The Great Cable and Interconnect Swindle: An Etiology

[Julf]

Basically my argument is this: the unique characteristics of audio have provided an environment where this phenomenon has been able to gain traction. You would not expect to, and will not find an analogy in photography for example. Due to current legislation the companies who peddle this expensive stuff are able to get away with not proving up their claims. Audiophile land is a bit of a backwater, so this is not surprising. It seems the con has gained enough momentum now for it to have reached the stage where the sheer number of believers has given it an elevated status.

 

Totally agree with both your initiative and your argument - but unfortunately it opens up a veritable pandora's box of issues...

[Julf]

So what we have here is a blog where flat earth folks can congregate, and make themselves feel better in their ignorance, laughable.

 

And what are the odds that they are allowed to do that without the threads getting clogged up with people ridiculing them?

 

Even more amusing is that it is quite simple to test whether an audiophile cable actually improves performance, as The Cable Company maintains a rather extensive library of lending cables, which can be tested in one's system for very little cost. Additionally, many good dealers will allow for at home cable demos. The intellectual "excersize" of once again describing how cables "could not possibly make a difference" is moot considering the extensive evidence to the contrary.

 

I suggest another test that can be done easily - and provides answers that can be verified beyond individual listening rooms. What we need is someone who has a high-quality, high-res ADC/sound card to record the output of a good DAC playing a piece of music - and record it twice, once using a cheap, generic cable and once using a fancy, audiophile cable. Better yet, record the piece three times - picking one of the two cables randomly for the third recording. Then make all three recordings available online, and let CA members do a blind ABX. Do this with enough people and multiple music samples, and you should get a pretty reliable result. But what would we then argue about?

[Julf]

In general nobody is deliberately trying to fool anybody in audiophile land.

 

I hope you are right. But I do find that Mark Twain's old "It's easier to fool people than to convince them they have been fooled" is all too appropriate in the audiophile world.

[Julf]

But it didn't appear we would receive too many taking part in the poll. Even though anonymous, I received about as many direct responses from those not wanting to use the poll as those who voted. Plus people didn't like the two choice variety of testing I used. I suppose a 3 choice version like Julf proposed would find more acceptance though analyzing the statistics of that are messier. But I haven't been motivated to attempt that again.

 

I would definitely do it, but right now I don't have a good enough ADC available, nor do I want to have to buy an expensive audiophile cable to compare to. But more than happy to do some of the legwork with logistics and statistics if somebody can do the recording part.

[Julf]

Amplifier oscillation is a real, electrical, problem with some cable/speaker combinations-the amplifier output stage-cable-and speaker creates a circuit, and some combinations thereof can make an amplifier unstable (oscillation). This is not some audiophile myth. Many amplifiers have zobel networks across their outputs in an attempt to keep them stable even under very different load conditions ( a zobel is an RC network, like what we have in MIT cables). I am not an EE, but I am sure Julf can confirm.

 

Yes, amplifier oscillation can be a problem with badly designed amplifiers, and zobel networks are widely used on transmission systems, but properly designed amps should not rely on external components (cables) to take care of that.

 

I have no interest in doing this kind of test, as it can only prove if their is a measureable difference, it cannot prove if their is a significant sonic difference that is not measurable by this test. Additionally, a sound card is woefully inadequate for testing high end audio systems, an AP2 is needed, and few folks have access to this equipment. And, in any case, what matters is what the system sounds like. Ultimately, we are going to enjoy our systems by listening to music, and that means system changes should be evaluated by listening to music. If one hears a difference, then that difference is relevant to that listener on that system, it is as simple as that. There is no need to "prove" anything any further than that. This is why I suggest that people listen for themselves, and, if you do not trust what you hear, then there is no reason for you to pursue high end audio reproduction anyway, get another hobby.

 

So you have no scientific curiosity? You don't think proving once for all that USB cables *do* make a difference has any value?

[Julf]

I reckon for any such test to have any meaning, the cables need to be shipped around to each tester, along with the test files on a USB stick, and played on the tester's own systems.

 

What issues do you see with the test I proposed? You can still use your own amp, speakers and room (by far the most important components), and listen as many times as you want and as long as you like. Is that not enough? And if it isn't, is something you can't hear except in absolutely ideal conditions, something worth worrying about?

[Julf]

That's what's missing isn't it. I have a pretty good ADC in my Lynx L22 pro sound card. With a set of binaural mics, I can record my stereo/room and make comparisons like I am starting here: http://www.computeraudiophile.com/f8-general-forum/wanna-listen-my-stereo-12492/ Folks may be suprised at the resolution of the recording. Maybe another experiment to see if we can hear any sonic differences.

 

Appreciated the binaural files! So, could we twist your arm to help with this test, that might actually shine some light on the murky issue of USB cables?

[Julf]

But the above statement is in error. I have been paid in the past to listen test various components and prototypes, and I took this work quite seriously. I took precautions to make sure I was not "fooled", and developed a specific methodology which worked for me.

 

Sounds like we could all benefit from your methodology! Any chance of you sharing it with us? (and to avoid any cultural miscommunicatiosn - there is absolutely no sarcasm in my comment!)

 

I have noticed that many human beings seem to have a noted problem when it comes to expectations, and there seems to be a lack in the ability to operate, when necessary, outside of expectations. The ability to be objective seems to be almost a lost art in our society at large, hence we seem to have an entire culture where people often base their views and life, not on their own, individual, experiences, but based on some kind of perception of how they are viewed by others, and where they might, "fit in". It appears now that this way of living has gotten so bad, that we have people who can no longer even think, or act, for themselves, such a shame. I am not in a "camp", I report on my (considerable) experiences, not based on a "doctrine", but if and when there is a doctrine which does mesh with my experiences, I am not reluctant to agree with it.

 

I have to say that that paragraph is something I would like to print out on a poster an put it on my office wall. In courtesy to the copyright owner, I hereby acknowledge that I owe you a beverage of your choice if you ever find yourself in Amsterdam!

[Julf]

a Zobel network is designed to work optimally with a selected loudspeaker impedance.

A few amplifiers use external Zobel networks IF required.Perhaps the manufacturers of the more expensive amplifiers should provide a way for the consumer to optimise the Zobel network to suit the speakers being used ?

 

Some loudspeakers have built-in Zobel networks in their crossovers. I have not seen fancy loudspeaker cables come in "4 ohms" and "8 ohms" versions - wouldn't that be the logical thing if they really are intended to work as matched impedance transmission lines?

 

 

Julf refuses to accept the possibility of file degradation due to uploading and downloading etc. despite quite a few subjective reports to the contrary.Award winning Recording Engineer Cookie Marenco even provides her DSD downloads as Uncompressed Zips to reduce file degradation.Barry Diament even refuses to provide Lossless DLs to avoid file degradation,

and provides his high resolution material on DVD's burned to order with the user's preferred Non Lossless format.

 

I am not entirely sure how this is relevant to the topic at hand, unless it's about silly audiophile superstitions in general.

[Julf]

As far as listeining goes, I suspect that each of us may find things we can do differently, that is, what works for me, may not be the same for you.

 

Thanks, Barrows - all very good advice.

 

But a few things: blinded does not work for me, it just introduces stress which makes it harder to remain objective.

 

You have a good point about the stress, and I totally agree with it. But if you can't resort to blinded tests, how do you protect yourself from perceptual biases?

[Julf]

Audio Power Amplifier Design Handbook 5th edition by Douglas Self on pages 227 and 228, gives further information about Zobel Networks as NORMALLY fitted INSIDE amplifiers.

 

Thanks, I am familiar with zobel networks and their uses. I was pointing out that some loudspeakers use them in the crossovers to even out the impedance curve, and thus some loudspeakers already have the zobel network built in.

[Julf]

I was talking about Paul's suggestion of passing around a USB memory stick.

 

OK, so your objection to the test is that downloading the files will change their sound, while transferring them on an USB stick won't?

[Julf]

I reckon for any such test to have any meaning, the cables need to be shipped around to each tester, along with the test files on a USB stick, and played on the tester's own systems. That means for any kind of analysis to be done, the equipment of each system has to be recorded, and the environmental conditions, as well as the tester's opinions.

 

At this point I am not suggesting trying to find out how and why the audible differences happen, but simply see if there are any audible differences. I would appreciate your views on why a series of ABX tests on hi-res files recorded from the output of the DAC on a decent system, but using different USB cables, wouldn't show if there are audible differences or not.

[Julf]

Paul is simply suggesting a method that should ensure that everybody receives an identical copy of the comparison material.Even if you refuse to accept any reasons for differences, surely that method should be acceptable ?

 

The only issue with the USB stick approach is that it severely limits the number of participants, and increases the time it takes to conduct the test, just to accommodate a rather controversial position.

 

I know we have had this conversation many times over, but if two files compare bitwise identical, where can the audible differences be stored? In the fifth dimension? And as operating systems only copy the bits, wouldn't your audible differences disappear just copying the file from one disk to another?

[Julf]

"And as operating systems only copy the bits, wouldn't your audible differences disappear just copying the file from one disk to another? "

 

Not according to Martin Colloms or the authors of the TAS Report on .flac vs. .wav.

 

So where are you suggesting the audible differences hide in the file?

 

FLAC vs WAV is a different issue. What you are talking about is not a difference in file formats, but audible differences between two bit-identical files, based on their history. So kind of audiophile homeopathy, right?

 

Despite what you may wish to believe, I am not the only C.A. member who hears similar to the TAS report either.

 

No, I most definitely believe you are not the only one.

 

Or would you prefer to see a quick result that is open to criticism because the Objective side has set the rules ?

 

So if we used USB sticks, would you consider the test a valid one?

[Julf]

I am not getting into the other area again.

 

Ah, so hinting at things is OK, but actually discussing them isn't?

 

On the subject of older people still being able to hear these things

 

A subject that has nothing to do with this thread.

 

Israel is a >80 year old ex U.S. A Professor of Music. He details differences he hears with a recabled headphone using Cardas cable. According to presently accepted hearing dogma, and according to you guys, what he reports should be impossible, shouldn't it ?

 

Uh, why? And what does this have to do with anything?

 

I believe that the use of the USB stick as Paul has suggested, has the best chance of obtaining meaninful results provided that the equipment used to generate these differences is suiutable.

 

OK! Thanks!

[Julf]

I suggest that you use the Search facility, as if I go there again, Chris may have my scrotum for a " Beanie"!!!

 

I assume he has good reasons for that...

[Julf]

Not only is he reporting things that present hearing dogma would suggest is impossible due to sharp hearing decline with age

 

I don't think there is any "dogma" that states that hearing high frequencies is impossible at an advanced age. I think the generally accepted fact is that on the average, most people do seem to have some high frequency hearing loss in their later years.

 

but he reports hearing differences between 6' long headphone cables that Dennis and others seem to believe is not possible on technical grounds.

 

So you consider one person reporting subjectively hearing a difference objective proof that there are real, audible differences?

[Julf]

When you say you down want it downloaded, it that because you are saying bits can get lost? Or are you saying something else may get lost (something the EE's haven't the tools to measure yet?)

 

As sandyk seems unable to address direct, factual questions because Chris supposedly won't allow the discussion (while seemingly being OK at *hinting* at the things),. let me state what I have understood sandyk's position to be - hopefully he will correct me if I get it wrong.

 

Basically he is saying that there is some audible difference, stemming from the past history of the data, that bit-wise file comparison programs can't detect (so clearly it isn't bits getting lost, but the content being somehow different despite every bit being the same).

 

This audible difference is supposedly caused by downloading/copying files over the Internet. Packing the files in a zip file supposedly helps protect the files against this corruption.

 

On the other hand, sandyk has also claimed to have files that, while being bitwise identical, sound different, because the content, a bit-perfect rip of a CD, was in one case ripped using a standard power supply and in the other case using a custom power supply - this despite both files comparing the same using bit-for-bit comparison programs. This was "independently verified" by having other people listen to the files after the files had been either downloaded or emailed to them, so clearly the transfer over the net didn't obliterate the differences.

[Julf]

Now some of you you even want a thread of your own where the Objective side can discuss these things free of annoying replies from the other side.

 

Yes. Outrageous, isn't it!

 

You can just about bet your left one, that if the subjective side tried to do the same ,it would be met with howls of protests and extremely sarcastic replies, as is almost always the case where there are any kinds of subjective reports,no matter what the subject, and the thread would end up getting locked by Admin. This is a regular occurrence on MANY different forums.

 

I wouldn't bet on that one - there has been several cases on CA where the OP has asked for purely subjective impressions and no technical discussion, and in general that wish has been respected.

[Julf]

... a smokescreen for not wanting to do the real work, IMO.

 

Indeed. It's easy and cheap to throw around words about why a test is "not good enough", instead of actually spending the effort to run a test that can provide some valuable information, even if it isn't perfect.

 

Pitchforks and torches went out of favour in Europe once we sent our narrow-minded religious fanatics over to the (former) colonies.

[Julf]

I suggest that for a different reason than you assume - precisely the reason that Alex stated. It eliminates any possibility of a tester listening to the wrong copy of the get files. For example, a MP3 copy of the file they have on their system.

 

I don't think that was the concern Alex had - he was clearly concerned about the contents getting corrupted during a network transfer. Somehow I don't see how transfer using an USB stick would reduce the risk of confusion compared to a download, especially is the files are distinctly named and tagged.

[Julf]

I can see you don't see that, because you are not thinking it through.

 

Ah, thank you for pointing out what I was doing wrong.

 

I must have been confused by Alex writing:

 

As per what Cookie Marenco has found , the files on the USB memory stick would be best saved to it as Uncompressed Zips.

Yes, there is still a smalll degradation, but not as bad as going through several servers and perhaps 100s of optical repeaters.

Not that I expect you to accept any of this ! Paul is simply suggesting a method that should ensure that everybody receives an identical copy of the comparison material.Even if you refuse to accept any reasons for differences, surely that method should be acceptable ?

Somehow I took that to imply that the issue was that the files would be degraded less using an USB stick compared to "going through several servers and perhaps 100s of optical repeaters", and that the distribution method you suggested would be better as it would ensure the files would be identical instead of corrupted by the net.

 

I am glad you are so much better at understanding what Alex meant, and corrected my mistake. Much appreciated.

 

Just passing around files to tet cables is a useless test, and would produce meaningless results. Won't stand up to even the most trivial peer review.

 

Are you sure you aren't secretly an European naysayer?

[Julf]

And even forward thinking liberal European leading England was convicting people of witchcraft in the 1940s. Didn't repeal the 18th century witchcraft laws until 1951 in fact.

 

Uh huh...

 

Indeed. On the other hand...

 

"Jane Wenham was among the last subjects of a typical witch trial in England in 1712, but was pardoned after her conviction and set free."

 

"The last execution of a witch in the Dutch Republic was probably in 1613. In Denmark this took place in 1693 with the execution of Anna Palles. In other parts of Europe, the practice died down later. In France the last person to be executed for witch craft was Louis Debaraz in 1745.In Germany the last death sentence was that of Anna Schwegelin in Kempten in 1775 (although not carried out). The last known official witch-trial was the Doruchow witch trial in Poland in 1783."

 

On the other hand, it seems the last documented witch trial in the US was the 1878 Ipswich Witchcraft Case in Massachusetts.

 

I think all we can conclude is that history is written by those who write history, and that no one has a monopoly on Chauvinism (in its original and primary meaning, "an exaggerated, bellicose patriotism and a belief in national superiority and glory").

[Julf]

Guess he was right. Either that or you simply choose to misinterpret.

 

Or maybe I simply looked at the whole paragraph instead of just cutting out one half?