More Audioquest cable: this time the blind testing

[ThirdDrawerDown]

Following on from @@PCOWandre'd thread about the AudioQuest Vodka Ethernet cable teardown, which led to it being dismissed by the SNA folks hereabouts.

 

Ars Technica did some blind testing.

 

There's much in the comments about silly audiophiles, along with some interesting points about test regimes.  Good to see James Randi onto this stuff.  There's some reference to a "true believer" audiophile community, an epithet which obviously has some local currency in the USA.  If the community has outposts south of the Equator then I haven't received the invitation or the funny-shaped key.

 

 

We spent considerable time prior to the test discussing whether the listening subjects should in fact be conference attendees or whether they should be verified "golden ears" listeners—listeners with some demonstrated amount of sonic acumen who might be more attuned to detecting tiny variations in sound.

 

We decided to go with regular conference attendees on the strength of the claims—again, coming back to the flowery language in the product reviews indicating that the differences should be apparent to anyone who listened. If the product were marketed and reviewed differently, perhaps a different test would be applicable—but it isn’t.

 

I've met a number of people through Stereonet who fall into the 'golden ears' category.  They've educated me to what to listen to and for, and can put into words things that to me are just a feeling. Even so, my wife puts it nicely: "it's like a blanket has lifted off the music".  

 

Anyway, this test would have been more interesting with golden ears listeners. It's like relying on the average person walking into a bottle store for whisky or wine reviews, rather than a distiller or wine-maker.

 

Another poster points out:

 

t's also entirely possible the cable did make a difference, but since there's an $0.10 DAC and headphone amp being used in the middle, any difference can't be heard.

I don't think it does, but I also don't think anyone is going to remotely buy $340 cables to use with a laptop headphone jack out.

 

 

 

[Addicted to music]

Conclusion: 6 out of 7 tested can’t tell the difference.....

[~Spyne~]

Let's add just a hint of context here:

6 out of 7 skeptics at a skeptics convention report they can't tell the difference.

 

I'm not disputing the result one way or the other, but to perform this test at a skeptics convention, for me, detracts somewhat from the validity of the results.

Just like if the test was performed at an AudioQuest fan convention where people who attended were generally of the mindset that cables did make a difference, and the results came back that 6 out of 7 people reported they could hear a difference (one way or the other), we would all scoff at the test and say the subjects were biased from the outset, the results are invalid.

Two sides of the exact same coin, if you ask me.

 

Run the test on a truly random group of people, from various backgrounds and more importantly, use a higher number os test subjects.

[Addicted to music]

Let's add just a hint of context here:

6 out of 7 skeptics at a skeptics convention report they can't tell the difference.

I'm not disputing the result one way or the other, but to perform this test at a skeptics convention, for me, detracts somewhat from the validity of the results.

Just like if the test was performed at an AudioQuest fan convention where people who attended were generally of the mindset that cables did make a difference, and the results came back that 6 out of 7 people reported they could hear a difference (one way or the other), we would all scoff at the test and say the subjects were biased from the outset, the results are invalid.

Two sides of the exact same coin, if you ask me.

Run the test on a truly random group of people, from various backgrounds and more importantly, use a higher number os test subjects.

+1 agreed

But, I'm someone who uses 6.00mm electrical wiring as speaker cable, well I got it for free!

Edited August 1, 2015 by Addicted to music

[davewantsmoore]

 

 

However, the audiophile contention is that some amount of electromagnetic interference or noise is transmitted up unshielded Ethernet cables, through the Ethernet port, and into the computer’s DAC (the digital-to-analog converter)

 

So why didn't they test that? .... u know, like with an oscilloscope.   <facepalm>

[firedog]

Conclusion: 6 out of 7 tested can’t tell the difference.....

 

Did you actually read the article? 

we conclusively proved that those six people, when listening to a short audio sample in a semi-isolated environment with this specific equipment setup, could not discern any difference between the cables, putting paid to the claim that there should be a "plain as day" difference. On the other hand, it’s unclear that we’ve proven anything more than that

 

. These "scientific" people couldn't even conduct a test that meets the most basic scientific standards. So they showed that 6 people failed to discern a difference - in their test. But the data has no application to any more general conclusion than that. You can believe what you want. But this "test" doesn't back up any general conclusion with any scientific authority. On the level of general applicability, the "test" merely added 6 anecdotal data points of no general value. 

 

Beyond the problems revealed in the article, the so called test was a failure already in conception. I'd say it was a demo or a show, but certainly not a scientific test. And that's before you get to the other problems with the implementation.

[firedog]

Following on from @@PCOWandre'd thread about the AudioQuest Vodka Ethernet cable teardown, which led to it being dismissed by the SNA folks hereabouts.

 

Ars Technica did some blind testing.

 

There's much in the comments about silly audiophiles, along with some interesting points about test regimes.  Good to see James Randi onto this stuff.  There's some reference to a "true believer" audiophile community, an epithet which obviously has some local currency in the USA.  If the community has outposts south of the Equator then I haven't received the invitation or the funny-shaped key.

 

 

I've met a number of people through Stereonet who fall into the 'golden ears' category.  They've educated me to what to listen to and for, and can put into words things that to me are just a feeling. Even so, my wife puts it nicely: "it's like a blanket has lifted off the music".  

 

Anyway, this test would have been more interesting with golden ears listeners. It's like relying on the average person walking into a bottle store for whisky or wine reviews, rather than a distiller or wine-maker.

 

Another poster points out:

 

I'd say you'd need to have some subjects who have demonstrated an ability to discern small, but known audible differnces between tracks - not related to changing cables.

 

But the conceptual question remains: are we trying to determine if most people, or no people can tell a difference? Say they do a large test and only 2 "experienced" listeners can differentiate between the 2 cables - but pass an additional test or two. In other words, not chance. 

 

What's the conclusion?  the cables don't make a difference? for most people? or The cables usually don't make a difference, but specific individuals can tell a difference?

[Addicted to music]

Did you actually read the article?

. These "scientific" people couldn't even conduct a test that meets the most basic scientific standards. So they showed that 6 people failed to discern a difference - in their test. But the data has no application to any more general conclusion than that. You can believe what you want. But this "test" doesn't back up any general conclusion with any scientific authority. On the level of general applicability, the "test" merely added 6 anecdotal data points of no general value.

Beyond the problems revealed in the article, the so called test was a failure already in conception. I'd say it was a demo or a show, but certainly not a scientific test. And that's before you get to the other problems with the implementation.

When it comes to performances in electrical wire construction, why am I wasting my time reading the article, I did post previously that I use free 6.00mm electrical cable for speaker cable. Audioquest can market what they want, I'm in no position to outlay hard earn cash for cables that are designated a limited standard. If I were to gain improvements in my rig, it won't be cables....it's like say vinyl has better bandwidth than CD, that's why it sounds better, yet the human hearing is limited to around 20khz, and as we age this degrades as some of our receptors for hearing high frequency continues to get destroyed, the louder and the length of time we are exposed to high SPL the greater these receptors 'die out' I'm sure that most of the test subjects are in there 40-70. So whatever test they conduct, would you trust these test? And how detail a test is conducted, there is always critics for every one done... Edited August 1, 2015 by Addicted to music

[davewantsmoore]

I'd say you'd need to have some subjects who have demonstrated an ability to discern small, but known audible differnces between tracks - not related to changing cables.

 

Not if you are attempting to check a claim of "night and day difference"  (which they were).

 

 

....  but people are right to say that the general applicability of the test to ethernet cables picking up or transmitting electrical noise, is completely useless...   but no sane person would use a listening test for that anyways.

[Addicted to music]

Not if you are attempting to check a claim of "night and day difference"  (which they were).

 

 

....  but people are right to say that the general applicability of the test to ethernet cables picking up or transmitting electrical noise, is completely useless...   but no sane person would use a listening test for that anyways.

It's all additional free marketing for Audioquest.....

[bob.cob]

Whether you're talking about hi res audio, fancy ethernet cables or Pepsi - anything requiring blind ABX testing for the average mug to discern a difference - well I think we all need a reality check   and worry about things that actually matter.