$1,000,000 cable challenge

[musicoverall]

Same can be said for many proponents of overpriced snake-oil cables. Ultimately good DBX testing would be wonderful, however people will misinterpret results to their hearts content. I can easily see how a good test would be interpreted all wrong.

Assuming no difference between cables the chances are 50/50 of getting it right. And depending on the # of tests it's going to take quite a few runs to get a significant result out of this if the difference between cables is not blatantly obvious.

I would love to see average 10 guage pitted against the most expensive and "well reviewed" 10 gauge available and have it tested on a variety of systems (covering all types of speakers and amplification, tube and SS). The actual gauge is less important than the cables being compared are of the same gauge just to eliminate one extra possible bone of contention.

Well, your first comment *should* be testable, but won't be until the audio world can agree on who the "many proponents of overpriced snake oil cables" might be. As of now, who belongs in that group is as subjective as anything could possibly be.

I don't think people "misinterpret" such ABX tests, per se. I think they simply take them for what the tests mean to them. A null result on a cable test is fairly close to meaningless but, to an objectivist, its close to proof in many cases. Whatever makes people feel better. A correct interpretation of a null result is, quite simply, "the testee was not able to distinguish the two devices under test on this day, with this system, with the music chosen, with the system used, and with the testee's particularly state of mind at the time". It doesn't mean the two devices sound alike, no matter how many trials were involved. It's hard to read much into that.

The problem as I see it is that there isn't a universally accepted testing methodology for cables. I'd like to believe in ABX tests but I cannot. If Fremer passes his Randi challenge (as if the test will ever happen!), that would help.

[Feanor]

... A correct interpretation of a null result is, quite simply, "the testee was not able to distinguish the two devices under test on this day, with this system, with the music chosen, with the system used, and with the testee's particularly state of mind at the time". It doesn't mean the two devices sound alike, no matter how many trials were involved. It's hard to read much into that.

....

In essence a DBT/ABX cannot prove that the tested components sound identical.

[emorphien]

In essence a DBT/ABX cannot prove that the tested components sound identical.

If the success of detecting one cable over another was close to the rate for random chance, then the test indicates there is no audible difference between the two cables.

It's very easy for someone to willingly misrepresent or misinterpret data even from this kind of test, although not as easy as other testing methodologies. And people will believe it, even if the conclusion is completely made up.

[musicoverall]

If the success of detecting one cable over another was close to the rate for random chance, then the test indicates there is no audible difference between the two cables.
.

Respectfully... no, it doesn't. It indicates that the person under test could not determine differences at that particular time. Another time, he might. Another person might. Any inferences or decisions made on one particular test (regardless of the number of trials) MUST relate only to the testee and not the devices under test. Most of the DBT's I've read about are woefully devoid of details regarding the test methodology. Reading anything into these tests is done at the reader's own risk.

[emorphien]

Respectfully... no, it doesn't. It indicates that the person under test could not determine differences at that particular time. Another time, he might. Another person might. Any inferences or decisions made on one particular test (regardless of the number of trials) MUST relate only to the testee and not the devices under test. Most of the DBT's I've read about are woefully devoid of details regarding the test methodology. Reading anything into these tests is done at the reader's own risk.

That's exactly what it says, but yes it is for that person and the equipment used in the test. For this reason, a test like that needs multiple test subjects and as many combinations of speakers and amplification as possible in order to strengthen the clarity of the results. In reality the number of speaker-amp combinations really necessary to pretty much settle the issue likely isn't that high. Sure you'd want to cover solid state and a couple tube amp designs, as well as the varying planar speaker designs, etc. But I think a pretty good results could be had with 5-8 setups and a handful of listeners (experienced and not) through a series of trials.

Many DBTs I've read about in the audio world are devoid of details because the results are not what the tester wanted. This is true in any field where DBTs might be used, warping the results or hiding details is a skill in and of itself, even if not the most virtuous of skills.

The problem with a DBT like this is there are many people both selling and buying these products who don't want to be proven wrong and even if the testing methodologies and results speak for themselves, these people may never believe them.

[musicoverall]

That's exactly what it says, but yes it is for that person and the equipment used in the test. For this reason, a test like that needs multiple test subjects and as many combinations of speakers and amplification as possible in order to strengthen the clarity of the results. In reality the number of speaker-amp combinations really necessary to pretty much settle the issue likely isn't that high. Sure you'd want to cover solid state and a couple tube amp designs, as well as the varying planar speaker designs, etc. But I think a pretty good results could be had with 5-8 setups and a handful of listeners (experienced and not) through a series of trials.

Many DBTs I've read about in the audio world are devoid of details because the results are not what the tester wanted. This is true in any field where DBTs might be used, warping the results or hiding details is a skill in and of itself, even if not the most virtuous of skills.

The problem with a DBT like this is there are many people both selling and buying these products who don't want to be proven wrong and even if the testing methodologies and results speak for themselves, these people may never believe them.

I'm not so quick to state unequivocally that Testee A scored only 10 of 20 on Tuesday and is therefore labeled as someone who cannot hear differences in the DUTs. I think that's a conclusional leap of faith, and I'd prefer to see how he does over time - at least 2 more tests on 2 other days. I've had several days myself where my state of mind was not conducive to hear much in the way of subtleties. For this and other reasons (mostly due to lack of detail), people tend not to believe DBT results.

I also don't agree that non-experienced listeners should be used, unless the goal is to show how listeners of different experience levels score on DBT's. Most inexperienced listeners I know do fine on the overall sonic picture but can't focus on certain "audiophile" things like imaging and soundstaging, etc. I'm not saying that's a bad thing; hell, a lot of inexperienced listeners I know enjoy music more than a lot of audiophiles I know and can get to the nitty gritty of the music, if not all the subtle aspects of the sound.

I guess our disagreement is at the heart of a lot of people's problem with DBT's as a testing mechanism. Opinions on how and who to test, and what to make of the results, differ. I think most people wouldn't mind too much being proven wrong but rarely do the tests "speak for themselves", as you posted. They might if they provided more detail but the ones I've seen are lacking and, as such, are easy to disbelieve.

[emorphien]

I'm not so quick to state unequivocally that Testee A scored only 10 of 20 on Tuesday and is therefore labeled as someone who cannot hear differences in the DUTs. I think that's a conclusional leap of faith, and I'd prefer to see how he does over time - at least 2 more tests on 2 other days. I've had several days myself where my state of mind was not conducive to hear much in the way of subtleties. For this and other reasons (mostly due to lack of detail), people tend not to believe DBT results.

I also don't agree that non-experienced listeners should be used, unless the goal is to show how listeners of different experience levels score on DBT's. Most inexperienced listeners I know do fine on the overall sonic picture but can't focus on certain "audiophile" things like imaging and soundstaging, etc. I'm not saying that's a bad thing; hell, a lot of inexperienced listeners I know enjoy music more than a lot of audiophiles I know and can get to the nitty gritty of the music, if not all the subtle aspects of the sound.

I guess our disagreement is at the heart of a lot of people's problem with DBT's as a testing mechanism. Opinions on how and who to test, and what to make of the results, differ. I think most people wouldn't mind too much being proven wrong but rarely do the tests "speak for themselves", as you posted. They might if they provided more detail but the ones I've seen are lacking and, as such, are easy to disbelieve.

If testee A only scored 10 out of 20 on one day, that means on that day he could not discern a difference. If on days two and three the scores were similar, I'd say you've got some pretty conclusive evidence. If I were setting it up, I'd have people test over two or three days because there are plenty of people who share your opinion (and i have days where my listening just doesn't seem right as well). The cable mania has gotten to the point that good sound practice isn't enough, overkill would be necessary to convince anyone of anything. I'll get back to your comment about why people don't believe DBTs later though.

I think non-experienced listeners should be used, but not exclusively. It will be interesting to compare their results against the so-called "golden ears."

The problem is many people don't understand the mechanics of DBTs, both those performing them and analyzing the results, and people out in the world reading about the results. Whether different systems will really benefit from anything more than changing gauge is questionable, but the need to test on various types of systems is real because people believe it depends on the system. Any existing belief shared by enough people needs to be tested to see if it holds water.

The biggest problem with DBTs as a testing method, and any other human trials is that it is easy for people to discount, however the interpretation of the test results is clear. Whether or not it holds a global truth depends on how the test was performed. A success rate of half absolutely means no statistically significant difference, but if the listeners ears were clogged or if he was tired, then that only says that given those conditions he couldn't tell. Hence the need for both multiple listeners and a few days in order to perform the trials.

After a while though, enough experimental care has been taken and there will still be naysayers who will never believe it. You yourself stated something that hints at your own opinion on the matter:

I'm not so quick to state unequivocally that Testee A scored only 10 of 20 on Tuesday and is therefore labeled as someone who cannot hear differences in the DUTs.

You are suggesting that if the difference can't be heard, that it is the listener's fault. However it is just as likely that it means there is actually no audible difference that a human can detect. However to be able to confirm one or the other multiple subjects are needed.

[musicoverall]

In essence a DBT/ABX cannot prove that the tested components sound identical.

Since DBT talk is so boring, I'm going to change gears a bit and say that your system is about the nicest "down to earth" audio system I've seen among the posters here. I'd be willing to bet you get a tremendous amount of enjoyment out of it, and don't waste much time worrying about little ancillaries like cables.

I don't see many of the Monarchy monoblocks around - how do you like them? Interestingly, the last pair I saw was driving Maggie 1.6's. Killer combo. Anyway, congrats on what appears to be an extremely well thought-out system. And it's mostly "down to earth", although I think that preamp has tubes in it! LOL!

[Feanor]

Since DBT talk is so boring, I'm going to change gears a bit and say that your system is about the nicest "down to earth" audio system I've seen among the posters here. I'd be willing to bet you get a tremendous amount of enjoyment out of it, and don't waste much time worrying about little ancillaries like cables.

I don't see many of the Monarchy monoblocks around - how do you like them? Interestingly, the last pair I saw was driving Maggie 1.6's. Killer combo. Anyway, congrats on what appears to be an extremely well thought-out system. And it's mostly "down to earth", although I think that preamp has tubes in it! LOL!

My system is necessarily down-to-earth given my limited funds. For a lower-mid-range system, you will note the balance amongst equipment fit the profile speakers-first profile pretty well. 3 of 4 reviewers at The Absolute Sound choose the MG 1.6QRs as their choice under $2k. The Monarchy's are outstanding value at under $1k, (a price you can or used to be able to get from the factory on, let's say, "refurbished" units). The Sonic Frontiers LINE 1 MSRP was $2.5k, but I got mine 2nd hand for $900. I wasted the most money on the modified Assemblage DAC in whose sound improvement over my Sony SACD or my M-Audio external sound card is very, very slight to say the most.

Since you ask about the Monarchys, I think they're really nice and I'm totally satisfied, though if I played at higher volumes I'd probably like more power. They are a high-bias, low-feed back design and are, IMO, detailed, dymamic, and very musical (to borrow that over-used cliche).

No, I haven't lavished a lot of money on cables. Many of my interconnects are Blue Jeans which I highly recommend to sensible people. My speaker cables are actually 4-wire, 14 ga. Monster used in biwire configuration; theyt cost me <$60, and are absolutely indistinguishable (by me) from the $250 Nordost Flatline Gold MkII I used previously.

But, heck yes, I have one indulgence. I have rolled several sets of tubes through the LINE 1. Yes, I believe the ones I use now, (Amperex USA white label PQ's in the gain slots), sound significanty better than the stock Sovteks. But could I distinguish the PQs from the Sovteks in a DBT? Let's just say I wouldn't bet a $1M on it.

Here's a link to my configuration ... http://ca.geocities.com/w_d_bailey/StereoDiagram.jpg

[musicoverall]

The 1.6's are my choice as well. As a longtime owner of the 20.1's until a year or so ago, I think Maggie's are the way to go. I sold them, not because I found something better, but because I needed the money to finance a very large estate purchase of several thousand LP's and 78's. I'm making do with a much lesser system and, you know what? I kept telling myself I'd have an expensive system again by now but I'm enjoying the music so much that I haven't upgraded one piece of the hardware. To tell you the truth, when I do rebuild, my system will be a lot more like yours than it will be like what I used to own.

Since you mention CDP's, I'd have to say they are a lot like cables to me. Is there a difference? Yes. Would I bet my life on it? No. My favorite CDP is the one that navigates scratched CD's the best and my old Sony XA20-ES sails through damaged discs. The unit is 12 years old but was little used for half that time. It works wonderfully and sounds just fine.