Observations of a controlled Cable Test - Page 2

He sure saved me a lot of typing.

But here's another discussion (two posts) of the interaction of expected size of effect, p value, and number of trials (with a lot more typing...by someone else). It's also a good precis on good audio DBT practices:


"What is a Blind ABX Test?"
http://www.hydrogenaudio.org/forums/...howtopic=16295

and here's a link to a useful excel table of binomial probabilities of guessing

http://www.kikeg.arrakis.es/winabx/bino_dist.zip

basically from this you can read the p value that the result's due to chance, for any combination of #right/#trials up to 100/100.

For 8 trials, the p-values are shown below (from 1 to 8 correct). Mike would have to get 7/8 to be in the ballpark of standard scientific acceptability (p<0.05)...assuming a 95% confidence interval is appropriate for this hypothesis.

1.000
0.996
0.965
0.855
0.637
0.363
0.145
0.035
0.004

Well, Mike can now auction off his specially prepared Monster cables - the ones that sound so close to the Transparents, you can't tell them apart.

Yes, must be worth 30K at least...

Damn, but at those prices I'd want to hear a difference not only blind but blind drunk too.

It just sunk in that for the price of a pair of 25' Transparent Opus cables (43k according to gizmodo) you can buy (in the US) a 5.2 system made up of 5 x B&W 802D and a couple of JLAudio f113 and still have change. Holy cow. No, really.

Yes, exactly. So the common notion that the box may or will cause an audible difference is not founded in anything and the higher probability is that the box is for RF filtering and not audio band alterations.
As to their web site claiming the difference, that is par for the course no matter what they place on the wire, or just having plain wires. But, it could very well be that is where the common notion came from, jumping to an unfounded conclusion.

As posted elsewhere, someone placed this cable or somthing similar on a spectrum analizer and didn't see the box doing anything in the audio band. I just don't have a link to that message

Yes, but that doesn't stop hearing differences under sighted conditions with 1 minute or more of setup time
But you are correct, acoustic memory has been shown to be very short, on the order of seconds. So, even long segments of music is detrimental.

That never stopped a company like this to put one in anyhow How else but a new gimmick to convince the gullible ones?

i want to thank my friends Joe and Ted for all their help yesterday; they both had sore fingers, knees and backs after all the cable switching. i particularly want to thank Chris; first of all for taking the time to help, for contributing his ideas to come up with the protocol, for being open-minded, but especially for his detailed, complete and accurate description of what happened. i type real slow and would have really struggeled to answer all the 'what happened?' questions.....Chris's write up and followup posts have saved me a good deal of trouble.

sorry for the delay in posting my perceptions.....i must admit to at least a bit of reluctance in discussing what happened......as the result was clearly much different than i expected. i do not feel bad about doing the test or the results (other than a slightly bruised ego) since i do feel it is the truth......at least the truth as to the protocol we used.

yes; i have, to some degree, changed my perspective on cable differences....but...my mind is still processing the results and what they mean for me. i hope that i can coherently relate the various thoughts that go thru my mind.

as Chris mentioned; the controls were successful at keeping me from knowing which cable was which. for each test i felt confident about my choice (except #6...see below).

my methodology was to use a demo disc with multiple cuts and switch quickly from one cut to another; listening to a cut from a few seconds to maybe 30 seconds.....coming back to the same cut a few times. i mostly used 5 or 6 cuts which (i felt) told me about various sonic issues. until the test was over i felt confident that i could hear real differences and that i had a 'lock' on which characteristic was which cable.

the tests were as follows;

1-Opus first.....i chose Opus first.
2-Monster first....i chose Opus first
3-Monster first....i chose Opus first.
4-Opus first....i chose Monster first
5-Opus first....i chose Opus first

after the first 3 tests i was getting a bit tired and the ear protection started causing a slight headache. at no time did i feel that was affecting my listening. after 5 tests i requested a break, and we went upstairs to play a bit of pool.

as we started the 6th test i could tell i was struggeling to hear differences. so i asked that i change CD's and re-do a 'sighted' test. i did not take my covered glasses off but they did tell me which cable was playing.

6-thrown out

then we did two more tests.

7-Opus first....i chose Opus first
8-Opus first....i chose Monster first.

when i made my choice known for #8 i was confident that i was 100% for all 7. then my friend Ted said 'that's it.....test over'. we had discussed prior that any result 7 out of 10 or better or 15 out of 20 or better would mean a positive result and to continue. once we got to only 3 out of 7 it was clear that we were not going to get a positive result.

OK......so how do i feel different?

about 1000 posts ago in the Randi thread Michael Grant asked me about how i would feel if i failed a blind test with my Opus cables.......would my opinion change?

my recollection is that i said that it would cause me to further investigate the issues with a much more skeptical viewpoint......but that until i did that additional investigation i would not dramatically alter my viewpoint. i think that comes closest to describing my feelings right now. i have some work to do. there is a chink in my subjective armor but it's not completely broken.

why did i fail?.....or put another way.....why did this test show no real difference?

was i overconfident?

yes; regardless of the eventual answer i was not respectful enough of the challenge.

looking at the specific test results i got both #1 and #7 correct......directly after the sighted trial where my aural memeory was the freshist. what if i were to have 'A' , then 'B', then 'X' every time? would that change the outcome?

did i spend enough time listening to the Monster prior to the test? i was very confident going in but now i think i had not taken enough time to really get into the Monster more.

i did not practice enough blinded.....for sure. if i had practiced i would know that my methodolgy would not show any difference. it's also possible that no amount of practice would be able to prove i could hear the differences blinded.

in my mind i am not confident that i will ever be able to hear reliable differences between the Monster and the Opus to pass a Blind test. OTOH i am also not sure i won't be able to do it.

and that is where i'm at right now.

i have lots more thoughts.....but as i said i type slow and i wanted to get something out.....i will do my best to answer any questions and add other thoughts.

and thanks everyone for all the positive comments and keeping to the high road.

the price of Monster cable just went up to $500.00 a foot.

It should be interesting to hear from MikeL regarding how he perceived the sound quality from his system during the days following the test. Will it sound as good as percieved from the sighted Opus or worse as noted with the Monster cable. Was alleged placebo adding to or subtracting from the experience.

the system was sounding great last night after the test and this morning again.....in fact, as good as or better than ever.

i have not re-inserted the Monster since the test.

i need to just enjoy for awhile before i climb back on the 'cable-wars horse'. this is all a fun thing for me......not work. i need to re-charge my investigative juices.

Yep if you read the BS on the website they claim these boxes do have an effect on the sound versus those without boxes. In fact they say it so revelaing it needed special tweaking. So they claim it effects what you hear then it indeed can be measeured. So they lose. The box is hocum.

Having an effect on the sound, and having an effect on the sound that is audible are two different things.

So, actually, it was only *7* trials , with 3 correct (that's a p=0.773, i.e., a 33% probability that the correct answers were not due to chance -- versus a typical target value of 95%)

The most interesting part of this turns out not the be the stats, which are telling but meager, but the self-report from Mike that he was sure he'd got 7/7 correct. This is really classic, and very common, overconfidence in the accuracy of subjective perception. It's much the same confidence he's expressed all along, regarding the Transparents. I hope Mike now has a deeper appreciation of the roots and bases of audio skepticism regarding sighted reports.

Mike,

You have my utmost respect and admiration for being willing to do this test, and for the way you've conducted yourself following the posting of the results. You really had nothing to win and a lot to lose, and certainly get my vote for the biggest balls on AVS. Chin up and enjoy your wonderful system !!!

It is ALL in the mastering.

Hi

Just wanted to congratulate MikeL and Chris Wiggles... Just a great experience, one that has me thinking very hard... I do not take the results as conclusive... I have always been of the opinion that cables differences are not as large and easily discernible as we, audiophiles would like to think... Subtle? Yes... Non Existent? I continue to doubt...

Mike needs downtime to come up with a reason why cables still sound different..

As for the networking boxes, I can offer this: I believe that the vast majority of them are RC or LC or RL networks to compensate for abnormally high capacitance that these esoretic cables posess. The network boxes are there so that some amps won't go into oscillation due to driving a highly capacitive load.

So while the network itself won't affect the audio spectrum as the values used are active at frequencies far above the audible spectrum, I suppose that the manufacturer could claim that a non oscillating amp sounds better than one that is oscillating wildly at 100Khz. Never mind the fact that generally speaking, a speaker sounds better with the tweeter working rather than it being blown by said ultrasonic amp oscillations..

Yes, but the capacitances are nothing like Alpha Goertz and besides, they provide the resistor and capacitor when you buy their wire.

Mike, thanks for your thoughts. I definitely hope that you'll consider continuing this crusade. Go after those details of the test that didn't suit you. Everything that you think you can do to improve your chances of passing a blind test---from more practice, to more comfortable testing conditions, to shorter changing intervals---do them. Of course, the blindness has to stay in; but within that constraint you should do whatever you feel can help.

Congrats to Mike for being so brave with this.

Mike, look at the bright side: You'll be saving a TON of money in the future by not buying boutique cables anymore.
Instead, you'll be redistributing that hefty amount into your already wonderful system.

Enjoy.
PS Like Curt already said, don't try to find a reason why you failed the test. Just accept it because it is the truth...

The interesting part of this and what i am most interested hearing mike continue to comment on... is the fact that this has challenged (to some extent) his belief system.

So while we can by no means confirm that this test was conclusive, it can conclusively be said to put to rest the comments of "night and day, system sounded broken before..." etc. hyperbole.

I think that we can comfortably say now on all sides that differences are VERY subtle at best. Anybody who looks at this with a fair and balanced view (not fox news of course) would not say that exotic cables have been forever dis proven, however they have been perhaps brought much closer to reality.

What i am surprised to see is the lack of folks trying to nullify the test.

From Mike Lavigne, maybe. From other 'audiophiles': as if.




It's only been a day. I've alerted the true believers on Audio Asylum. Give it a few hours.

Hi All, especially Mike L.,

For anyone who hasn't been as subject in, and possibly watched from side a true, double blind test, I strongly suggest you be somewhat respectful of those who have. The way the test is conducted and many seemingly minor factors can make such things VERY difficult. I have sat through blind amplifier comparisions as well as one of Tom Nousaine's cable tests as well as an interesting comparison of a cheap/common op-amp circuit. A few of these used the ABX box that QSC had built. To say that a blind test is a difficult listening condition is a gross understatement.

Harman Motive, even more than the home audio end, has done an amazing amount of testing on trained and untrained listeners. In OEM autosound systems, prooving the money you spent was worth it is even more important as compared to home audio where spending more *must* mean it's better. Some guys I went to college with have been involved in plenty of this sort of testing, and to say much of it is sobering is a gross understatement.

Practice and gaining some level of comfort with the test is an important factor to consider, both in general terms and for the specific case. Please note I am NOT defending the $25k cables on the market (some of them really muck with the signal! ), but rather intending to add some levity to the discussion. Many objectivists are also guilty of assuming that the speakers many enthusiasts love follow common behavior. Some speakers have wildly varying impedances, some amplifiers actually try and deliver on the DC-Ch5 ideal, and some components and cables have rather peculiar grounding practices. I'm not suggesting that these are competently designed products, but such occurances are more common in exotic hi-fi than many like to acknowledge.

I see the other thread is locked.

Thank you Mike L for stepping up and taking the time to do this to answer all the questions.

This all just shows how incredible and complex the human mind is when it comes to combined senses.

The testing can be a huge strain and very stressful, I remember my roomates laughing at me all the time!

So is Mike L keeping the Monster cables and selling the other ones?

Why would he do that? Were I in Mike's shoes I wouldn't be so quick to switch camps. No doubt this test was an eye-opener for Mike. But had the test come out differently I dare say most of us would be demanding more evidence. (For example we'd be asking for frequency response measurements on the cables.) I think it's only fair that he do the same, and I hope he continues the search for it.

I'd like to add my thanks to Mike Lavigne for his efforts and for his willingness to subject himself to this rather public test. Also to Chris Wiggles for helping to conduct the test and for his most informative posts.
I recently did a simple test of my ability to hear the differences between audio formats, and having done that, I have some appreciation for the effort that Mike went through.

Sorry, it was just a joke.

Of course he wont sell them because they do make a difference to him in a normal environment. I did post before that I believed he does hear a difference, I believe I would hear a difference, our damn brains play tricks on us all the time

If the testing was done right (it was) and he did hear a difference then I would not demand more evidence from MikeL at all. I would believe his speaker cables do sound different.

Again, I think its awesome he stuck his neck out and did this. I think all audiophiles should do blind tests to understand how we dont really have control over our senses.

My above post out of the way, and having seen some of the similar testing, I do hope this testing leaves many more enthusiasts seriously questioning the claims of the marketeers. I make no assertion as to conclusions from the test other than the differences not being as grossly blatant as many would like to believe. There are many greater offenses in audio systems to pursue with the vigor I see given to cabling.

Personally I lost most interest in the cable controversies long ago when I started dabbling in loudspeaker design. There are so many other minor changes to a design or room that can be both clearly measurable and audible, that you quickly get to the point of "why bother?" My personal suspicion is that the problem boils down to the fact that classic hi-fi systems provide limited means of tayloring, fitting or flavoring for your space and preferences, that enthusiasts are left grasping for anything that may allow them to fine tune the character of their system.

Of course it's just a theory.

That would only be a request to check likely reasons for an 'unlikely' result -- standard science practice. (I was asking for that even before the test!)

Mike's result wasn't unlikely or mysterious, assuming the Transparents really don't measure very different from the Monsters in the audible band. But if he wants to show that something was wrong with the DBT, his best bet right now would be to ....ask for frequency response measurement on the cables! And hope that they really do measure quite differently in the audible band, in his system. Then his FAILURE becomes unlikely and mysterious.

Isnt' it great how that works?



I hope he knows when to stop.

This should be posted in capital letters above any 'cable' forum.