[b0dhi]

Quote:

Originally Posted by royalcrown /img/forum/go_quote.gif b0dhi: ...

I agree with much of what you say here. I generally do give credence to blind tests, where they're used correctly. A blind test is also extremely useful as a training tool. I hope my post didn't come across as favouring sighted tests - I consider sighted tests almost useless as "tests". So my criticism is not of blind tests as such, but only how they are mis-applied as support for conclusions they are not designed to elucidate.

Quote:

That being said, however, it's equally silly to assert with confidence that the experience is caused directly by something in the external world - whether that be a cable, an amplifier, or whatever - without compelling evidence that it is actually caused by said object.

I agree. The cause, I think, is the fact that most people don't realise that their senses are only representations of the real world that can sometimes be wildly incorrect (LSD anyone?). On balance of (incomplete) knowledge, that person does see compelling evidence for their belief, so I can't really fault them for anything other than ignorance.

Quote:

"Throwing out one's senses without sufficient reason is supremely foolish, and I would say veers well into scientism." I highly disagree, and it's not a matter of scientism, it's a matter of Cartesian epistemology, which is still (IMO, and according to most analytic philosophers I've worked with) sound. In fact, he makes a pretty compelling argument that most (if not all) knowledge that we think stems from sensory experience is in fact determined by the 'faculties of the mind' (i.e. using logic/rationality and not empirical data). See Descartes's wax argument.

I'm tempted to respond to this in depth but these discussions tend to expand and consume all in their path, so I'll just say that I'd like to emphasis the "without sufficient reason" in the quote above

 

[randomasdf]

Quote:

Originally Posted by FirebottleRon /img/forum/go_quote.gif Find one that proves they dont.

Talk to any electrical engineer.

 

[royalcrown]

Quote:

Originally Posted by FirebottleRon /img/forum/go_quote.gif Sounds like some of you guys just have a problem thinking you can know what other people can hear, or not hear. Did any of the science buyers ever find a test that proves all cables sound alike? I think I will continue to be a listening buyer and hear for myself what makes a difference and what does not. I listen to my system to enjoy music, if fornicating with cable testing is more fun for you than why not just stick with iBuds? SmellyGas, can you show us a test preformed by audio engineers, statisticians, researchers, and other people with doctorates that proves that every cable sounds alike? Or is talking about it all you can do? Seems like a common problem lots of people have when it comes to cables.

If you check other threads, I've posted numerous cable tests, including one done at MIT, that all came up negative.

Quote:

Originally Posted by b0dhi /img/forum/go_quote.gif I agree with much of what you say here. I generally do give credence to blind tests, where they're used correctly. A blind test is also extremely useful as a training tool. I hope my post didn't come across as favouring sighted tests - I consider sighted tests almost useless as "tests". So my criticism is not of blind tests as such, but only how they are mis-applied as support for conclusions they are not designed to elucidate. I agree. The cause, I think, is the fact that most people don't realise that their senses are only representations of the real world that can sometimes be wildly incorrect (LSD anyone?). On balance of (incomplete) knowledge, that person does see compelling evidence for their belief, so I can't really fault them for anything other than ignorance.

Fair enough, and I see the distinction a bit more clearly that you were conveying in your original post (that I missed). I guess the ultimate standstill that the debate reaches is whether or not one thinks a cable test is representative of reality to the degree of precision that our (uncontrolled) sensory experience is representative of reality. That kind of a statement is almost impossible to prove, however (how do we test that, and is that test not representative? It's kind of regressive).

Quote:

Originally Posted by FirebottleRon /img/forum/go_quote.gif I'm tempted to respond to this in depth but these discussions tend to expand and consume all in their path, so I'll just say that I'd like to emphasis the "without sufficient reason" in the quote above

Indeed, and point taken. I'll leave the philosophy for class time

(I also want to say, right now, that I love the sardonic triportsad emoticon).

 

[nick_charles]

Quote:

Originally Posted by FirebottleRon /img/forum/go_quote.gif Post a link or can it.

Would you accept the writings of the late John Dunlavy ?

 

[ServinginEcuador]

Quote:

Originally Posted by Berlioz /img/forum/go_quote.gif Probability wise, the chances of him getting the order correct are quite low, but it's totally possible.

The odds are 1-in-46656 if he used six different cables and was able to ID them correctly in one pass through the lot.

 

[nick_charles]

Quote:

Originally Posted by ServinginEcuador /img/forum/go_quote.gif The odds are 1-in-46656 if he used six different cables and was able to ID them correctly in one pass through the lot.

That depends on whether each cable was always randomly presented or just presented once. Also we know the tests were done in batches of 3

So test 1 probability is 6 x 6 x 6 if any of the six can be presented in any trial but if each cable appears only once then it is 6 x 5 x 4 (having got cable 1 correct , there is a choice of 5 for cable 2 and 4 for cable 3)

but if the subject knew the identity of the three cables under trial the probability is 3 x 3 x 3 (if random) or 3 x 2 if each cable presented only once.

So basically we need much more detail about the protocol to work out the actual probability.

 

[JadeEast]

Quote:

Originally Posted by FirebottleRon /img/forum/go_quote.gif I listen to my system to enjoy music, if fornicating with cable testing is more fun for you than why not just stick with iBuds?

A BS statement as if people who are interested in an objective explanation are some how deficient in the ability to enjoy the aesthetics involved in music or gear appreciation. If music listening was paramount you'd be more than happy with an ibud.

 

[ServinginEcuador]

Quote:

Originally Posted by nick_charles /img/forum/go_quote.gif So test 1 probability is 6 x 6 x 6 if any of the six can be presented in any trial but if each cable appears only once then it is 6 x 5 x 4 (having got cable 1 correct , there is a choice of 5 for cable 2 and 4 for cable 3)

Yeah, without the test methodology we'll never know how to figure this out.

But, with six cables, even if they appear once, the odds of probability are still 1-in-6 six times. The possible number of combinations, a simple factorial calculation, just tells us how many ways there are to organize these six cables. I didn't think that you can use the factorial calculation to give odds of someone identifying one out of six cables?

The probability of a cable being in a particular location is 1-in-720, but the chances of guessing its exact location is 1-in-46656 IIRC.

Boy, this takes me back to some college math from 20 odd years ago, but I'm totally open to being wrong in remembering the use of each calculation.

 

[nick_charles]

Quote:

Originally Posted by ServinginEcuador /img/forum/go_quote.gif But, with six cables, even if they appear once, the odds of probability are still 1-in-6 six times. The possible number of combinations, a simple factorial calculation, just tells us how many ways there are to organize these six cables. I didn't think that you can use the factorial calculation to give odds of someone identifying one out of six cables?

Again without knowing how the test was done you cannot say.

For instance trial 1 P(1) with three cables has to be treated as independent of trial 2 P(2)which used 3 different cables this is known, so you have
(P(1) + P(2))/2 not P(1) x P(2) i.e 1 in (6x6x6) + 1 in (6xx6x6) /2 unless the identity of t1(x.1) t1(x.2) and t1(x.3) are known then it is

(1 in (3x3x3) + 1 in (3x3x3)) /2

Unless the order was predetermined in which case it is
(1 in (3x2x1) + 1 in (3x2x1))/2

In this most extreme case the probability drops to 1 in 6 overall

 

[SmellyGas]

Quote:

Originally Posted by ServinginEcuador /img/forum/go_quote.gif The odds are 1-in-46656 if he used six different cables and was able to ID them correctly in one pass through the lot.

Nope. We don't know his exact methodology, so we cannot run any stats on it. the OP could have easily said "you're right!" after each cable, allowing the listener to use process of elimination to help guess the remaining cables.

Furthermore, I could just as easily type up "my own experiment" here and you would also just have to take my word for it. I can guarantee you that the results will be amazing.

Addendum:
Didn't see post from nick_charles' post above. I'm basically saying the same thing.

 

[SmellyGas]

Quote:

Originally Posted by FirebottleRon /img/forum/go_quote.gif I listen to my system to enjoy music, if fornicating with cable testing is more fun for you than why not just stick with iBuds?

Someone else here already addressed this faulty logic.

Quote:

SmellyGas, can you show us a test preformed by audio engineers, statisticians, researchers, and other people with doctorates that proves that every cable sounds alike? Or is talking about it all you can do? Seems like a common problem lots of people have when it comes to cables.

The problem really is that any blind, controlled listening test, whose results are published in a peer-reviewd journal, and analyzed using basic statistics, is not readily understandable to anyone who isn't involved in similar lines of work or commonly reads these papers. Imagine trying to read a study on cigarette smoking and cancer, and the abstract reads: "in a large, multicenter retrospective study with matched controls, blah blah" are you really going to understand that, and interpret the implications of the study design and the analysis? Is such a study (or a lot of them that say the same thing) going to convince a smoker to kick the habit? Or will the smoker say "oh, science is wrong. i know my Marlboro lights don't cause cancer because my grandpa smoked them. furthermore, the idiot researchers only tested Slim Jim brand, not Marlboro Light."

On the other hand, there was a nice experiment done in Sound & Vision Magazine by Tom Nousaine. "The Great Chicago Cable Caper." I think (or hope, at leaset) that a lot of people have already read this:
http://www.nousaine.com/pdfs/Wired%20Wisdom.pdf
Keep in mind that a magazine makes money from advertisements, and there is actually a financial disincentive for them to publish this negative article.
His methodology and analysis are explained in very simple form, and should be more readily understandable. You can start there if you really want to know the truth. I'm pretty sure you don't, though. Most people don't.

 

[sanderx]

Quote:

Originally Posted by PhilS /img/forum/go_quote.gif ^^ No, this is not a semantical argument at all. Sometimes people need to listen to the substance of what people are saying, instead of picking at nits.

So Phils, what kind of test would it be for you to accept that there is no audible difference between cables?

 

[BobMcN]

I don't think any engineer would argue that different metals conduct electricity differently,

- or that guage plays a role
- or that solid vs stranded construction plays a role
- or that different dielectrics interact to influence the inductance and capacitance and resistance
- or that shielding influences the signal
- or that different geometries affect inductance and capacitance and resistance

So why is it so hard to think that different cables can affect the sound differently, unless of course they are constructed identically?

Sure, an engineer will also say that regardless of the construction, if they measure identically (within audible tolerances per his/her textbook of electrical properties) there should be no difference. To that I answer, what if even you can hear a difference ... will you still believe that there is none because the textbook says there shouldn't be? (to which he will say he was influenced by the color of the cable or some other such thing, no doubt.)

Now a theoretical scientist might postulate that we don't have adequate tools to measure this properly. Sure, we now believe in quantum particles, but we still can't see them. We can only measure them indirectly and draw conclusions based on indirect measurements that are based on theories, whcih may be proven inexact over time. So perhaps the perceived differences in cables need to be understood and measured indirectly, based on an assumption that there is something here that is not clearly understood.

But of course there is no real economic impetus to do so. Maybe if someone could just build a superconducting interconnect for us to try out

Bob

 

[SmellyGas]

Quote:

Originally Posted by BobMcN /img/forum/go_quote.gif I don't think any engineer would argue that different metals conduct electricity differently, - or that guage plays a role - or that solid vs stranded construction plays a role - or that different dielectrics interact to influence the inductance and capacitance and resistance - or that shielding influences the signal - or that different geometries affect inductance and capacitance and resistance So why is it so hard to think that different cables can affect the sound differently, unless of course they are constructed identically?

Great. So if two cables, though minutely different in LCR properties, when played through a loudspeaker, have identical frequency vs. amplitude (as far as you can measure), phase, and distortion, then what physical or imaginery property could cause ACTUAL differences in audio reproduction?

 

[royalcrown]

Bob: I think if blind test revealed differences between cables, nobody would argue that they sound the same even if they measure the same. The evidence that's compelling, at least for me, is that not only are there no measurable differences, but there are no audible differences either.

Sure people hear differences while unblinded, but that's like acing the SAT when you have the answer sheet - your answers aren't coming from your own intelligence or knowledge. You know which component is which when you can see them - but that's because you're not using your ears to identify them so much as you're using your eyes. However, when only the ears are tested, we don't see any differences between cables. Audibility wrt measurements stem from listening tests in the first place: those "audible tolerances" you mention weren't arbitrarily or electrically defined, they are there because lots of testing was done to reach that level - all done using subjects' ears.

I really don't think the measurement aspect of cables is what's controversial - the real question is how those measurements translate to audibility. However, to assess this, we need blind testing, which some people here reject. The root of the problem, as I see it, lies in blind testing.