Ars prepares to put “audiophile” Ethernet cables to the test in Las Vegas

[wgscott]

Hasn't this already been done for individuals? I seem to remember reading reviews/articles where the professional listener did such a thing.

 

I don't know.

 

One of my problems with the whole subjectiveist/objectivest/DBT debate is that I think there are experienced/professional listeners like Barry D. that have learned to hear things about 99% of listeners can't.

 

Which really should be enough to convince the skeptics. The differences have to be present in order for them to be identified by anyone, including your dog. If Barry (sorry to personalize this) can hear differences that others cannot, and the skeptics (me included) want to make the claim that his hearing is super-normal, then that becomes the extraordinary claim that needs to be tested.

 

I had this argument on Hydrogenaudio a few years ago. It didn't go over well.

[mmerrill99]

I don't know.

 

Which really should be enough to convince the skeptics. The differences have to be present in order for them to be identified by anyone, including your dog. If Barry (sorry to personalize this) can hear differences that others cannot, and the skeptics (me included) want to make the claim that his hearing is super-normal, then that becomes the extraordinary claim that needs to be tested.

 

I had this argument on Hydrogenaudio a few years ago. It didn't go over well.

I just read your posts on HA related to the same topic & you made some very good points there.

I agree that those who are trained in recognising such audible differences should be the test subjects & the test should be done by professionals who know how to do perceptual testing - but I didn't think this was the case here. Barry, despite his claims, may not be one such trained person. As is often the case such tests are set-up to challenge a person or group of people - which is not the way to begin to setup such tests

 

I saw you post on HA that you are a biophysical chemist & I doubt you would be inclined to accept tests done by people with no training in your area of expertise. Perceptual testing is an area that requires expertise in if valid tests are to be done.

[Paul R]

I had this argument on Hydrogenaudio a few years ago. It didn't go over well.

 

This particular discussion *never* ends well. Even though the logic of it, from all sides, seems pretty crystal clear to me.

 

-Paul

[jabbr]

There are two issues:

 

(1) Flaws in the experiment

 

(2) Flaws in the interpretation (which would persist even in the hypothetical case that the experimental design itself is perfect).

 

Can we just focus on one of these at a time.

 

I'll take your word for (1), although I confess I don't fully understand why. But lets assume, just for the sake of discussion, that we can eliminate the objections in (1), so now what happens when one test subject alone reliably can identify differences.

 

These are very important points in case people aren't following.

 

What happens, is that when a difference is heard, people who have a preconceived idea that this is not physically possible need to challenge their own assumptions.

 

For example, we know that various Ethernet cables are perfectly capable of transmitting Ethernet data with a low or near zero error rate. (I have posted somewhere a very easy way to measure Ethernet errors). So if the assumption is that differences in Ethernet cables is due to differences in Ethernet packet drops, this is very easily tested. Let's assume they are close enough to rule this out as an explanation for audible differences. Does that mean that it is impossible that audible differences don't exist? Certainly not. So why not consider out of band signals transmitted by the Ethernet cables. Since these signals are not part of the Ethernet spec, there is no reason to suspect that different cables will transmit these other signals in a similar, hardly identical fashion.

 

Now suppose what I've said is the explanation -- this suggests that what we really ought to be focussing on is eliminating these out of band signals, rather than picking cables which modulate these out of band signals (in probably random fashions). That is to say that if a $1000 Ethernet cable is capable of diminishing an out of band signal by 60%, would anyone rather have a $15 cable that can completely block the same signal?

[mmerrill99]

Yes, agreed, Jabbr - once the underlying mechanism of operation is known, people can produce better solutions. At the moment, we are still in discovery stage (as we are with a number of areas in audio) so this is not yet the case.

[jabbr]

Yes, agreed, Jabbr - once the underlying mechanism of operation is known, people can produce better solutions. At the moment, we are still in discovery stage (as we are with a number of areas in audio) so this is not yet the case.

 

 

Another way of working has been called a "blackboard" where various hyps are thrown up with rough probabilities of "correctness" and solutions are generated from each hyp and the hyps are reordered based on new information that adjusts the probs. Eventually low prob hyps are discarded and hopefully the system comes up with a final solution. Alternatively called "fuzzy logic". Working with imperfect information is challenging.

 

Let's say that if I say I hear a difference, that gets a prob=0.2, whereas if Barry D hears something, prob=0.8 and random DBT prob=0.6 but well done DBT prob=0.95 and then a formula updates the sum prob based on all these... so we say, jabbr hears something, put on back burner, but Barry D hears so let's investigate, perhaps start really good DBT

[Paul R]

These are very important points in case people aren't following.

 

What happens, is that when a difference is heard, people who have a preconceived idea that this is not physically possible need to challenge their own assumptions.

 

For example, we know that various Ethernet cables are perfectly capable of transmitting Ethernet data with a low or near zero error rate. (I have posted somewhere a very easy way to measure Ethernet errors). So if the assumption is that differences in Ethernet cables is due to differences in Ethernet packet drops, this is very easily tested. Let's assume they are close enough to rule this out as an explanation for audible differences. Does that mean that it is impossible that audible differences don't exist? Certainly not. So why not consider out of band signals transmitted by the Ethernet cables. Since these signals are not part of the Ethernet spec, there is no reason to suspect that different cables will transmit these other signals in a similar, hardly identical fashion.

 

Now suppose what I've said is the explanation -- this suggests that what we really ought to be focussing on is eliminating these out of band signals, rather than picking cables which modulate these out of band signals (in probably random fashions). That is to say that if a $1000 Ethernet cable is capable of diminishing an out of band signal by 60%, would anyone rather have a $15 cable that can completely block the same signal?

 

Whoa there mate - it is very very unlikely indeed you have true out of band transmissions traversing your home network. What most people call out of band transmissions now are just transmissions on a different port, or with a packet that is in some manner, unexpected.

 

Not on a different frequency band.

 

No real physical difference in the transmission of on out of band packet these days. Not on twisted pair or wireless home networks.

[jabbr]

Whoa there mate - it is very very unlikely indeed you have true out of band transmissions traversing your home network. What most people call out of band transmissions now are just transmissions on a different port, or with a packet that is in some manner, unexpected.

 

Not on a different frequency band.

 

No real physical difference in the transmission of on out of band packet these days. Not on twisted pair or wireless home networks.

 

Don't mean to use the wrong terminology. By "out of band" I just meant stuff that isn't the Ethernet bits. I would consider any electrical signal on the wire that isn't the Ethernet bits to be "noise". I am just saying that it is possible that different cables -- all of which transmit Ethernet perfectly well -- might be transmitting trash differently and that is how the "SQ" could reasonably be different.

[joelha]

I don't mind people disagreeing with me at all.

 

When someone says they have faith in a god doesn't phase me none.

 

When someone says some deity of a higher plane just friended me on facebook and they are coming over later to crash at my place. Wellllll...

 

plissken,

 

I appreciate what you've written here. I've often believed there's a religious (or anti-religious) component to these debates/discussions.

 

I'm hoping you regard your example as extreme. If not, then it goes to show how far apart we are in our perceptions of what is going on in our hobby.

 

But I don't think it's an accident that your analogy was religious in nature. I suspect many who must have measurements to validate listeners' perceptions in our hobby can't accept religion in their lives because faith is also something which one can't measure.

 

That point doesn't make either of us right or wrong. It's just an observation.

 

Joel

[Paul R]

plissken,

 

I appreciate what you've written here. I've often believed there's a religious (or anti-religious) component to these debates/discussions.

 

I'm hoping you regard your example as extreme. If not, then it goes to show how far apart we are in our perceptions of what is going on in our hobby.

 

But I don't think it's an accident that your analogy was religious in nature. I suspect many who must have measurements to validate listeners' perceptions in our hobby can't accept religion in their lives because faith is also something which one can't measure.

 

That point doesn't make either of us right or wrong. It's just an observation.

 

Joel

 

While I agree with you, the flip side is also true... some people cannot face the idea that there might not be a God, or any higher being out there who cares for us. That is indeed a hard thing to face, and there are plenty of analogies in the audio wotld. On all sides, of course.

[mmerrill99]

Don't mean to use the wrong terminology. By "out of band" I just meant stuff that isn't the Ethernet bits. I would consider any electrical signal on the wire that isn't the Ethernet bits to be "noise". I am just saying that it is possible that different cables -- all of which transmit Ethernet perfectly well -- might be transmitting trash differently and that is how the "SQ" could reasonably be different.

Exactly, all differential signalling is prone to common mode noise. Differential signalling, as is used in ethernet, USB, LVDS, etc. uses two signals which are 180degress out of phase with one another. Adding these two signals together should result in a linear DC voltage. CM noise arises when the two differential signals are unbalanced which causes a non-linear & fluctuating DC voltage = common mode noise. This can arise for many reasons - some of which have to do with the the cable .

- Uneven/unbalances twisting of the twisted pair - one wire will be longer than another giving rise to a phase shift in one of the differential signals (the 4 pairs of wires in ethernet cables have different twist rates so as one adjacent pair doesn't interact with the differential signalling of another adjacent pair)

- Running cables near power lines or electrically "noisy" devices

- Avoiding the minimum bend radius

- skin effect of the high frequency signals

 

There are other factors outside of the cables themselves that cause unbalanced differential signalling but these cable factors can vary from one cable to another as can the cable shielding

 

The use of ethernet transformers on either end of the transmission can reduce some but not all of the common mode noise

[joelha]

While I agree with you, the flip side is also true... some people cannot face the idea that there might not be a God, or any higher being out there who cares for us. That is indeed a hard thing to face, and there are plenty of analogies in the audio wotld. On all sides, of course.

 

Paul,

 

I agree with you. No question.

 

The only reason I made the point is that it often seems that those who are opposed to evaluating products based on experience only are more emotional and more persistent about their views than those they disagree with.

 

Joel

[Paul R]

Paul,

 

I agree with you. No question.

 

The only reason I made the point is that it often seems that those who are opposed to evaluating products based on experience only are more emotional and more persistent about their views than those they disagree with.

 

Joel

 

Semantic analysis from this kind of writing is hard, at least for me. There are none of the normal tells that clue one into the way the writer is feeling.

 

Which leads me to suspect that both sides feel more or less persecuted, and react accordingly. The truths probably somewhere int he middle of course, but occasionally, one gets handed a zinger in audio. Such as USB cables sounding different.

 

Then all heck breaks loose, with wild theories, impossible happenings, conspiracy theories, and everyone drawing erroneous conclusions based upon incomplete data and incomplete understandings.

[bluesman]

Then all heck breaks loose, with wild theories, impossible happenings, conspiracy theories, and everyone drawing erroneous conclusions based upon incomplete data and incomplete understandings.

 

[firedog]

The test has taken place. Why no results?

[wgscott]

The test has taken place. Why no results?

 

Sometimes it takes a bit of time to alter the data to conform to the expected result.

[davidbeinct]

Sometimes it takes a bit of time to alter the data to conform to the expected result.

 

I used to work with a guy who had a t-shirt that read "When All Else Fails - Manipulate the Data!"

[plissken]

plissken,

 

 

I'm hoping you regard your example as extreme.

 

But I don't think it's an accident that your analogy was religious in nature.

 

I was angling for efficiency of what I wanted to express. Nothing more nothing less.

[plissken]

As a person more in the objective camp I'm willing to concede that I have biases that play into perception.

 

I've sat for blinded testing and not been worried about it at all because I don't have an agenda I'm looking to fulfill going into it.

 

To borrow a phrase I can not trust someone that doesn't trust their own ears. Or just like Mystery Men where the only way the invisible man can turn invisible is if no one watches him.

 

In regards to Ethernet and someone saying the differences are plainly apparent and easily discernible. I find it all sorts of irony that the statement can't be tested with out a lot of arm waving going but, but,....

[senorx]

I'd say the presence of "Randi," sets this up as a "debunking" of the ability to hear differences in ethernet cables. It's hard to have faith in the results of a test when it is being conducted by people who seem to already have an agenda. It is also not clear to me that different necessarily means better, maybe, maybe not. Maybe better for someone and worse for someone else. As far as I am concerned, regardless of the results of this or any other test, an Ethernet cable sold for $300 or even $75 per foot is a rip off. One will, however, never catch me arguing that a person should be criticized for willingly allowing themselves to be fleeced. I've got no horse in that race. Couldn't care less.

[Tony Lauck]

I completely agree & it is the usual problem that is encountered with perceptual testing by amateurs - the test is full of experimenters bias which is guaranteed to give a null result.

 

As you said the null hypothesis isn't clearly established - it looks like it's a test to answer the question you suggested "whether the {average, typical, normal} person can hear a difference" & I agree with you - the answer is known to be "no". The same as would be achieved for any blind testing of almost anything that doesn't have gross, obvious differences - wine, etc.

 

What is disingenuous about this is that the supporters of blind testing who know full well that the test is flawed will stay quiet about these flaws, preferring instead another null feather in their already festooned cap which they pretend not to be wearing ("because null results prove nothing"). If null results prove noting why do they waste everybody's time with flawed blind tests which are guaranteed to return null results

 

Leave perceptual testing to honest experts in the field who aren't agenda driven, says I.

 

Any demo can be rigged. If the cable vendors are funding it, I wouldn't be in the least surprised if there are "positive" results. If the Great Randi is funding it I can be sure the $1,000,000 dollars won't be paid, etc... There people are professionals, if not professional scientists. They do their due diligence.

 

As to agenda driven, the phrase is, "There are no unbiased experts."

[Tony Lauck]

I think you are misunderstanding me. Let's say you do the tests on 1000 people. 999 people's results support the null hypothesis. One does not.

 

What do you conclude, or do next?

 

I think the best thing would be to test that 1 person several more times, perhaps under more careful conditions, etc., to see if this is simply a statistical anomaly, or whether they can reproducibly detect the differences. (If it is a statistical fluke, then the more repeats you try, the less likely you are to see a positive outcome survive).

 

If that one person can reproducibly detect differences, then potentially audible differences do in fact do exist.

 

It could be that someone could sit the 999 people down and train them for an hour and then the test could be repeated and 500 people could pass. (Similar things have happened in blind tests of a CODEC.)

 

The phrase "support the null hypothesis" is misleading. The correct phrase is "fail to reject the null hypothesis". Two different phrases, two different meanings, two different implications.

[wgscott]

The phrase "support the null hypothesis" is misleading. The correct phrase is "fail to reject the null hypothesis". Two different phrases, two different meanings, two different implications.

 

I feel like I just got slapped by Karl Popper. (You are of course right, although he might have said "corroborate".)

[plissken]

I'd say the presence of "Randi," sets this up as a "debunking" of the ability to hear differences in ethernet cables. It's hard to have faith in the results of a test when it is being conducted by people who seem to already have an agenda.

 

I feel the same exact way when vendors put on dog and pony shows.

 

The difference being with The James Randi forum they will openly detail the setup and methodology. Something I can't say for AQ or Chord.

[esldude]

 

To borrow a phrase I can not trust someone that doesn't trust their own ears. Or just like Mystery Men where the only way the invisible man can turn invisible is if no one watches him.

 

.

 

See my experiences lead me to the opposite conclusion. I once believed I could trust my own ears, it became apparent I couldn't. And not because I have lousy hearing ability (it does seem at least roughly average), but because the way human hearing works you can't trust anyone's hearing in many circumstances.

 

I agree about the hand waving on ethernet differences. To me to be worth the high end money, it should be something as obvious as a 3 db drop at the frequency extremes. Something that is not necessarily stupendously obvious, yet enough it would be quite simply heard in a few tens of seconds of listening with comparison to the other version of the signal. That sort of difference isn't as obvious as some believe, but is something that can reliably be discerned.