Is Audiophiledom a confidence game?

[jabbr]

1 hour ago, pkane2001 said:

But, the bigger picture is that there is no objective, peer-reviewed test of any design or methodology that demonstrates a difference in sound between two ethernet cables. None.

Nor am I aware of any peer reviewed published study that says that Ethernet cables can’t affect SQ.

 

Theres no actual science to say one way or the other. 

 

I don’t waste my time comparing Ethernet cables and am very skeptical of claims promoted by people who are selling snake oil, but there is no scientific basis to call someone delusional for hearing a difference.

 

Thats the bigger picture.

[jabbr]

Let’s me also say that as someone who is intimately aware of Ethernet including potential pitfalls, it is my view that potential areas where cables might get affect SQ are best handled by optimizing the interface itself, including DAC isolation, use of fiber etc, and that fooling around with cables that are reasonably good, or specifically trying to create “audiophile” Ethernet cables is an utter waste of time.

[pkane2001]

19 minutes ago, jabbr said:

Nor am I aware of any peer reviewed published study that says that Ethernet cables can’t affect SQ.

 

Notice that I didn't say that, although there are measurements that bear that out.

 

I didn't call anyone delusional. What I said was:

Quote

There are those that speculate about ground loops and noise transmission as the mechanism for making these cables audible. This may happen in some systems. That's fine,  but the way such issues get resolved is not by swapping ethernet cables for more expensive ones, but to provide proper noise isolation in the circuit or to use a different transmission mechanism that isolates noise. 

 

As you can see, I'm open to the possibility that noise transmission might be the reason for audible changes in some systems, but swapping ethernet cables to accomplish noise filtering is the wrong approach. Do you disagree?

[jabbr]

Just now, pkane2001 said:

As you can see, I'm open to the possibility that noise transmission might be the reason for audible changes in some systems, but swapping ethernet cables to accomplish noise filtering is the wrong approach. Do you disagree?

I entirely agree and to clarify: was responding to the larger discussion between these 2 threads rather than quoting you in particular 

[plissken]

Here is what I would consider a piece of evidence:

 

Create 0dBfs wav file of silence (yes it's a thing). Play back that wav file over a network and capture the output and inspect it. 

 

While playing back that file swap out two, like constructed, spec passing, Ethernet cables and show me the difference in the capture. 

[crenca]

1 hour ago, jabbr said:

Nor am I aware of any peer reviewed published study that says that Ethernet cables can’t affect SQ.

 

Theres no actual science to say one way or the other. 

 

I don’t waste my time comparing Ethernet cables and am very skeptical of claims promoted by people who are selling snake oil, but there is no scientific basis to call someone delusional for hearing a difference.

 

Thats the bigger picture.

 

46 minutes ago, jabbr said:

I entirely agree and to clarify: was responding to the larger discussion between these 2 threads rather than quoting you in particular 

 

jabbr, what would you call it then if not delusional?  If it is as pkane2001 says it is, that while we can not scientifically prove what these good folks are hearing is some placebo/expectation bias/etc. but that we have a large body of experience and foundational EE "principal" (so to speak - phrase it as you wish) that points that all these diverse and contradictory experiences are most likely can not be justified as actual electrical/mechanical phenomenon, what would you call it?

 

Now, take the above fact and see it in the light of an industry, a culture called "audiophiledom" that absolutely is drunk on these kinds of experiences of the unknown and strange (in my system, I can hear Close Encounters of the Third Kind!") in otherwise fairly well understood electrical systems, and what do you have?

 

I submit, you have a confidence game.

 

Remember, the confidence game is not primarily built on the deception of the manipulator, but on the natural trust and social conditioning of the victim.  Is Audiophiledom a coalition of victims who like their victimhood and defend it endlessly?

[jabbr]

50 minutes ago, crenca said:

jabbr, what would you call it then if not delusional?

 

First of all the only thing that causes some emotion in me are claims by cable manufacturers and folks wandering over pretending to ask questions but really pushing their agenda of selling snake oil: they aren't delusional, they are manipulative.

 

I have absolutely no emotion about honest folks who claim to hear differences in cables. I have some curiosity about whether they are experiencing a placebo effect, or alternatively whether there is an as yet inadequately explained physical basis for what they hear -- could very well be a bit of both -- I've suggested that alteration in leakage currents could very well be an explanation.

 

I have made many of my own cables and clearly hear differences among RCA interconnects. I attribute these to differences in shielding and the local noise environment. I have experienced clear ground loops related to USB cables (and have mitigated). I believe that fiberoptic Ethernet sounds "cleaner" to me, and attribute this potentially to leakage currents. There are other reasons that different connection types might sound different however. I only have well made Ethernet cables either those that I've purchased or made myself (this is really easy).

 

I strongly advise people against wasting their money on audiophile Ethernet cables. I suggest that Belden 10g with REVConnect connectors is SOTA for copper as far as I know, and would get this. I would also eliminate cheap SMPS power supplies and use either batteries, supercaps or good isolation transformers (I have both Topaz and Equi-tech in various configurations). 

 

So to summarize, I think folks may be hearing a symptom of another "problem" in their system that are best dealt with by measures such as above. I agree with @plissken that you can make a well made Ethernet cable that should be equal or better than any multihundred or thousand dollar "audiophile" cable (REVConnect connectors aren't the cheapest so if audiophiles want to "splurge" here is a great option). Frankly I'm not interested enough in whether copper Ethernet cables might sound different to spend any time testing this with my ears. This really doesn't interest me (ala "How I stopped worrying and learned to love the bomb"  

 

No I don't think people are delusional for hearing differences in cables (even if its placebo)

[jabbr]

1 hour ago, plissken said:

Here is what I would consider a piece of evidence:

 

Create 0dBfs wav file of silence (yes it's a thing). Play back that wav file over a network and capture the output and inspect it. 

 

While playing back that file swap out two, like constructed, spec passing, Ethernet cables and show me the difference in the capture. 

 

There is a huge difference between proving that two things are the same and proving that they are different.

 

To prove A and B different, all I need is one verified datum of difference. To prove the same I need all possible data.

 

If the above test shows a difference, then there you go -- the cable makes a difference when connected. If not, the above test doesn't prove anything.

 

For example, this test could possibly only work with a wave file of pure 1kHz?

[Rt66indierock]

7 hours ago, Teresa said:

 

 

 

Level matching only works if both devices or music files under test have the same dynamic range. After average levels are matched, the device with the lowest dynamic range will be judged the best as it will be the loudest during many parts of the music, as humans pick the loudest as sounding the best in A/B tests. This is the opposite of reality as the one with the widest dynamic range should sound the best.

 

I submit that except for level differences, large amounts of noise or distortion that it is impossible for most human beings to tell if two things sound different. I believe the best that we can do is live with the device for a few weeks to determine if we like the effect of the device in our system or not.

 

Teresa I was talking about volume matching. John Darko(Digital Audio Review) liked the MQA version of a song on The Nightfly because it was 2 dB louder than the other version he was comparing it to. Its an old trick of the industry louder is preferred. More people than me beat him up pretty good at RMAF.

 

I got to disagree with second paragraph. Every room I listened to music in at RMAF sounded different. 

[plissken]

8 minutes ago, jabbr said:

 

There is a huge difference between proving that two things are the same and proving that they are different.

 

To prove A and B different, all I need is one verified datum of difference. To prove the same I need all possible data.

 

If the above test shows a difference, then there you go -- the cable makes a difference when connected. If not, the above test doesn't prove anything.

 

For example, this test could possibly only work with a wave file of pure 1kHz?

 

The reason for the 0dBfs silence is to expose the noise that is not part of the wav file. A sample rate doesn't apply here. 

[jabbr]

4 minutes ago, plissken said:

 

The reason for the 0dBfs silence is to expose the noise that is not part of the wav file. A sample rate doesn't apply here. 

 

Okay, can't go through all this, but the leakage current of Ethernet is known not to be 0 A. It has been measured in medical environments and that's why isolators are made. Right there -- connected vs unconnected will give a measurable difference in leakage current.

 

 

There are many many variables. You'd need to systematically go through and test against each of the known variables and then you'd simply be left with the variables you haven't tested for.

 

[plissken]

1 hour ago, jabbr said:

 

Okay, can't go through all this, but the leakage current of Ethernet is known not to be 0 A. It has been measured in medical environments and that's why isolators are made. Right there -- connected vs unconnected will give a measurable difference in leakage current.

 

 

There are many many variables. You'd need to systematically go through and test against each of the known variables and then you'd simply be left with the variables you haven't tested for.

 

 

Ok. So does a boutique Ethernet cable obviate the need for a KVA rated isolator?

 

All the absence of leakage current will show is when we disconnected the cable if it's showing up in the output of the DAC.

[Ralf11]

6 hours ago, mansr said:

...

(Jud, please spare us your standard reply about the limitations of echoic memory; it's not relevant.)

 

 

and not correct either

[esldude]

3 hours ago, jabbr said:

 

There is a huge difference between proving that two things are the same and proving that they are different.

 

To prove A and B different, all I need is one verified datum of difference. To prove the same I need all possible data.

 

If the above test shows a difference, then there you go -- the cable makes a difference when connected. If not, the above test doesn't prove anything.

 

For example, this test could possibly only work with a wave file of pure 1kHz?

Well you have to start somewhere.  Looks like plenty of people clearly don't want to actually test anything.  The placebo is too addictive.  

 

I find it curious so often we hear of people saying I don't want to spend time listening for X this way.  Yet they usually are the same who say they listen for weeks to evaluate the effect until they are comfortable with it and make judgments that way.  Sounds inherently contradictory to me.  

[esldude]

6 hours ago, lucretius said:

 

Jud hasn't been around for a while.

I have noticed that.  I wonder if it is related to moving to New Mexico?

[esldude]

9 hours ago, firedog said:

 

Or they just don't like doing such tests and aren't really interested in putting time and effort into it. 

Yet (not speaking of you in particular) they put orders of magnitude more time and effort and money into the same thing in a way that can never be anything other than comforting placebo effects.  Spending 30 minutes listening carefully is too much time and effort while spending weeks listening and worrying and thousands in money is not too much?

Quote

 

I can't properly "test" anything at home. I compare the best I can and  make up my mind. Then I don't worry about it anymore, and ignore what everyone else says. I'm interested in whatever sounds best to me. Don't care if someone else thinks it makes sense or not. Obviously, I don't want to spend thousands on snake oil-but exclusive of that I don't really care. 

So where do you get the ideas you do try out?  I don't know your normal methodology, if you could improve it why wouldn't you? Again don't know how you do it, but I almost never can get anyone to do level matching.  That one thing really improves results.  Somehow that is too much.  Touchy, feely makes me feel good approaches are the rule.  

 

Quote

 

I've spend a bit of money on various power improvements. Do I know from DBT's that they sound better? No - but I think they've improved the SQ for me. Maybe I'm fooling myself. Some people would say so.

But I'm satisfied that within the limits of my room and my finances, I've done the best I can, and I've prevented additional episodes of  "audiophillia nervosa". Any of you can tell me you've DBT'd one of my devices and it has no real effect. I won"t care.

 

Yet most people do care if someone tries a device and says it has a real effect.  Even when the method determining that effect is horridly flawed.  That is what seems strange, the ill will toward better methods.  Methods that can dramatically shrink the nervosa issues. 

Quote

 

 

Even the placebo effect is "real" to the person hearing it. We keep forgetting we actually hear with our brains and how they interpret sound, and not with the receptors in our ears.

They have to account for "the placebo effect" in medical research b/c it actually exists - i.e., it can actually cause improved outcomes. The person in the trial who is given the sugar pill and gets better doesn't really care about the pill in the end. He cares that he got better. His/her experience may not be applicable to anyone else, but the outcome is still an actual outcome for him/her.

 

So maybe that is the secret.  If you get cured of nervosa once that is all well and good.  But if you can get the nervosa now and again, and have a sure fire method that always cures it (though always temporarily) then you get the repeated placebo high of fixing your problems.  A cure that never ends and an emotional payoff that happens over and over.  Something like forgiveness from sin in a religion. 

[plissken]

I've offered, in the best faith I know how, to explore this in a transparent fashion that asks neither party to participate without some recompense for their time *if* the respective claims don't hold up to scrutiny. I've even taken that rather extraordinary step of going and doing this on the claimants own setup. When I offered this to ML with an increase to $10,000 to his $2,000 he told me to eat shit. 

 

What I've seen to date:

 

1. No technical counter-argument that dismantles my suggested testing rig that is based loosely on Rene van Es's documented setup. 

 

2. No takers for what should be an easy $2000. 

 

My conclusion thus far has been singly that people are deluding themselves based on the above. It is what it is. It's a bit of rationality kicking in as to why no one will take me up on the offer. 

[jabbr]

1 hour ago, plissken said:

What I've seen to date:

 

1. No technical counter-argument that dismantles my suggested testing rig that is based loosely on Rene van Es's documented setup. 

 

2. No takers for what should be an easy $2000. 

Don’t draw any conclusions based on your proposal. I don’t agree that your suggested testing rig is valid but don’t really want to go back and forth on nitty gritty details and meandering arguments about how to do the testing. I gave a broad outline. 

 

Dont assume that what you call an “easy $2000” would be a good use of my time. I turn down consulting offers all the time. I also can say that what you are testing wouldn’t come out the way you claim.

 

In summary: you may be right or you may be wrong. This issue is not important enough for me to spend hours on. 

[plissken]

Just now, jabbr said:

I also can say that what you are testing wouldn’t come out the way you claim.

 

What am I testing? and I'm surprised that you are clairvoyant. 

[fas42]

12 hours ago, marce said:

But we come accustomed to the sound...

 

Some may become acustomed to the sound, but I don't. Once I discern the distortion artifacts of a particular setup, its 'signature', I keep hearing it, with every playback. And that annoys the hell out of me ... to the point where I just won't use the system - seriously; that is, it then becomes a background muzak machine until I can do what's needed to sort it.

[jabbr]

2 hours ago, esldude said:

Well you have to start somewhere.  Looks like plenty of people clearly don't want to actually test anything.  The placebo is too addictive.  

 

I find it curious so often we hear of people saying I don't want to spend time listening for X this way.  Yet they usually are the same who say they listen for weeks to evaluate the effect until they are comfortable with it and make judgments that way.  Sounds inherently contradictory to me.  

Again, I’m not interested enough in copper Ethernet cables to even listen for differences. I’m happy with the cables I have and life is short.

 

I find that if I try to listen for SQ differences to too many variables at the same time that my brain gets fried. I try to listen only for major schematic changes which interest me.

[jabbr]

1 minute ago, plissken said:

 

What am I testing? and I'm surprised that you are clairvoyant. 

I’m not sure what you are testing — but you repeatedly claim to draw conclusions based on the fact that no one has taken you up on your offer. I’m just suggesting that conclusions not be drawn because I for one am not really that interested in whether Ethernet cables can affect DAC sound, so at least in my case that doesn’t mean that I agree with the testing protocol.

[plissken]

Thanks for the clarification.

 

I'm testing claims. 

[Audiophile Neuroscience]

8 hours ago, pkane2001 said:

 

 You seem intent on picking on Plissken and his proposed test. Fine, nobody said that it's perfect,  and maybe it deserves to be picked apart. No argument.

 

Maybe you mean "picking on Plissken" because I have somehow singled him out as representing an ideology or belief that is counter to my own. However, I have no qualms with properly conducted research and my theory is that expectation bias is alive and well. Plissken singled himself out when IMO he crossed the line, calling a professional a blatant liar, with proof, and was therefore challenged.That's not unreasonable IMO.

 

Now, when challenged about Chris C doing sighted reviews his attitude appeared very different. "taken with a grain of salt" was his answer. IMO a reasonable and defensible viewpoint. So why is his"scientific" conclusions so utterly different if not bias, specifically against ML?

 

8 hours ago, pkane2001 said:

But, the bigger picture is that there is no objective, peer-reviewed test of any design or methodology that demonstrates a difference in sound between two ethernet cables. None.

 

That would be my confident guess and for the reasons that there are no valid tests as yet. If you don't have the test you can't have an outcome either for or against. There is the whole "falsifiability" thing and when does an absence of evidence start becoming evidence? I concede that.

 

8 hours ago, pkane2001 said:

 

What there is, though, is a very real and extensive knowledge of physics and computing that says there will be no difference, and a lot of experimental and measured data showing that there is no audible difference (if the cable is functioning within specs)

 

Let's break this down

1) suppose I said you are a blatant liar by making that statement and I have proof. A test I devised but no one has taken but I am happy for you to take it and prove to yourself how wrong you are. You decline but I say no matter I have proven you a liar.

 

2) IMO you have misconstrued the situation. The physical facts are indeed facts. They suggest there will be no audible difference IOW a *theory*. It may be a very good theory and you may posit that if such difference occurs it is *not* due to xyz (bit differences, waveforms, whatever). The strength of the theory goes up the more you can eliminate variables.

 

I agree you have a strong theory based on objective common sense reality.The whole trouble with 'objective reality' is that it turns out to be malleable in light of new evidence, sometimes.

 

3) So there is point 2 , BUT in this case however you are just interpreting the data to suit your theory, the theory that certain undeniable facts say there is no audible difference. You may be *hearing* that's what the facts say (analogous to hearing audio) but as we know, expectation bias affects "hearing".  The conclusion of inaudibility is a leap in logic. As stated, there is no way as yet to scientifically test this conclusion. Theories are great, they are the stuff of predictions and basis for future research. You cannot short circuit this process because it conveniently suits your belief system. Falsifiability of theories is a strong counter point but IMO not definitive.

 

Objectivity is not just about data points and the notion of "common sense realism", in an epistemological sense it includes objective attitudes and methodology.

 

 

8 hours ago, pkane2001 said:

There are those that speculate about ground loops and noise transmission as the mechanism for making these cables audible. This may happen in some systems. That's fine,  but the way such issues get resolved is not by swapping ethernet cables for more expensive ones, but to provide proper noise isolation in the circuit or to use a different transmission mechanism that isolates noise. 

 

Well okay, fine, unless the more expensive ethernet cables do just that eg address noise isolation or whatever - and you have eliminated all other possible variables, and your tests are completely accurate.

 

8 hours ago, pkane2001 said:

 

If you are a scientist in more than a name, you must understand the need for a proof beyond hearsay when the claim is unsupported by the existing, established science.

 

That is precisely my claim. ....If you are a scientist in more than a name, you must understand the need for a proof beyond hearsay when the claim is unsupported by the existing, established science. I have explained this above. I am happy to discuss further if you want.

 

 

8 hours ago, pkane2001 said:

 

If you think that ethernet cables are audible, please propose an objective, repeatable test methodology that will prove it rather than picking on someone who is trying to do exactly that. 

 

I have never stated that I do hear differences in ethernet cables. I do not find it incumbent on me to develop such a test. I am not making claims one way or another, simply pointing out sloppy science is not the way to resolve anything IMO. I would be delighted if someone comes up with a valid test and demonstrtes its validity etc. You cannot just toss it around and say well, yeah, that seems about right. Its like using an uncalibrated instrument.

 

I am sincerely puzzled at times as to why this seems so difficult to grasp. I am not being adversarial. It's just that if you invoke science then you must abide by it.

 

 

[jabbr]

3 minutes ago, plissken said:

Thanks for the clarification.

 

I'm testing claims. 

I’m making absolutely no claim that any so-called audiophile Ethernet cable changes the SQ, nor would I purchase one. I know you haven’t heard a difference with Audioquest which would be my expectation.

 

Again I think the way we differ is that I’m not certain there are no SQ differences but if there were I would consider them artifactual.