Ethernet Madness!

a 12m run of AudioQuest's Diamond Ethernet cable will run you $7,995

I think it started on Reddit then spread like wildfire to a number of websites including Ars Technica, The Register, HotHardware, and more. Not to mention the forums. What's all the fuss? Expensive Ethernet cables. $10,000 Ethernet cables (actually the price is $8,000 and all of these sites got it wrong).

I reviewed four of the AudioQuest Ethernet cables back in 2013 but a few quotes and a link in the above-mentioned articles has helped bring new packs of well-credentialed network engineers and others to AudioStream to call bullshit (thanks for traffic hit btw). Their point being there's absolutely no way an Ethernet cable can effect sound. Data is either being transmitted or its not. There is no in between. "Bits are bits, after all." With that in mind, the majority of these same people cry foul and suggest the only possibile explanation for perceiving a difference between Ethernet cables is a) rooted in psychological interference like confirmation bias or placebo effect, b) I am a paid shill, or c) I am insane. I'm personally going with d) none of the above and will stubbornly stick to my guns that the differences I've experienced are in fact audible differences. I'll leave it to more knowledgeable people than me to figure out why.

One such person/commentor, CG, who appears to know his way around this issue more than your average network engineer, has offered up another possibility—noise. You can read more about this thought-provoking possibility and the rest of this sordid affair in the comments of my AudioQuest Vodka and Diamond Ethernet Cable Review.

Since I wrote this review, I've had an opportunity to hear more comparisons between the AudioQuest Ethernet cables including their entry level Pearl ($25/.75m). The most notable demo took place in Denver at RMAF 2014 (see my report) where AudioQuest's Steve Silberman took us through a wireless connection, a wired Ethernet connection using a standard Belkin cable, and then the AudioQuest Pearl. In each instance, with each change, the sound improved. I noted "increased clarity and resolution and generally a more musical sounding system."

This demo was performed to a room full, standing room only, crowd. It was clear from the crowd's reaction that I wasn't the only one hearing these changes. John Darko of Digital Audio Review also wrote about this same demo which you can read about right here. There are many other reports around the web of people hearing a difference between Ethernet cables but that's hardly the point. Right?

So what's going on with Ethernet cables that's got everyone in an uproar? Clearly, $10,000 is to most people a ridiculous amount of money to spend on most anything you can't at least drive. But as difficult as it may be to swallow, there are people for whom a $10,000 investment in their hi-fi, like an Ethernet cable, doesn't even register as a blip on their financial statement. Is this wrong? Is it amoral? Is it plain stupid?

The possibility of different Ethernet cables causing an audible difference aside, I suppose we have to ask is there an amount above which it is wrong to spend on a hobby. Hi-fi is, after all, a hobby. It is not a science experiment or a race toward "good enough". Last time I checked, people spend lots of money on lots of things that other people feel don't make any sense including fine food, fashion, art, cars, yachts, jewelry, cigars, spirits, music, antiques, Pez dispensers, sweet sixteen parties, and more so obviously people are free to spend their own money on what they choose (and the Internet was clearly invented to give voice to every opinion on this very subject).

Does anyone need a $10,000 Ethernet cable or a $100,000 wrist watch (especially when your cell phone tells you what time it is)? Of course not. But are we then saying we need to regulate how much people spend on their hobbies? Of course not. We are in fact stuck between a rock and a very expensive place. These things exist and people actually buy and enjoy them. Does that fact somehow tarnish every hi-fi hobbyist and every wrist watch wearer? Of course not. There are extremes in every hobby and the farther out you go, the stranger things get to the uninitiated. Ever seen a furry convention?

As far as I'm concerned, if you use Ethernet in your hi-fi, your opinion regarding Ethernet cable is best informed by listening. Do it blind, do it naked, do it dancing, do it however you like. AudioQuest's Ethernet cables start at $25 and you can return them if you find they make no difference. Or don't try them, I don't really care.

When all is said and done, if I stopped listening to all of the things that people who know better believe make no difference, I'd be relegated to listening to MP3s on my iPhone with a "decent" pair of headphones.

stevebythebay's picture

When prices for interconnect get this high should a person consider an alternative to a NAS? How about another connection technology like Thunderbolt? Or for a device capable of multi-terabyte internal storage on a well established BUS technology? Guess it may eventually come down to a company or two building purpose-built music players that achieve the optimal delivery of bits without anything but the bits, in time, etc. Of course there will still be a connection to a DAC. And we all understand that each DAC seems to "like" one form of connection over another, be it USB, AES/EBU, etc. So it goes...

Michael Lavorgna's picture
System building considerations certainly apply and if the notion that Ethernet introduces noise into the audio system is correct, one way to avoid this scenario is to keep the network out of the hi-fi.
rompolompo's picture

With CAT5 and CAT6 six shielded cables there should be no noise getting out or in. The voltage and current running on fast ethernet is around 1.5 volts. If someone hears a difference within ethernet cables then good for him but a scientific paper is required to show why.

vindaloosoup's picture

I find it a little odd that manufacturers of high quality ethernet cables do not publish much information and tend to rely on reviewers to charter these waters alone. Michael is probably doing the best job right now.

I wonder if there is a legal reason they cannot claim improvements in sound, and rely on the industry to do that for them?

Stephen Mejias's picture

Quote: I find it a little odd that manufacturers of high quality ethernet cables do not publish much information and tend to rely on reviewers to charter these waters alone.

I'm not sure what you mean. Would you please explain? What kind of information would you like to see published? Michael Lavorgna referenced several articles that explicitly criticized the information that we have published.

I was surprised and saddened to find that when I made a few attempts to provide those other publications with additional information, or even to merely thank them for sharing their thoughts, my comments were deleted. As many readers of AudioStream know, I spent nearly 14 years working for Stereophile before joining AudioQuest last March as VP, Communications. I have great appreciation, admiration, and respect for brave, honest, thorough journalism. And in my new role with AudioQuest, these feelings have only grown stronger.

I have such deep appreciation for any journalist -- and anyone, in general -- who is willing to stand by their beliefs, to publish those beliefs for others to read, question, ponder.

In light of all that, I am similarly deeply proud to be a part of a company that stands by its own beliefs. AudioQuest takes great pride in exploring possibilities, on questioning, on discussing, on sharing, on educating. On our website, I think you'll find that we've published a great deal of information -- not only on our cables, but on audio, in general. And we're always working on more.

We thank Michael and all of AudioStream's readers for their comments. Throughout what has been an otherwise unpleasant experience, it's been really wonderful to see the thoughtful, respectful, insightful comments here. Thank you! As ML says, listen blind, listen naked, listen while dancing, or however. Just listen.

Happy listening! (And Happy Valentine's Day.)

Stephen Mejias's picture

I'm sorry -- I forgot to sign my post.

Stephen Mejias
AudioQuest, VP Communications

CG's picture

Journalism and the Internet aren't always entirely synonymous.

vindaloosoup's picture


Your website contains ZERO documentation for the Audioquest Ethernet line of cables. Nothing. Zilch. Nor does Audioquest make *ANY* claim to improved sound when compared to other cables meeting specification. My hypothesis is you can't - most probably because there is no difference. If you did make such claims, I put it to you Audoquest would be at odds with consumer law for false advertising.

Instead, it appears that Audioquest make good use of paid reviewers to peddle the claims that Audioquest legally can't. Whether they truly believe it, are riding the gravy train, or a making it all up is anyones guess. But it appears to the educated to be a bit of a racket.

I have no issue in someone selling high quality cable, but to venture to say "widening of the sound stage" and "increased perceived clarity" and other such nonsense reeks of wine taster talk and starts to resemble more of a religious belief.

I have all but one question Stephen - why does Audioquest choose NOT to make any claims in their advertising or on their website that makes direct reference or claim to improved sound quality, even between the different cables in the same ethernet product line?

Stephen Mejias's picture

Hi vindaloosoup.
Before I address your other points, I'll begin with what I think is your main point -- that we do not publish information regarding our Ethernet cables. I'm still not sure what you mean by that. Have you seen this page of our website?

You can click through all of the models of our prepared Ethernet cable: Pearl, Forest, Cinnamon, Vodka, and Diamond.

Or have you already seen these pages, but decided that we haven't published the specific information that you would like to see?

Stephen Mejias
AudioQuest, VP Communications

qsysopr's picture

The Transmission Control Protocol (TCP) sends data packets between two or more components. I wonder, why the AQ pearl and other do have a "direction". Transmission is made reliable via the use of sequence numbers and acknowledgments. I doubt seriously whether a CAT6 cable can be superceeded by one cable with "direction" in a user's home network. If I am wrong, please let me know.

highstream's picture

If the reason(s) is not known and thus there aren't scientific papers, but differences can be heard, does that mean the latter aren't real? This comment applies to all sorts of phenomenon in the world, from virtually all of audio to steroids and other performance enhancers for athletes. It's interesting that people question the former all the time on forums and in comments sections, but readily accept the latter as real w/o scientific studies showing why. As for ethernet, I have heard differences between OEM, cable company, Blue Jeans, Audioquest and Meichord ethernet cords with PS Audio PWD and Directstream dacs.

sg60's picture

The rich can spend their money any way they want but do we seriously have to waste so many inches of columns/blogs on products that only cater to this 1% of the 1%? Almost every audiophile magazine or blog is guilty of allocating a totally disproportionate amount of space to products that only a tiny handful of people can actually afford.

Michael Lavorgna's picture'll find a balanced mix of products in all price ranges. I'll also reiterate that the AudioQuest Ethernet cables start at $25.

The reason these non-audio websites are talking about a $10,000 Ethernet cable (and they a) got the price wrong, and b) why talk about a 12m cable?) is because it's an easy, cheap, and relatively mindless way of drawing traffic.

ianfollett's picture

Will you agree that confirmation bias is a possibility and that the way to avoid it is blind testing? Would you be willing (hypothetically) to do a blind test? And if so would you accept the outcome of that test?

I understand and in some ways agree with your argument for the value of subjective experience and I don't accuse you of having ulterior motives. But if you're not willing to do a blind test, isn't it more fair to say that knowing your data is traveling over these cables is more enjoyable than knowing its traveling over off the shelf ethernet cables?

Michael Lavorgna's picture
Let's begin here. Show me ten blind tests that relate specifically to audio that have proven, without question, anything. It would help if some number of these test results were shown to prove what had heretofore been viewed as a subjective impression.

The reason I ask is because I've seen "blind test" thrown around so often and more times than not, it's just pseudo-scientific puffery. For example, in the comments of the review I linked to above, someone challenged me to take a blind test with the added provision that we would bet $2000 on the outcome. Show me where in the scientific literature, betting on the outcome of a blind test is recognized as a valid approach.

But if you're not willing to do a blind test, isn't it more fair to say that knowing your data is traveling over these cables is more enjoyable than knowing its traveling over off the shelf ethernet cables?
I see no reason to equivocate. Anyone who reads my review is very well aware of how I arrived at its conclusion - through subjective sighted listening. I have also, since the time of that review, had the opportunity to compare AudioQuest Ethernet cable to standard Ethernet cable and the results were the same. So, not only did I initially spend a few months working through the review process, I have since had other experiences that support my initial findings.
ianfollett's picture

"Without question" is a very high standard. Even people (like myself and the rest of the scientific community) who consider double blind testing the gold standard for experimental evidence (audio experiments included) will tell you you can't prove anything with absolute certainty. Instead you can prove that an assertion is very unlikely to be false. With that said, there have been many experiments with audio done which have produced very strong evidence for all sorts of things. I’m guessing your request that I show you examples of these experiments was rhetorical but if you are genuinely interested in the subject, is a good place to start.

Your assertion that you have made your testing procedures abundantly clear is fair. I’m not accusing you of making false claims. I (and I think other readers) would like to know if you actually believe that you can tell the difference between the “premium” ethernet cables you have reviewed and a standard CAT5 cable without being told beforehand. If you can, a blind test could prove that. Not “without question,” but to a near statistical certainty.

Of course you don't have an obligation to do a blind test. But, hypothetically, if you did one and the results were negative, would you accept that you couldn’t actually tell the difference? If your answer is no, then you are rejecting the scientific method. Clearly that’s your right. But it’s a very controversial point of view which defies logic and I think your readers would like to know you hold that bias.

Michael Lavorgna's picture
...politeness. It goes a long way in making this a real conversation.

For me, a blind test would possibly show that I may be able to discern a difference under a very limited set of circumstances (the test setup). It seems to me, this test is about proving that I cannot hear a difference rather than offering anything of substance about Ethernet cables effects on audio. This type of test holds very little interest for me.

I am aware of Hydrogen Audio. I've visited a number of times and I have to say, with all due respect, that I find that place to be the death of joy. All that means is that type of discourse is of no interest to me and I fail to see how it furthers our enjoyment of music.

hifiguy24's picture

You hit the nail on the head. Well done. More civil than I can usually manage reading this site that I swear daily I will stop reading.

I've realized I read these ridiculous hi fi sites for the same reason people watch soap operas. It's such a shitty and overwrought drama that I just can't avert my eyes.

rompolompo's picture

AudioQuest designs their products based on some type of science. They use measurement tools to capture, track and fine tune their research. Whatever method they choose, it should be traceable from first sketch to the final product. As such, they should know how changing the design will affect the final product. There is no magic or voodoo in the design. As a consumer, one would like to know what he is paying for and not just let to be told that there is a difference. This applies to cars, watches, pens, etc.. etc... My point is that cables were designed and engineered by known methodologies and using A/B/X test is a proper way to qualify the designers final result. This is done all the time in many scientific circles. The fact that Stereophile refuses to even consider this and frequently tries to diminish this as irrelevant does not bode well to its credibility.

Michael Lavorgna's picture
How is it we're talking about Stereophile on AudioStream?
deckeda's picture

As I tried to say in some comments on Ars about these cables, anything high-end related has no place being published there. It was the same with their Pono review, they stated upfront their intentions to mock and express grave caution and "save" us from ourselves.

These aren't legit reviews and the tech writers aren't qualified to write about them as such. Whether or not expensive and outrageous and unreasonable labels should be assigned is therefore irrelevant.

But forget all the bling that not even a noticeable fraction of audiophiles would buy. These tech sites seriously harm music reproduction with their wide swaths of condemnation for audiophiles.

Would they run a piece about great-sounding budget speakers for example? No, because they aren't in a position to recognize what those would be. It's preposterous to think they are remotely able to understand what a super-niche product would be for.

But because unlike other expensive endeavors, "it's electrical" they feel completely and totally qualified.

Don't lets get lost in objective vs subjective and blind test rabbit holes here. The questions aren't worth asking. They don't apply.

Michael Lavorgna's picture
Never. For the simple reason that it wouldn't have any draw. Dissing Neil Young and expensive audiophile stuff are easy, large, and emotionally charged targets. Just look at the comments on those sites.
Deli's picture

Beyond stereos, another one of my interests is cars. When reading those publications they’re able to provide objective comparison facts such as acceleration times, braking distances, and cornering abilities. They can prove a Mustang performs better than a Focus in those areas.
In audio, objective facts are provided with speakers and amplifiers in the way of frequency responses and distortion levels; however, I don’t remember ever reading any performance facts about cables beyond what they are made of and what someone “hears.”
If these same basic journalistic qualities can not be performed on cables, then maybe they don’t deserve a space in your pages because you can’t prove anything. And, to ask consumers to just try it for themselves seems like a lazy copout. You’ll quickly find yourself on a slippery slope with a short rope.
There is a place for subjective opinions in reviews. I would never make a car buying decision based solely on stats; I would want to test drive first. The fact that one has a few tenths of a second on the other to 60mph wouldn’t be my deciding factor and is a great opportunity to consider the reviewer’s subjective opinion.
Imagine a sports car review where the author picks his favorite, but can’t give you any factual reason why he liked one more than the others. I don’t want to test drive every car out there; give me some comparison facts so I can narrow down my choices and then test drive the top two or three. I want some verifiable performance differences between my repurposed printer USB and a boutique one!
I’m an English teacher, and if I student turns in a paper with just their unsubstantiated opinion, they’re not going to pass. That doesn’t cut it! If that is the best audio journalists can do when reviewing cables, that is pretty sad.
Oh, and Michael, why when asked about running “a piece about great-sounding budget speakers" did you feel it “wouldn't have any draw?” Maybe I am in the minority, but I would love to read about budget speakers and how closely you can get to champagne tastes on a beer budget. I fear there is a massive case of diminishing returns in hi-fi, so I’d love to see if I am right. Anybody else curious? I’d hate to think your hesitance is for the sake of advertising revenue.

Michael Lavorgna's picture
"...seems like a lazy copout."
I disagree and recommend that everyone should listen before they buy any piece of audio gear, preferably in their own home/system.
Oh, and Michael, why when asked about running “a piece about great-sounding budget speakers" did you feel it “wouldn't have any draw?”
You misunderstood what I wrote. I was responding to a comment by Deckeda wherein he asked "Would they run a piece about great-sounding budget speakers for example?" They being the sites I mentioned in the beginning of this article, not AudioStream. I have reviewed a number of powered desktop speakers. I do not review speakers for in-room listening since this is the domain of our sister publication Stereophile.
stevebythebay's picture

This discussion in many ways runs along a similar line to debates about Strads and modern fiddles. Blind taste done by seasoned players tend to put the modern instruments ahead of the classics. I'll leave it to each person's ears to decide what suits their tastes, despite the outlandish costs in either classic instruments or audio systems.

kvh's picture

"When all is said and done, if I stopped listening to all of the things that people who know better believe make no difference, I'd be relegated to listening to MP3s on my iPhone with a "decent" pair of headphones."

Nooooo kidding (imagine every aspect of your life optimized away in this manner; what would that be like?) Thanks for writing about this because I've been simmering away; so much audiophile bashing going on... it's so easy, and so lazy. I'd love to see Ars visit a cable manufacturer, listen to stuff, and talk to people, and write that up. THEN bash it all they want, if they reach the same conclusions. But this kind of armchair trashing involving an interest the author is clearly not into is hardly what I expect from an Ars Technica article. The world is full of foolish things that some people think other people spend too much money on (this same Ars writer reviewed and purchased a $450 joystick last year).

I just don't get the hostility toward discovering ways to improve the experience of listening to music. Ways that are affordable, enjoyable, interesting, even fun. And that this exploration doesn't automatically involve becoming a depraved maniac who lusts after $10,000 ethernet cables.

The smallness of thinking we know it all... Give me beauty, mystery and magic over the smug certainty of what we think we know, every time.


Michael Lavorgna's picture
Good for him! I hope he enjoys it.
The smallness of thinking we know it all... Give me beauty, mystery and magic over the smug certainty of what we think we know, every time.
Great post Matt.


vindaloosoup's picture

Interesting article Michael. I've been considering upgrading my ethernet cables for a while now, but my understanding of ethernet is very low!

I'm interested in the Audioquest line, but they don't appear to make any claims in their advertisements or website on any effects their product may have on the quality of the reproduced sound. They focus on build quality and construction, but not the sound. I don't mind paying extra for a nice quality product.

Like I said, a very interesting subject. I'd like to see more information provided by Audioquest. I'm sure they have the answers most of us are asking but they remain mute.

It's up to professional reviewers such as yourself to try and figure this one out.

Keep plugging away, I enjoy your commentary.

dchatenay's picture

The least I would expect from a journalist when they're comparing equipment -especially cables- is at least a blind test (double-blind, even better). It would eliminate the chance of cognitive bias. It's not hard to do, it only requires another person to swap equipment, keeping it hidden from view during the test, and revealing the equipment at the end of the test.
Doing so would make me more likely to trust the test results.

Michael, we trust your opinion, and those of other Audiostream/Stereophile writers. You were immensely helpful to educate me -and I suspect legions of audiophiles, by sharing your knowledge. On this one though, I don't trust your opinion. I understand how Ethernet works, how data is transferred between two machines. The only explanation I can think of to a difference in sound is that the digital player used is sensitive to variations in network packet timing, which means its network and DAC code stack are poorly isolated/integrated (bad coding). Which means the digital player is terrible, not that the network cable is better. It's possible. But it's also very unlikely. And without blind-testing different combinations of audio cable and digital players though, there is no way to know for sure.
So I will stick to a basic network cable...

Michael Lavorgna's picture
I completely understand your point but would suggest the one way to know for sure is try some different cables for yourself. I also understand if this is of no interest to you.
blaven's picture

Michael, I am curious if a better ethernet cable would affect sound for something like an Aurender S10/W20 server that has local based music storage. In such a case, the ethernet cable only functions to transfer files to the internal hard drive and allowing one to use an iPad remote, but otherwise is not carrying any data on a moment to moment basis.

I am not sure it is totally analogous, but I have seen reviews that people have noted a difference in Firewire and USB cables carrying data from external hardrives to computer, that later passes that signal out on a separate USB cable, via separate bus to the DAC. I believe that this was attributed to noise and possibly field affects. I believe it was noted even when that data stream from the external drive was not active and the computer was driving the DAC directly (but with the Firewire cable still attached to the external hard drive).

Thanks for your thoughts

Michael Lavorgna's picture easy test would be to disconnect the Ethernet cable from the Aurender and see if you notice a difference.
BradleyP's picture

With a spread in price as we see among AQ ethernet cables, I see some possibilities.

1. Their engineering does measurable things under certain circumstances, and it does more of those things as you go up the price scale. In that case, AQ must know what is going on. That they don't tip their hand is clearly a strategic decision. I would be happy if they came clean about what is actually going on in the cables. Any other respectable company in any other industry would be very up front about their engineering and specs.

2. It's possible that the actual cost of materials and R&D is not much different among the cables, but their best cables are skimming the market.

3. It's possible that AQ has tried different materials and geometries and gotten different results but has no idea why.

4. It's all BS.

bobflood's picture

everything scientifically and justify everything based on some lowest common denominator just sucks the life and joy out of this or any other hobby. I respect science very highly but this is music and by its very definition music is not about science; it is about emotion and pleasure.

If this line of reasoning keeps up, soon we will have robots playing the instruments because they can be scientifically proven to have better timing and accuracy. In a scary sort of way we are getting close to that with the way studio music is created today.

Let's keep all the science talk limited to true scientific subjects like the big bang theory or evolution.

If you don't think that any product has a value proposition for you then don't buy it, but stay out of the purchase decisions of others. Unless of course, you are OK with them telling you what you should or should not buy.

24bitbob's picture

Hi bobflood,

I'm sure you've heard of 'snake oil'. The HiFi business is drowning in it.

On science (engineering I'd call it), it's not hard to find evidence where the choice of materials, component choice, layout, circuit design, etc. all influence the quality of the sound reproduced. Most of that design effort depends on verifiable, objective criteria. The role of engineering in audio reproduction cannot be overstated, and indeed it's acknowledged in sister publications such as Stereophile and InnerFidelity. Without engineering, the HiFi business is dead.

Which are the verifiable, objective criteria on which cables are designed? Or are we supposed to accept that science only matters up to a point?

I suspect that cable design emanates from the marketing department more than it does the test lab. I'd love AQ, or any other cable manufacturer to disabuse me of that idea. If they do, I'll part with my $25 or $200, depending on the information offered. I've bought 3 different USB cables and can't tell the difference between them. I'm a $400 digital cable skeptic.

I tried, though.

bobflood's picture

with your comment about engineering. These products have been "engineered". This whole argument is always about something entirely different though. Those that make these arguments use the scientific method to try to discredit the engineering because they don't like the price.

What they all miss is the scale at which our hobby runs. There is no mass production of any of these products because there is not enough demand to scale up and reduce production costs and spread the cost of the engineering over a high unit sales volume.

This is not going to change. Listening to music is barely a hobby at all anymore. We are never going to get a large number of people interested in doing what needs to be done to get high level music production into their lives so we will always have to deal with high prices.

My main point really was that a purchase decision should remain personal. If someone is swayed by marketing and feels that the product offers real value to them, it is their choice to make.

24bitbob's picture

Thank you. I understand the point you make, and don't dissent.

Beyond that though, if it hasn't already reached that point, a segment of the HiFi industry is stretching credibility, to the point of being mocked and derided.

At a time when there is a push to encourage wider acceptance of at least lossless music, if not, 24/96, credibility matters. I'm not sure that the questionable touting of expensive Ethernet cables serves that end*.

That said, Galileo and Leonardo Da Vinci were derided in their day. Maybe science will come to the rescue and explain exactly why this or that sounds better.


ktracho's picture

I remember once listening to orchestral music played through my $80 DAC (which I still use today) on my $600 tube amp connected to $200 PSB speakers about 7 years ago, and thinking "Boy, there's something wrong with how this sounds. I don't remember it sounding this bad. Is something malfunctioning?" I started looking around, and I discovered I was using $30 Monster Cable interconnects (marketed as high-def cables particularly appropriate for CD players). I had totally forgotten I had switched out the $100 AudioQuest cables I had been using before. I can assure you I did not need any scientific explanation to explain why it sounded much better after switching back to AQ cables. In fact, I would dare you to find any scientific measurement that can explain the difference in sound. I'm not saying a $8K or $10K ethernet cable is worth it, but based on my experience, I also can't dismiss that there could be a difference, whether or not there is a scientific explanation for it, or whether or not the manufacturer chooses to disclose his secret sauce so his/her competitors can copy his techniques.

Patrick Butler's picture

For anybody who has sold audio for longer than thirty seconds, convincing anybody to part with thousands of dollars for anything takes real work. Trying to convince someone to buy something that they are already skeptical (or hostile) towards puts you at a considerable disadvantage. We conveniently ignore this form of cognitive bias when we flippantly use the term to explain why folks claim to have heard things for which there isn't already a canned scientific explanation.

Got news for you- the folks with the kind discretionary income to spend $10k on a cable are not idiots, and to a man there are many things they'd rather be spending their money on (rare scotch, a new watch, Russian call girls.) They didn't get this wealthy by being idiots, and every one of them that I've ever sold to start by being skeptical- that's the default state.

They end up buying (or not buying) products because talented salespeople who knows their trade, setup demonstrations where customers can make up their own, "biased-against-buying-a-damn-thing" minds. People don't buy because "they think they heard something under the influence of a Jedi mind-trick" or other namby-pamby explanations. People buy because they had a "holy shit!?" moment that defied their skeptical minds.

If you can't scientifically explain why it happened, maybe it's time to strap on a pair and start doing some research. That's science. Who knows what you'll discover.

Wavelength's picture


Look originally we thought SPDIF cables could not make a difference, proven wrong. Then we moved to USB cables, "no USB cables can't sound different", again proven wrong.

Why don't you think Ethernet can't make a difference? I can think of many reasons why it does.

For that matter, doesn't everything make a difference in some way?

Sure these are expensive, but I found even the inexpensive Forest Ethernet to make a difference. We use them at shows now for connecting basic stuff. We don't even use NAS in the system and it makes a difference.


vindaloosoup's picture

Hi Gordon, would you be able to point me to any credible source where it was demonstrated SPDIF and USB cables all meeting specification make a difference? Genuinely interested.

Wavelength's picture


Stereophile I am sure has tested SPDIF cables in the past. Originally when SPDIF appeared by a mistake in the beginning by an engineer at Philips that used the SPDIF and two RCA's as a testing method for sealed box CD players, it was never meant for usage. Later IEEE adopted the it and companies started making SPDIF cables. The problem was that companies did not realize what SPDIF was and how to optimize it. With SPDIF you are dealing with 2 basic things, clock recovery and data. Sure the damn thing is suppose to be a 75 ohm system, but an RCA can never be 75 ohms so let's just forget that fact and move over to the other 2. Clock recovery has a big effect on the receiver's ability to create a Master Clock. On a CD player (just like Adaptive USB) the Master Clock is used as a flow control for the data being sent by the transport. Fixed clocks such as ones used in asynchronous USB->SPDIF converters supply the SPDIF receiver with a nice fixed clock. But transports don't do that and the cable therefore due to capacitance, shielding and so forth can have an effect on this derived Master Clock. Then of course there is the data stream, also very import as the protocol NRZ can be one that is hard to interpret at higher sample rates.

USB came in I think as a shocker to the digital world when I made the Cosecant 11 years ago. At first people thought yea fad, heck I did also. Then it just took off and look at it today. The USB cable companies first cables were derived from their information on SPDIF cables. BUT!!! USB is very different, it is bi-directional. Last year I bought a Tektronix's USB compliance testing system. Before that I really only used USB protocol analyzers. But this system really shows cable and interface differences like no other.

Isosynchronous USB is the main protocol for audio. There is no error correction here. What my testing on cables has shown is that many of these so called ultra $$ cables are actually not as good as $10 Belkin cables. With Asynchronous comes what is called a feedback pipe. This pipe sends flow control information back to the computer so that the flow rate of data stays consistent enough so the DAC does not have pops and clicks (over or underrun of available samples). Some of these expensive cables on the TEK system shows ERRORS on these feedback pipes which means the computer will toss these packets and the flow control can be compromised.

USB cables also have power and ground and these can be effected by cables. As well as computer noise... Some cables because of their shielding or not shielding of the cables pushes the computer noise back into the computer. Were as some cables push the computer noise to the dac and that can have a more costly effect on the DAC's output.

Ethernet can be considered it's own problem, with it's extended length and speed as well as dual pairs of data. All the things associated with the IP protocol, routers, bridges and other computers and devices on your local net can have an effect on the over all sound of your system.

Say you have a NAS driven system, then the DLNA controller tells the renderer where to get the song the renderer has to go to the NAS and get the track data. Well along the way your computer and other stuff is also hopping on an off the link. Cables here can make a huge difference, especially if you are mixing 10/100 and 1G devices on the same local net.


CG's picture

Not to mention that at the digital transmission frequencies, the cables and other interconnects are bona fide transmission lines. If the impedance at the transmit and receive ends don't match each other and the cable impedance, there will be reflections. This can and often will compromise the digital signal integrity. A problem with any sort of cable is trying to keep the impedance constant over the entire length, just due to manufacturing variations. Connectors can be a challenge, too. It isn't so easy.

A subtlety associated with this is that a digital signal ideally is a square wave or approximation of same. That means that the spectrum really is that of a sine wave at its fundamental plus all the odd harmonics up through maybe the 7th or 9th or so. So, these components also are subject to reflections. That's also presuming that the signal is good approximation of a square wave, which often it isn't. This all adds additional challenges. (Lots of high performance communications systems often distribute sinusoidal clocks so that the signal can be more easily filtered, and you only need to deal with a single frequency with regard to reflections, which cause jitter in the time domain.)

I look at all this very simply. Ethernet and other high speed transmission protocols operate at less than 100% efficiency by design, so that encoding can be used as well as retransmission, if necessary. That implies that the guys who designed the system expected a noisy environment. Otherwise, why go through all that crap? The noise is electrical noise. This same noise doesn't magically say to itself, "Oh look - here's the digital receiver. We're just going to stop now and vanish." So, since we desire to convert these bits into an analog signal at some point have to do what we can to prevent any noise from the digital transmission segment from corrupting the analog signal. Simple as that. (When I say noise, I mean in the broad sense - any undesired signals. That goes way beyond hiss from your loudspeakers.)

dallasjustice's picture
CG's picture

Nice argument, counselor.

However... What you left out is that John Swenson has made extensive measurements of what he describes. That's not so easy to do and it takes test equipment not found on most people's test bench. That he has not published all the details and all the results does not mean they don't exist or are just balogna. Honestly, how many readers of an audio enthusiast forum could reasonably be expected to successfully interpret them? I wouldn't think this is a fair expectation.

What puzzles me is that there often is an absolute demand by a large segment of the audience for rigorous engineering information and data. Yet, when such is presented, it's immediately dismissed. Just why is this?

The subject of how digital noise can affect analog performance in a mixed signal environment is hardly new. It goes back a couple decades. A more recent paper, which just happened to be the first that came up in a Google search can be found here >

Audio applications in many ways are more challenging that many other mixed signal processes since the bandwidth is relatively high comparatively and the dynamic range - highest level to noise floor, with spurious tones counting as noise - is pretty high as well. Tones and such that can be filtered out in other applications through processing are processed by the listener's brain, which often tends to be unforgiving.

But, you have found an answer that works for you. That should be all anybody looks for. Bravo! Everybody should be looking for their own answer.

MrkiMile's picture

Proven wrong? USB cables? How, where? :)

Come on lads, you sound like those people from Penn and Tellers' "Water Tasting" show (

It is not possible. Place super sensitive microphones connected to expensive oscilloscopes and MEASURE the difference between differen USB/Ethernet/HDMI/whatnot cables.

Sadly, I'm surprised that you still can't buy $100k super-high-quality-audio home routers as they 'route separate frequencies trough separate digital channels to create extraordinary sensory experience! (It will make bats tingle!). - you're so much fun! :)

Wavelength's picture


Really microphones? What we are testing does not necessarily happen in the audio realm, but the effects of these anomalies does effect the device it is connected too.

I have tested cables that are over $2K that cause dacs to pop and click because they don't pass the feedback back to the computer correctly. Remember USB is a poll select protocol. When the computer selects a device to transmit it must do so in an appropriate time frame. Some cables do not work well in this regard because of what I call overhang. The select or what USB calls an IN request is sent to the device. The cable has to go quiet before the device sends the response. At 480MHZ this time period is very fast.

What happens with poorly designed cables is the response from the device becomes clobbered and therefore the computer sees the packet as an error.

Can this be heard from a microphone? NO! does it cause audible problems, YES.


Deli's picture

Doesn't a microphone pick up audio? I will admit, our ears are far superior to a manmade microphone, but if something is audibly wrong, why wouldn't the microphone pick it up?

Wavelength's picture


Yes microphones pick up audio. But in this case not everything is in the audible range. Allot of the problems we are talking about here are RF problems which a microphone will not pick up.

As designers we cannot just think of what is audible, but what effect these connections make to each other and how they are carried out. Therefore new tests are being thought up all the time to see and fix problems like this.


CG's picture

Be honest - there's really no point in anybody trying to explain any of this, is there?

Michael Lavorgna's picture
There are many reasonable people out there who appreciate and learn from useful comments including yours and Gordon's. I've received emails from some of them related to this topic as well as Pono/high res. It's just that most of these people do not feel comfortable commenting here, and for good reason.

People like "MrkiMile", a DBA/Software Enginner by trade, create a toxic environment with their smug, smartass, attitude even though they are not qualified or mature enough to intelligently discuss the topic.

CG's picture

Everybody is entitled to their opinion, at least in my book,

I guess that not everybody feels that way.

Michael Lavorgna's picture
Not everybody feels that way.
CG's picture

I think part of the confusion at the block diagram level is that the various things we are comparing are not really comparable.

For example, an SNR of 35 or so dB really works great for digital transmission. The bit error rate will be very low - almost negligible. For Ethernet, 90 mV of noise would give you that. 90 mV of noise riding on a 2 volt audio signal is about 30 dB SNR. Not so good.

In video, a really great SNR equivalent to when the video signal was analog is about 50 or so dB. Not so great for audio.

The real key thing to consider is that at some point all these 1's and 0's get turned into an analog signal. That's what it takes to drive loudspeakers and headphones. You just can't get around this part. So, any noise that comes along with the digital signal and its connections can compromise the audio signal. The noise can be directly in the audio band. It can be outside it, where it can degrade the conversion clock or the actual DAC conversion process. Outside the band noise can even cause intermodulation products right in the analog circuitry that fall within the audio band.

If you design and build all your gear yourself, you can build noise barriers to your heart's content. That can solve the problem. If you mix and match commercial gear, you are left to doing the best you can with the available products. That includes cabling.

JR_Audio's picture

Hi Michael

I appreciate your effort and time to open this post, because as you mentioned, this kind of post can be discussed to death, without even listening to the cable or more important to the music.

As mentioned in one of your earlier posts, I also have the AQ diamond LAN cable in my hifi system and I enjoyed it very much (as I do also have the AQ Diamond Firewire and Diamond USB cable)

When I read your first “review” of the Audioquest Forrest and Cinnamon RJ45 cable, some years back, I was skeptical (as Gordon mentioned, we where also skeptical about SPDIF cable and then with USB cables), but then I tried the complete Audioquest line of LAN cables together with some other LAN cables of other manufacturers and ended up with using the AQ Diamond RJ45.

I do have only vague explanations, why this all is audible (without changing any Bit), but for the reason, I am not a LAN cable developer, I just prefer to listen and choose that, what brings me the most enjoyment in listening to music. If a product does get me more pleasure and does invite me more often to listen to the music, than this is what does count for me on the long run.

Happy discussion, but for most, happy listening.


creeble's picture

Reminds me of an old adage from the computer business in the 1980's. Seems perfectly suited here.

Do you know the difference between a high-end audio salesman and a used car salesman?

The used car salesperson *knows* when he's lying.

Patrick Butler's picture

The knowing part is what makes it a lie.

Michael Lavorgna's picture terms of hi-fi and putting together an audio system are, or were at one time, audio salesmen.
HiTech Guy's picture

Sorry for what appears to be an inflammatory subject, but it's a play on a phrase often used to "try to get to the heart of the matter" ... (Eagles/Don Henley reference).

I believe the problem in discussing any part of the auto stream from studio recording to listening is that it depends on your use case and goes from there.

A friend who can burn $100 bills to light his Cuban cigars and I have developed a phrase of "good enough" for when and where you listen to music. For example:

Casual /Ambient Listening

He spends far more on his "kit" than I do, but we both use SONOS for casual listening throughout the day even though it is limited to 44.1 kHz (indeed barfs at anything over that). I've been using Play:1's and they are remarkable in 2 ways: my wife can clearly hear the difference in our bathroom compared to the radio we were using and they meet the requirements of her decorator's view of the world by being small and looking good. At $199 each they also hit a nice price point.

More Serious Listening in a Typical Room

Here the ante is upped and more $$$ and quality are involved. He uses kit from Cambridge Audio (StreamMagic 6, a very good Denon Amplifier and floor standing Monitor Audio Speakers with Subwoofer)

In my case it is a Bose 2.1 System (fire your arrows now) because it is "minimalist" in our most common listening room and meets "the wife" test.

The quality of the sound in both is better than SONOS even when playing 44.1 kHz material because the equipment used is truly better. In both cases, the rooms are terrible from an audio standpoint (hard surfaces, lots of glass, open on one end), so spending more will not really improve the sound given the rooms used

Dedicated Listening

This is the high end use case. I don't happen to like cans (they mess up my hair!), so I invested in a good pair of custom moulded CT 6 IEMs from a small company in Florida called Sleek Audio ... The difference in everything is dramatic, regardless of source, bit-rate, sampling frequency or room acoustics plugged directly into the jack on a PC or Mac. A side benefit in listening is to my ears custom moulded IEMs also block out ambient sounds dramatically.

Other differences are: I use Amarra (with iTunes) when listening, and it shifts sample rates automatically. An A/B test here turning Amarra on and off provides differences that my ears like. Easy is better than hard! I also use an entry-level external HRTmicrostreamer DAC. Each one independently improves the sound and together it is even more improved All in for IEMs, DAC and Amarra, the investment was just $700. Not cheap, but much better than before.

Another version of this would be in a Den which typically has fewer hard surfaces and is more enclosed or a dedicated media room where speaker might be substituted or IEMs or cans. I also have this and have on loan from aforementioned friend a pair of older Monitor Audio Silver 2's (no sub) and a Jamo MPA-101 Amp ... Still get great sound when working on other stuff. I also have a pair of "old" Bose 201's that I switch to periodically and they sound good too (more slings and arrows). On the Bose, I am using simple 12 gauge copper wire; on the Monitor Audio Silvers, I built a set of 6 foot Bi-Wire cables with banana plugs for less than $60.


For those of you who took calculus in college, this is the second derivative, with the use cases being the first derivative. In any of these use cases, you can spend anywhere from $1,000 to tens of times that amount. Each of us has to decide what is "good enough" or the discussion shifts from informative to obsessive quickly.

Happy listening (as I am doing now)! ... Dedicated Mac Mini (soon to be stripped down to essential software and modded to use an internal solid state drive and external Firewire HDD for music), HRT Microstreamer, Jamo amp to Bi-Wired Monitor Audio speakers ...

HiTech Guy's picture

As evidenced by hundreds of thousands of words on what we can hear and not hear, there is considerable interest in "improving" our listening experiences, however we choose to define that.

I would be very interested in your take on how to allocate a digital audio budget on a percentage basis regardless of amount. From what I believe, it should be divided into two parts: Computer/Server and Output from Computer/Server and in each a ranking of most value/use to least. Is one or the other the first step and what percent of a budget should they get? 50/50? 60/50? 40/60?

Output from Computer/Server

What part of the chain is most important and then what percentage of available funds should be allocated to the rest of the components of the digital audio chain: USB, SPDIF, Optical cable; DAC, Amp or DAC/Amp combo; Speaker wire (single or bi-wire; Speakers and headphones (if used in the system); or speaker alone and headphones alone. Note: should USB cables have a "conditioner" inserted in the chain to reduce noise?

Choosing, Modding and Building a Source System

Similarly, should a dedicated computer of some kind be used for transmitting the source material, and if so, where does one get the most leverage: multiuse computer and dedicated computer. If you take the dedicated computer route, what are simple next steps (and costs). Some next steps seem to be obvious if you take the dedicated route (Sonic Studio:
... Swap out or get an SSD installed for applications only strip software to the essentials, use USB only for connection to the DAC or DAC/Amp, use a FireWire or Thunderbolt connection to an External HDD which stores your music library) but others are not as obvious like get: a linear power supply, better AC and DC (internal cables), logic board shield. The fist part can be had for about $400 over the price of the server. The latter can run $3,600 or more all in.

Because of the complexity, it may be necessary to construct things in price bands:

Less than $1,000 ... Pick one: Source or Output
$1,000 - $1,999 ... Beginning Source and Output
$2,000 - $2,999 ... Next level

And so on. But at some price point percentages alone my suffice since the amounts are large enough.

Are you game for doing this?

Michael Lavorgna's picture
My initial reaction is there are too many variables involved for this to be a useful exercise. For example, some people need multi-room/multi-system capabilities while others only have one dedicated hi-fi, others want to be able to access streaming services, others just want to play music from their hard drive or NAS, and on and on.
HiTech Guy's picture

That's why I have come to the "font of all audio" knowledge ... Who is reasonable. Seriously, I know it is a challenge, but breaking it up into parts might help. For example a single room, then dedicated media room, personal den and Family / Living room. Multi-room could be dealt with later

Because you have been doing this for some time, what I would call "the dominant variables" come almost naturally to you and patterns, or missing patterns, emerge from evaluating lots of equipment, software, etc.

This was true for me when i was at a leading high technology industry analyst firm ... I think you forget more than most of us will ever know.

Anyway, thanks for at least considering the subject matter. If you want to create a private dialog to bounce ideas around. Feel free to use my email to contact me.

bobflood's picture

kid in us ("My cable is better than yours ... Oh yea, wanna fight ...) but they are fun to read. And, what the heck, they bring more traffic to the site than the download of the week does.

Seriously though, these system were never designed for the uses we put them to (think e-mail, spreadsheets and printers) so they will always be compromised to some degree. This hobby just is not big enough to get things designed for us alone. As Kal said in an article, " We in audio always wind up optimizing whatever is thought will sell in the broader market". I may not have the quote exactly correct, but the point is clear.

marcusavalon's picture

My gut reaction to $10,0000 Ethernet cables is wow! a real bad case of the emperors new clothes however having done a little research I found some nice technobabble which may just explain why Ethernet cable can be an issue. I hope you enjoy, if someone can explain why the following is untrue then we don't need $10,000 Ethernet cables however if you really must have the best and what follows is true then perhaps there's a case for them.

Ethernet cables, as well as many other types of cables used in AV, are in essence antennas (thin cross section + significant axial length = antenna). You don’t have to look any further than your garage door opener: the antenna is literally a 6” length of wire hanging down, and typically they work pretty well. When you combine the antenna-like behaviour of an Ethernet cable with the increased transmission frequency of the signals it’s carrying, the effects and impact of alien crosstalk (AXT or ANEXT) become much more pronounced.

AXT is crosstalk caused by other cables (and connecting hardware) routed close to the cable of interest. While not a problem at the lower frequencies of Cat 5e and Cat 6 cabling, the higher frequency signals of Cat 6A makes AXT the limiting noise source for those systems. Active equipment can effectively reduce near-end and far-end crosstalk between pairs in a cable using cancellation technology that relies on crosstalk coming from a known source. However, the reason this type of crosstalk is called “alien” is because the source is unknown and cannot be cancelled out by the equipment.

The primary contributor of AXT is the in-wall cable runs. While you can physically mitigate it by separating the runs into separate joist bays, it’s not very practical. The same is true of electrical cables, especially in retrofit situations. 110 VAC 60 Hz is pretty noisy, and can greatly impact AXT, which is why the rule of thumb is to cross electrical cable with low voltage cables at a 90° angle to minimize cross sections.

There are three variants of a shielded cable: an STP construction that includes individual shields for each twisted pair, an S/UTP construction that has an overall cable shield, and an S/STP construction that has both an overall cable shield as well as individual shields for each twisted pair. Typically, the twisted pair shields are to reduce near-end crosstalk (NEXT), while the cable shield is to prevent AXT.

Usually all of the shields are electrically tied together, which makes termination simpler. However, they will only become grounded if the jack they are placed in has its shielding tied to ground. For that reason, if you are setting up a network with Cat 6 shielded cables, you should use Cat 6 shielded couplers and jacks.

junker's picture

Good thing shielded Cat7 cables are so cheap!

Bob Munck's picture

The human mind and sensory apparatus make up a strange and wonderful hugely-complex system. Our ears are not just microphones; they're microphones with free will and a great many conscious and subconscious opinions about the sounds they detect.

A proper double-blind test would present the subject with a variety of kinds of music in a variety of situations and for a variety of lengths of time. It would gather the subject's opinion on which cable -- normal or expensive -- was being used and keep statistics on the correctness of those opinions.

I am absolutely certain that the subject is currently hearing clear, distinct differences in the sounds produced by the two different kinds of cable.

I am absolutely certain that if Mr. Lavorgna or anyone else were the subject, those statistics would indicate that the choice was no better than random.

I am absolutely certain that if the subject believed that the test was honest, he would cease to hear the differences that he now hears so clearly.

Archimago's picture

Can't hear a difference nor find a difference in measurements using my Squeezebox Transporter.

If someone can, please demonstrate with real hardware.

CG's picture

Would you explain something, please?

For a lot of years, various folks have measured analog audio components like preamplifiers and amplifiers.

So far, the folks who claim to hear differences between amplifiers cannot often explanation those differences from the commonly taken measurements. There are exceptions, of course, but they tend to kind of extreme.

Another substantially sized group are certain that the amplifiers all sound the same, despite different measurements. The extreme examples excepted, of course.

Now, similar measurement techniques are used with digital audio source equipment. (Jitter measurements added.)

Why should we expect the results to be any different now with digital gear?

I know you've devoted a *lot* of time performing measurements. What do you make of all this?

Archimago's picture

Some people say they hear a difference, does that mean there WAS a difference? Especially when folks generally don't test or render their opinion based on controlled situations.

My perspective is that objective measurements of electrical waveforms like those out of a DAC work just fine. Things like cables varying the electrical output can be seen without extraordinary testing methods. Hence, *knowing* about cable effects is not complex like amplifiers that do measure differently but people can perceive different preferences (Art Dudley's recent Listening post on Stereophile for example). Or speakers where indeed there are many dimensions to account for preferences, room setups, etc...

Folks like those talking about noise (yourself included?) and such with ethernet cables??? Well, show us how this impacts the audible output from a good DAC especially when data transfer rate for audio is so slow and doesn't even stretch the limits of a decent ethernet network! What's the use in arguing about whether this theory "can" explain something if there's no reasonable demonstration. If it can be demonstrated, then we can confirm the theory and make things better, if not, then there's nothing to worry about, right? (Just as I am not worried about using a generic Cat 6 cable - details as usual on my blog).

Archimago's picture

It does NOT take "a *lot*" of time to run many tests!

I ran through my ethernet cable tests in less than 2 hours after putting the kids to sleep and enjoying the latest episode of "The Walking Dead" last night with the wife. This was even with listening to a couple of tracks here and there to see if I subjectively hear any difference with cable switches. No big deal!

CG's picture

First off, the data rate has absolutely nothing to do with it.

For example, if there is a crappy switching supply (not all switchers are crappy) in the computer, the router, the hub or whatever you're talking about, the noise current will follow whatever available paths it has available to it, in inverse proportion to the impedance of the path at whatever frequency the noise is at. That is pretty basic stuff (Kirchhoff's current law). I suspect you know that.

This noise combine with other currents in the overall circuit, again as expressed by Kirchhoff's current law.

Whether this is audible depends on a whole bunch of considerations, between the circuitry and the overall system set-up.

It also depends on the architecture of the DAC, the associated clock, and so on. High frequency noise currents may not be evident when nothing is being played - the old "I can turn my volume up all the way and I don't hear any noise" thing - because no input signal is being converted. It could well be a mixing thing causing intermodulation distortion. (This is hardly limited to audio digital to analog conversion, either.) Whether a given audio DAC does this is very dependent on so many variables I can't count them. A simple example: If you play nothing, just how do you measure jitter? Does it even matter?

So, this is not a cable issue per se. Cables are just one aspect of the entire system. Whether they are minor or not is a reasonable point for debate.

One more thing - just what is a "reasonable" demonstration? So far, I've never witnessed a single person who *knows* that cables don't matter be swayed one bit by any demonstration or measurement of any kind. I can't explain why that is, but that's been my experience.

My opinion of double blind tests or any variant is that they really aren't entirely helpful in deterring the goodness of an audio anything. Why? The "test gear" is hardly calibrated and certainly "drifts" by the hour, day, week, and month depending on all sorts of things out of the control of the tester. (In a medical test, the tester might ask the test subject for comments, but they hardly depend on these comments to determine the iron level in one's blood or the subject's blood pressure.) So, the test becomes one of testing the test participant rather than that of the amplifier or whatever. If the tester can monitor the test participants brain waves, then we might have something. Get back to me on that...

Instead, I think the much maligned "subjective" review has more validity. Why? It tends to average out the unreliability of the "test equipment" over time. Changes in mood, blood sugar, or whatever tend to smooth out with more samples. Besides, at least for a potential owner and user of the gear under test, what really matters is the enjoyment they derive over time. That's all that should matter here.

However, again, if you are satisfied with your results, good on you! I'd never criticize anybody for somebody making the decision that X is good enough for them. That's not being cynical or snotty, either. Everybody makes choices all the time over what is good enough for them in life in a million areas. This is no different.

Disclaimer: I don't even use Ethernet in my audio set-up. I really don't feel any desire to have a zillion tracks available on instant demand over a network. Copying files from my main storage drive to the playback computer via sneakernet is good enough for me. My approach is to keep the system as simple as possible.

Archimago's picture

Seriously dude.

Lots of words...

Lots of theories...

Did I even say "double blind" anything in this discourse?

No evidence that the theories reflect actual audio playback... (No suggestion what equipment under what situation one should detect a difference.)

And you don't even use ethernet cables!

Well, now you've "virtually met" someone who does not believe digital audio cables make any difference within reasonable operational situations (ie. lossless digital transmission which is almost always!) based on subjective experience and my own attempts at maintaining objectivity.

Steven Plaskin's picture
I find CG's comments to be informative and to the point. It's too bad you don't feel likewise.
Michael Lavorgna's picture
CG is one a few people here who actually have real-world knowledge to speak to this matter in an intelligent manner. If some of the people commenting here who simply "know better" were able to read without prejudice, they might actually learn something.
Michael Lavorgna's picture
Lots of words...

Lots of theories...

You are one of the people saying subjective listening in not sufficient yet when presented with objective data that may help describe the reason for a perceived difference, you object that its "Lots of words" and "Lots of theories". That's priceless.
CG's picture

You did not specifically say double blind tests. Nope. Not a suggestion. You instead said that you did a quick test that you thought was conclusive. I referenced DBT as an example of a controlled test. Most people who are concerned about controlled testing generally consider DBT as the benchmark for controlled testing. You don't? Great!

Otherwise, I just don't get it.

What does whether I use Ethernet cables in my audio system have to do with the effects they might have?

And... What I said was that I have never come across anybody who *knew* about cable effects ever had their mind changed by other argument or evidence. Still haven't.

And, talking about lots of words? Go find any blog I've authored that talks about audio testing. Or anything.

My only interest in even participating in this discussion was to address some of closely held beliefs of some posters that just aren't borne out in actual laboratory tests of digital communications and conversion systems. If nobody cares, or nobody believes any of it, I can certainly spend my time doing something else. I'm good with that completely.

Archimago's picture

Actually I do agree that DBT is the benchmark. Just that it was not the nature of these tests. Nor do I believe it's necessary for determining the effect of ethernet cables.

If someone were to design a good DBT of this, great! I'd love to see the results...

Yes, my blog has lots of words to try to explain things as best I can and fact check as best I can. But at least I show some data :-).

CG's picture

Yes, you have data. Useful? That's your opinion.

But, you win. I give up.

Steven Plaskin's picture

"Some people say they hear a difference, does that mean there WAS a difference? Especially when folks generally don't test or render their opinion based on controlled situations."

This is why I just let Archimago tell me the result and that's good enough for me.

Bob Munck's picture

"Some people say they hear a difference, does that mean there WAS a difference?"

That's essentially the "when a tree falls in the forest and there's no one around..." question. Our instruments detect and measure pressure waves in the atmosphere; our minds hear sounds. Those are two entirely different things.

Our brains have evolved to do amazing amounts and kinds of audio and video filtering on the signals detected by our eyes and ears. For example, there's a spot right in the middle of your field of vision where you're blind. Does it ever bother you? Probably not, because your brain fills it in. I used to build sonar systems for the Navy at NRL; until recently, the very best analysis we could do was to feed the audio directly into the ears of an experienced operator. Now that our systems are approaching the teraflop level of processing power, we may be able to do better in the computer, but I'll bet the guys out in the subs are still keeping their headphones on while they watch our displays.

So if you're asking if all those filters and analyzers in your brain make a difference, "Heck Yeah!" and "Huge!!" Note, however, that it's entirely possible that they add things to the sound that weren't actually there in the pressure wave.

Steven Plaskin's picture

Yes, but one does bring his cognitive processing apparatus into this game. It's called taste. But to suggest that the difference heard by some is really not there can work both ways. One could also say that the no difference crowd simply cannot detect differences.

But I do agree with your basic premise.

Bob Munck's picture

But to suggest that the difference heard by some is really not there can work both ways

That's why you do blind testing; to see if the subjects who hear a difference are actually detecting something or just being spoofed by their mental filters. If the former, there will be a statistically-significant correlation between what they report and the cable in use. If the latter, the correlation will be no better than random choice.

Steven Plaskin's picture
I understand DBT Bob. I really do. Thanks
solo2's picture

If someone has ten thousand in loose change and wants to spend it on ethernet cables or similar humbug because he thinks they make some difference, bless him. But I don't understand why all this ink and electricity is spent reviewing and discussing something that has a market of maybe a couple of dozen fortunate people. The Teac and Sprout reviews, on the other hand, are useful.

Michael Lavorgna's picture
Ranging in price from $29 to $1,195.

This post addresses the recent spattering of articles which chose to focus on a 12m AudioQuest cable which in reality costs $7995 but was mistakenly quoted as $10,000 in all of the above-mentioned articles. I thought it was important to answer the nonsense within these articles and stand up for what I believe to be a valid way to discuss matters related to the enjoyment of music on the hi-fi.

Steven Plaskin's picture
I have to agree. I enjoyed Gordon's and CG's comments about cables. But I'm with you-if it makes someone happy, good for them!
cablesnakeoil's picture

You're an audiophile, you might have bought your Audioquest ethernet cable for a grand, you may power your audio equipment exclusively using PSAudio AC12 chords, but this only works on a LAN. When you do internet streaming, you can't prevent those audio packets from travelling across those standard cables before reaching your router.

Introducing, the first Audiophile Internet Provider using only Audioquest Diamond cables, vastly improving the soundscape.

Michael: besides, if your music sounds better when 'streamed' through those fancy cables, how does it sound if you just transfer them over the wire and play the same file locally?

vindaloosoup's picture

It's cringe worthy observing old school audiophiles playing in a digital world where the rules and parameters are completely different.

Unfortunately however, its the very "journalists" who write about the decline in the audio hobby yet they have somehow become the hand puppets of the snake oil peddlers. Don't they see that this turns the majority of folk away? Don't they realise they are being made fun of and derided? That's the situation concerning the modern day "audiophile". Only the religious have remained faithful. The rest of us have moved on to truly enjoying great fidelity without the nonsense.

It is becoming more apparent to me that the "audiophile" blogs are nothing more than outlets being pimped by the snake oil vendors to carry out there dirty work. It's all rather too cosy, and it circumvents the legal ramifications of the manufacturers making these outlandish claims. You need to go no further than reading about the ruling against Chord for making the same outlandish claims in the advertising of their audiophile ethernet cables. For anyone interested in the legal ruling read this:

I'm beginning to think the jokes on the readers. The snake oil manufacturers are legally unable to make false claims. They have to advertise their product somehow to avoid legal trouble. Enter the audiophile blog, and either take advantage of the gullible, or give them something in return. The audio blogger has gotta make a living right?

Michael Lavorgna's picture
I love a conspiracy theory as much as the next guy. There's good guys, bad guys, and the really smart guy, which would be you vindaloosoup in this case, who has read through all of the ciphers and code and figured the whole thing out, saving human kind from eminent disaster.

When the truth of the matter is much less interesting, which is to say that I listened to a bunch of Ethernet cables and heard a difference between them. Over time, over months and I'm not the only one.

The rest of us have moved on to truly enjoying great fidelity without the nonsense.
Now we're getting somewhere. But here's a bet - I bet you that I enjoy listening to music as much or more than you do ;-)
cablesnakeoil's picture

This legal ruling is very interesting and I note that it sheds a different light on the following part of the article:

Does anyone need a $10,000 Ethernet cable or a $100,000 wrist watch (especially when your cell phone tells you what time it is)? Of course not. But are we then saying we need to regulate how much people spend on their hobbies? Of course not. We are in fact stuck between a rock and a very expensive place. These things exist and people actually buy and enjoy them. Does that fact somehow tarnish every hi-fi hobbyist and every wrist watch wearer? Of course not. There are extremes in every hobby and the farther out you go, the stranger things get to the uninitiated. Ever seen a furry convention?

I saw kind of a shortcut here as at the end of the day, we are not talking about the same thing. What those cables have in common with jewelry or other kinds of luxurious goods is a retail price that is somehow disconnected from the production cost. Now the luxury industry will not claim that you will start walkin
g on a lake if you wear Prada or be more often on time if you buy a Rolex.

I look forward to more of those companies being taken accountable for false advertisement.

Besides, I also wonder the same thing about the ethics of those reviewers, but I am convinced a lot of people strongly believe what they are writing, they just need to be more informed. For the specific case of rj45/usb/hdmi cables, I'm thinking that explaining those guys a few things about modern computer archit
ecture and even some kind of audio programming 101 could be beneficial.