an observation

[883dave]

Quote:

Originally Posted by royalcrown /img/forum/go_quote.gif The second link is dubious because one of the several blogs that linked to it mistyped? If you have a problem with the actual methodology as outlined in the test itself, by all means bring it up, but the original poster is not responsible for blogs that misreport. As for the following three, just because you can't access the articles doesn't mean they don't exist or are thereby discredited. One of the links was done at MIT by a professor there, and involved a sample size of 24. Another was a literature review done at UCSD and discusses the carver challenge. The author is actually a believer and I don't agree with 95% of what he writes about, but the section on blind testing, specifically the carver challenge, is informative. A lot of it is warrantless, but I linked to it to provide some balance and because what is substantial is interesting. The final link is the opposite - laden with blind test analysis, but written with a more skeptical slant. It is also a literature review from Brown University.

No the second link about the monster cable is dubious...read what it says.

They had Martin Logan SL-3 speakers and hooked "them" up with 2 meter speaker cables. Acording to the artice they used "Monster Ultra Series THX 1000 Audio Interconnects" not "Monster Ultra™ Series THX® 1000 Speaker Cable"


Then they took 4 coat hangers and made a pair of 2 meter speaker wires. Every speaker wire I have seen has a positive and a negative running to each speaker "unless they are run in series". How do you get four 2 meter wires (8 meters total) out of at best 4 meters of coat hangers? Did they only hook up one speaker with coat hangers? When they twisted them into a pair of speaker wires what did they use as an insulator?

To quote this as a DBT test is flawed in every way, except if you acknowledge two of the original five (not the morphed twelve) actually were able to tell monster cable from Beldin wire not once or twice but 7 times in a row. (cynical)

In one of the three links I can not access, is "the Carver challenge" in reference to Bob Carver and amplifiers? Are these links actually DBT on cables? Could you share some info ie. brand and type (interconnect, speaker, power) of wire used. And a brief synopsis of the results.
Thanks

 

[JadeEast]

Quote:

Originally Posted by 883dave /img/forum/go_quote.gif Then they took 4 coat hangers and made a pair of 2 meter speaker wires. Every speaker wire I have seen has a positive and a negative running to each speaker "unless they are run in series". How do you get four 2 meter wires (8 meters total) out of at best 4 meters of coat hangers?

I had the same issue when I first read that report.

 

[royalcrown]

Quote:

Originally Posted by 883dave /img/forum/go_quote.gif No the second link about the monster cable is dubious...read what it says. They had Martin Logan SL-3 speakers and hooked "them" up with 2 meter speaker cables. Acording to the artice they used "Monster Ultra Series THX 1000 Audio Interconnects" not "Monster Ultra™ Series THX® 1000 Speaker Cable"

You're telling me to read the article when it's clear you either didn't read it yourself or just skimmed over it. The article specifically mentions "Monster 1000 speaker cable." The author does mention "Inter connects are Monster M-1000's wires" but that doesn't mean the speaker cables are interconnects, it means that he used 1000 speaker cables and 1000 interconnects, with the interconnects remaining constant throughout the test and the speaker cables switched out.


Quote:

Originally Posted by 883dave /img/forum/go_quote.gif Then they took 4 coat hangers and made a pair of 2 meter speaker wires. Every speaker wire I have seen has a positive and a negative running to each speaker "unless they are run in series". How do you get four 2 meter wires (8 meters total) out of at best 4 meters of coat hangers? Did they only hook up one speaker with coat hangers? When they twisted them into a pair of speaker wires what did they use as an insulator?

Just for fun, I went into my closet and, to the best of my abilities, straightened out a coat hanger laying around. It came at just around 2 meters when straightened. I might have weird coat hangers, but these were just the crappy free ones you get when you dry clean clothes. Also, coat hangers are insulated by some sort of enamel - that's why they're white (or at least mine were). You can take a pocket knife and scrape away at it to remove the metal underneath - if you do that at the edges they can be soldered onto the necessary plugs. Also, I doubt they twisted the length, seeing as monster cable (and most speaker cable for that matter) consists of two straight parallel lengths.

Quote:

Originally Posted by 883dave /img/forum/go_quote.gif To quote this as a DBT test is flawed in every way, except if you acknowledge two of the original five (not the morphed twelve) actually were able to tell monster cable from Beldin wire not once or twice but 7 times in a row. (cynical)

The two people guessing correctly is probably a good reason to have higher trial sizes - I'd say 10 is the bare minimum, but others have mentioned even higher numbers. That said, however, I would like to hear what is substantively poorly done with the test that would make it inadmissible (though it's obvious you're going to have an impossibly high standard for these tests). Sure the coat hangers might not have been 4 meters, but how does that invalidate the results? The blinding was still done properly.

Perhaps more importantly, if people cannot even distinguish between monster cables and what are possibly the worst cables ever designed, that's pretty compelling evidence that there's even less of a difference between a well-made stock radioshack or equivalent cable and an exotic expensive one.

Besides, as much as you pick on the coat hanger test, there's still the test done at MIT, published in an academic, peer reviewed journal, that involved 24 people, and none could tell any difference.
Quote:

Originally Posted by 883dave /img/forum/go_quote.gif In one of the three links I can not access, is "the Carver challenge" in reference to Bob Carver and amplifiers? Are these links actually DBT on cables? Could you share some info ie. brand and type (interconnect, speaker, power) of wire used. And a brief synopsis of the results. Thanks

You keep cutting out my response and then acting as if I never responded in the first place. I'll say it again - if people can't distinguish between two amplifiers, it's not reasonable to assert that people can distinguish between cables, when it's uncontroversially true that amplifiers change the signal more than cables.

Also, if you go to your local library they'll likely allow journal access.

 

[b0dhi]

Quote:

Originally Posted by nick_charles /img/forum/go_quote.gif LOL I have a DAC/amp with an optical digital input so no need for an analog cable at all for the 2nd part. Actually you are known to have good ears you could be a listener for the blind tests ?

Sounds interesting. What's the protocol for the test?

 

[nick_charles]

Quote:

Originally Posted by b0dhi /img/forum/go_quote.gif Sounds interesting. What's the protocol for the test?

What I do this is this.

I connect an analog source to an ADC with the cables under scrutinty. I play several different samples, white noise, cymbals, and digital silence on source and make 10 recordings of each signal for each cable. For each batch of 10 I very carefully trim and align the samples to +/- 1/100,000 of a second. I run each of the 10 samples through a spectrum analyser (Audacity) using a 2048 fft and import each data file into Excel. Each set of 10 for each signal for each cable is agregated and stats Ave, min, min , max difference are calculated for each frequency - this smooths out random variation.

This is repeated with each cable. Then I load the means for both cables into a 2nd sheet and chart the FR of both and calculate differences between the two.

For listening I choose the samples of each cable that are nearest to the artithmetic mean for that cable, i.e most typical across the spectrum (6hz - 20Khz) and do FooBar blind listening tests. So far having tested 7 dfferent cables (budget, pricey, solid copper, stranded copper, stranded silver, silver plated copper, shielded unshielded, directional, non-directional) while I have found measurable (v. small) differences I have not found audible differences, I have made my samples available on my cable test thread elsehwere for others to try out.