random sample of 1500 to have enough who heard a difference to achieve statistical significance, we would reject the null hypothesis that people don't hear a difference. This would be good research not Randi's game.

Winstonsmith, yours is a very informative post. Thanks for the information you bring to bear, which highlights the controversy well. I must say, however, that I wonder why you went so far as to say, "If people want to waste their hard-earned cash on such products, let them. If they think they heard a $34k difference, I'm not going to tell them otherwise." I heard the Wavelength $350,000 amp at CES and regretted that I did. It was incredible and there was no way I could even consider it. I don't think necessarily that people are "wasting" their money, especially on say the Siltech cords. If they can afford them, more power to them, especially if they can hear the benefits. I do know some who buy for appearances, but I know very few of these.

Nevertheless, your post does attack the irrational and illogical arguments of many of those who hear no differences. I have always wondered why they bother posting as they convince no one. You may well have the answer.

I am always, unhappy that they seek to wrap themselves in the cloak of science when few have any understand of what it means other than a few concepts in electrical engineering and a vague notion of what the placebo effect is about and of concern to some scientists.

Shadorne, more information please. I don't know what a capitalized blind test means. What did he do as a result of not being able to "discern" between the cables. I suspect nothing. Perhaps he was more able hear a difference in the sighted test if the "blind" test was the usual 30 second, same/different test. This is the reason why I think these DB tests are invalid.

I would not be too critical of Winstonsmith until someone shows that there are not variations in the ability to discern differences. It seems to be unreasonable to expect that everyone's aptitude is the same as it is not on all other measures.

Dlanselm, was this DBT a 30 sec. exposure same/different format? You seem happy to have seen the light. That is all that matters.

This thread has taken a less threatening direction than most of those dealing with the usual "prove it sounds better" versus "I hear one is better" threads. I think this is largely because of Winstonsmith's use of DIY cables and good defense of the importance of achieving accurate sound. But in reality Dlanselm's personal experience really has no impact. Most of us still proceed with using the components and wires we hear as best and can afford.

Personally, I see no benefit with double blind audio tests, but single blind tests are fine. Who cares if the person running the test knows which wire is which? I have participated in several of these with in one case where a Bozak preamp was preferred by several prominent designers of other preamps. But these blind tests all used matched volumes but lengthy listening comparisons. Also none of these were seen as anything more than entertainment. I am certain no one ran out and bought a Bozak.

I have also participated in short duration same/different DBTs. I know the confusion that comes with these tests. I certainly found them curious and hypothesized why the strange results. Then I realized that the tests must invalidly capture what we mean by sounding different, especially in one where most of us then compared the two amps involved still in a blind comparison but with longer listening comparisons and the opportunity to go back and forth between them. We all now preferred one of the amps. To me this is a more valid comparison and certainly yielded different results.

Again, all of this is largely entertainment of little relevance to a buying decisions. Dlanselm obviously did change his mind as a result, but I doubt many of us would. I certainly found no interest in buying the Bozak although I did seek it out several times at audio stores. I also did not buy any of the other preamps in the test.

Why there is this constant debate about DBTs perplexes me. Frankly, I have more trust in my ears than I have in others.

Well, Mrtennis, all I can say is that I have no formula. I began all of this interest in the middle 1960s. Since then I have had 27 different speaker systems, 21 amps, 19 preamps, 10 turntables, ? cartridges, and 47 different ac filters or isolation transformers. Generally each has been an improvement. It has been expensive but that is also part of the fun. Speakers are the exception. I have concluded that there are only speakers that are better in some regards; none that meet all that I would desire.