[philodox]

Quote:

Originally Posted by Patrick82 I wonder if circle sounded round, smooth and warm. Triangle sounded sharp and detailed. Square sounded harsh and jittery with some emphasized detail?

Which shape are you Patrick?

which...shape...are...you???

 

[russdog]

Quote:

Originally Posted by SunByrne And for those of you who are statistically inclined, this produces chi-square(4) = 6.86, p = .14. Cannot reject the null hypothesis of no association. This suggests, but does not prove, that responses were effectively random. We'd need a bigger sample to have more confidence...

The absence of significance never proves that responses were random. The absence of significance only means that you can't assume they were not random.

You can get significant results with small sample sizes, but to do that you need to use non-parametrics. The answering scheme here supports non-parametrics, but chi-square ain't gonna help. The non-parametric stats tend to be obscure, but you can find them.

The problem here is that, even if we got significance, we still wouldn't know what the results mean. It's not as easy to come up with a well designed test as you might think, and this one is just an example of that.

 

[Patrick82]

Quote:

Originally Posted by philodox Which shape are you Patrick? which...shape...are...you???

I am neutral shape, what else?

The Canare has thicker wire so it should sound smoother. Silver should sound detailed. Rat Shack should sound harsh. It fits well with the shapes.

 

[Xanadu777]

Quote:

Originally Posted by anastassios I think you got confused with what is the triangle and what is the square.

The triangle is solid silver, according to the pic. In actuality the solid silver was chosen the least as being Rat Shack by what I consider a substantial margin to two cables I consider fairly equal.

So the question to ask to back this theory up, did the 14 start by picking the least desirable, which I think would be the natural thing to do (and equate it to Rat Shack).

I would be curious to hear from the participants which they tried to decipher first, silver, Rat Shack or star quad (or was it random). If it was anything but Rat Shack first, my theory is nullified. Other than that theory, it seems to be a jumble to me...

 

[jpr703]

Quote:

Originally Posted by Xanadu777 I would also like to see a test where two cables are exactly alike and one totally different, then have the peeps choose which is different...

That sounds like an interesting test to me since it eliminates a lot of the opinions and concentrates on sound differences. If you could conduct a test like this using a standard set of equipment, it would show you something.

 

[eyeteeth]

Quote:

Originally Posted by Patrick82 with Nordost Valhalla I have no problem because it is smoother and has more detail than stock cable, so with me it's either stock or Valhalla.

That's based on sighted or blind listening?

Quote:

Originally Posted by markl Funny, to me, it shows how cable nay-sayers/agnostics will knee-jerk jump on anything that appears to "prove" them right (as seen in this thread, there's been some, let's say, "pre-mature" celebration going on),

The "nay-sayers/agnostics" seem well behaved. Somewhat blase' actually, in a "tell us something we don't know" kind of way. I've always thought 'believers' should do more blind listening for the purpose of self-education. It can be a threatening exercise though as well as a humbling one.

It isn't that blind listening is THE answer...but why or how is it the wrong way to listen?

 

[Sovkiller]

Quote:

Originally Posted by eyeteeth It isn't that blind listening is THE answer...but why or how is it the wrong way to listen?

That reminds me of the last meeting in NYC, before the national one. I was with Ray Samuels talking and discussing about his amps, Ray is a very nice fellow and I like very much to talk with him, his conversation is really nice and instructive, most of the times, for the ones not familiar with him...I was always curious about them, markl always suggested me to try the HR-2 which he absolutelly loves, and following his advice, I did it....just to satisfy my curiosity...

Ray provided a quick blind test among three of his amps, as while listening them by separate I felt the HR-2 a little brighter, than the stealth and the hornet. He called me for that purpose, as soon as he got ready the setup, to prove me wrong, so he hooked there the Hornet, the HR-2, and the Stealth, honestly none of the members in the meeting that participate in that test, about 3 or 4 besides me, were able to discern between the three of his amps, hooked in his Meridian...after we finished I was totally honest with him, and I accepted that to my ears, or my ears were misleading me, or the 3 amps sounded identical, that was precisely what he wants to prove, and what he wants to hear, that the sonic signature of all his amps, is the neutrality, and all of them sound alike, just being really accurate in offering you, just what is feed in them, that were his words, not mine, and nor the ones from the members there, Ray himself stated that, that was regardless of what I may or not think as my personal opinion, which I do not consider relevant now.

Well later on while stated this here, some called me crazy and that this was not the optimal conditions to perform the test and that they hear this and that and differences and blah blah blah,...guys the manufacturer himself accepted that all his amps sounded alike, he set the test, conducted, and concluded it, and they still insisted in saying that the test was not good enough....

If I told that long story was just as a way of stating that people will never accept what they consider that proves them wrong, not even is this is said and stated by God...they will still believe what they want and period!!!!

 

[K2Grey]

I haven't studied this closely, but it seems to me that this does not show that cables make no difference. For that, you would want a test with 2 sets of cables. One set has 2 identical cables, the other set has 2 different cables. And then you would ask people to determine which set is which (i.e. which 2 cables are the same, and which 2 cables are different).

However, this test does show that there is nothing wrong with Rat Shack cables. So while this does not necessarily show that cables do not differ from each other in quality, it does show that if they do, there is no particular relation to price or material. So the cable naysayers are better off.

 

[Sovkiller]

Quote:

Originally Posted by K2Grey However, this test does show that there is nothing wrong with Rat Shack cables. So while this does not necessarily show that cables do not differ from each other in quality, it does show that if they do, there is no particular relation to price or material. So the cable naysayers are better off.

But the curious thing was that the Radio Shack cables were considered among the worst sounding cables, with a muddy and dark sound, and later on during the test they were taken as if they were the solid silver ones, or at least what they consider that a silver one should sound like, that according to what have been said here many times are supposed to sound all the opposite, clean, detailed and a little on the bright side, what an irony, eh?

 

[jbloudg20]

Very interesting, Ed! Nice job.

 

[Iron_Dreamer]

Kudos to Ed for conducting this test, even if he lost my data!

(but then again so did I, it's just so hard to remember which cable sounded like what, two years ago!

)

 

[SunByrne]

Quote:

Originally Posted by russdog The absence of significance never proves that responses were random. The absence of significance only means that you can't assume they were not random.

In a strictly Fisherian sense, no, of course not. But if you have large sample sizes and a minimum meaningful effect size, with power analysis you can make highly-informed statements.

Quote:

You can get significant results with small sample sizes, but to do that you need to use non-parametrics.

You can get significant results with parametrics and small sample sizes, too, if the effects are large enough.

Quote:

The answering scheme here supports non-parametrics, but chi-square ain't gonna help.

You have a point, of course--the observations aren't independent, and the expected cell frequencies are a bit low.

Quote:

The non-parametric stats tend to be obscure, but you can find them.

Depends on which kinds of nonparametrics you mean. Bootstrap-based methods are hardly obscure these days.

Quote:

The problem here is that, even if we got significance, we still wouldn't know what the results mean.

Well, we would in fact know something. For instance, here's a hypothesis which is still tenable in the face of the current data (I'm not saying I agree with this, just that it's still tenable): people can't really tell the difference and are just responding randomly. If we could reject the null on a test of association, we could reject that explanation.

Quote:

It's not as easy to come up with a well designed test as you might think

I couldn't agree more. In fact, I routinely recommend rejection for submissions to the scientific journals for which I review, usually on the basis of flawed experimental methdology/statistics.

I just didn't think this was quite the right context for being quite that fastidious.

 

[Edwood]

Quote:

Originally Posted by Jahn I expected you to bite into a big ass bell pepper and grin afterwards, but i admit to being thrown off when you didn't

LOL, yes.

I figured you or Jason would be the only ones to catch it right away.

-Ed

 

[markl]

Quote:

I'm sorry if it requires the actual turning on of one's brain to do some evaluation of the test's merit, but it does. If that's too complex for some minds to encompass, oh well.

Just wanted to say, that yes, in re-reading this comment, my smirking sarcasm with a wink in my eye doesn't quite translate as intended, does it?

I agree, that statement is utterly unnecessary and should not have been typed by yours truly.

Carry on!

 

[Febs]

Quote:

Originally Posted by Sovkiller guys the manufacturer himself accepted that all his amps sounded alike, he set the test, conducted, and concluded it, and they still insisted in saying that the test was not good enough....

I don't know the extent and nature of the criticism that was made of the test, but it probably is worth noting that the test was only a single-blind test. Ray had an expectation as to what the results of the test should be, which may have biased the results.

Don't get me wrong, I think that it's still an interesting observation, but I wouldn't offer it as scientific proof.