Expensive Ethernet Cables
I hemmed and hawed. I procrastinated. I averted, diverted, rescheduled, delayed, prolonged, and put off reviewing the AudioQuest Vodka and Diamond Ethernet Cables hoping that something would change. Namely, I was hoping someone would be able to tell me why, in no uncertain terms, they do in fact make a difference in the way my music sounds coming through my hi-fi. In many ways and for many reasons I wished they made no difference at all. I wish they were just some audiophile nonsense that I could plug in, listen to, unplug, and simply shrug at their utter ineffectualness. But that's just not how these things work.
The line of AudioQuest Ethernet cables are all rated as Category 7 Ethernet cables and include the least expensive Pearl ($29/1.5M), Forest ($49/1.5M), Cinnamon ($89/1.5M), Vodka ($339/1.5M), and the top of the heap Diamond ($1,195/1.5M). I reviewed the Forest and Cinnamon (see review) and found that they improved the sound of music played through them and the rest of my hi-fi. When Steve Silbermann of AudioQuest told me, "You've got to try the new Vodka and Diamond cables. They're sick." I couldn't resist. Besides my sheer dumbfounded wonder at the proposition of sick-sounding Ethernet cables, I was curious to hear them in my system despite my reluctance to get badgered, again, for writing about expensive Ethernet cables. The way I figure, someone's got to do it so that someone may as well be me.
One of the ways the AudioQuest Ethernet cables differ from one another and other standard Ethernet cables is in their use of silver. The least expensive Pearl doesn't get any added to its long-grain copper, the Forest adds 0.5% silver, Cinnamon gets 1.25%, Vodka 10%, while the Diamond sports, "solid 100% Perfect-Surface Silver". Silver as you may know is a very good conductor, better than copper, and whether or not this makes any difference in the sound these cables impart on a hi-fi system is anyone's guess.
Another difference between these cables lies in what AudioQuest calls a "Noise-Dissipation System" which according to AudioQuest, "...prevents a significant amount of RFI from reaching the equipment’s ground plane." The lower level cables do not have any but the Vodka gets a "Metal-Layer Noise-Dissipation System" while the Diamond gets a "Carbon-Based 3-Layer Noise-Dissipation System". The other difference between these cables is their connectors which is plainly visible in their appearance. More money buys you beefier, locking RJ45 ends.
The Diamond Ethernet cable also employs AudioQuest's Dialectric-Bias System (DBS). I'll let AudioQuest explain:
Dielectric-Bias System (DBS, US Patent #s 7,126,055 & 7,872,195 B1)All dielectric (insulation) slows down and smears the signal traveling inside the conductor, and when insulation is unbiased it slows down different frequencies at different energy levels by varying degrees. This is real problem for time-sensitive, multi-octave audio, and a significant distortion mechanism for all audio cables, digital or analog.
AudioQuest’s patented DBS creates a strong and stable electrostatic field, which saturates and polarizes (organizes) the molecules of the insulation. Saturated (full) insulation absorbs less and therefore releases less out-of-phase energy. Minimizing nonlinear time delays results in clearer sound emerging from a “blacker” background with unexpected detail and dynamic contrast.
Why Ethernet cables? If you store your music on a NAS (Network Attached Storage) device, Ethernet cables are part of your hi-fi. They are in your playback chain. If you connect to the Internet and listen to music streaming from it, you have Ethernet in your hi-fi. My particular setup involves a Netgear ProSafe 5-Port Gigabit Ethernet Switch which is located right in my listening room. There's a very long Ethernet run which is plain old Cat. 5e plenum cable (since it runs through walls and our attic) down to the router on our first floor . But my NAS devices where my music resides are also connected to that same NetGear switch. So the music travels from the NAS, through the switch, and out to my MacBook Pro.
My methodology for this review consisted of my usual routine more or less; listen, swap, listen. Listen longer, weeks at a time, swap, listen again. I mixed in a standard Category 5 Ethernet cable as well as the AudioQuest Cinnamon Ethernet cable and switched the cable between the NetGear switch and my MacBook Pro, as well the cable between the NetGear switch and the NAS.
That's Sick
I really could have written this review months ago. The perceived differences between the Vodka, Diamond, Cinnamon, and Cat. 5 cable are plainly apparent and easy to hear. I'd sum up these differences as more. You get an increasingly large sound picture as you move up the line, greater differentiation between sonic elements, and a greater sense of clarity. I would classify these changes as being better in each case.
I've spent a few months with these cables, and I swapped them in and out any number of times. With quick A/B swapping, which I admit to finding one of the most annoying ways one can listen since you're no longer listening to the music!, I was still able to perceive a clear difference but longer listening allowed me to more fully appreciate the greater clarity and relative ease the Vodka and Diamond cables impart. You can picture the changes when going from the standard Cat. 5 Ethernet cable to the Cinnamon, Vodka, and Diamond as adding more and more color to a faded image. It's as if the sound blossoms more fully with the more expensive cables.
These changes are also cumulative. With all Vodka cable in my system, from NAS to switch and from switch to MacBook, there was a greater level of perceived clarity and scale as compared to swapping out just one of the cables. If I were to pick which cable in the chain made the most difference, I'd say the one that ran from the switch to the MacBook. So if budget is a concern, and when isn't it, I'd suggest starting nearest your music server and working your way back from there. Of course, your results may vary and I'd also recommend experimentation.
Again, these changes are not subtle or slight. I did not have to do any sort of special listening to special tracks, put on a lab coat, or comb my thinning hair in a particular manner. All I had to do was sit and listen and the changes I've described were readily apparent. As plain as day, as the saying goes.
What isn't as plain as day is why these AudioQuest Ethernet cables change the way the music sounds coming through my hi-fi. I can in fact think of more reasons why they can't make a difference. Ethernet is packet-based, it includes rigorous error correction, it works wonderfully for all kinds of super important data and even ineffectual blathering like carrying all of those Facebook statuses, tweets, and Instagrams. If the bits were not relayed in tact on regular old Ethernet, we'd surely know about it. Our networked world wouldn't work! But it does. And these AudioQuest Ethernet cables make a difference.
I Wish I Knew Why But I Don't
I wish I knew why but I don't. I'm not even going to hazard a guess beyond suggesting that the construction of these cables must affect the way in which data is transmitted. My sneaking suspicion is it has something to do with time.
In the hi-fi hobby we are blessed or cursed, your choice, with an inevitable truth—The veracity of anything and everything is decided by listening. Further, the actual worth of something, its value, relates directly to how much enjoyment we get not from it but from what it does to our perception of our beloved music. The greater the musical enjoyment, the greater the worth. Whether we're talking about having a glass of wine (cheap or really precious), a comfortable chair, the right room temperature, or spending the requisite time and energy to position our speakers appropriately in our rooms, all kinds of things contribute to our enjoyment when listening to music on a hi-fi.
If you have your music library sitting on a network, connected with Ethernet cables, and you'd like to improve the sound of your music coming through your hi-fi, I'd suggest you give the AudioQuest Ethernet cables a listen. Whichever ones you feel make the most sense price-wise. If that means you're not interested, no big deal. The things we think we know get in our way all the time. Luckily with hi-fi we only risk losing out on greater levels of enjoyment.
Also on hand and in use during the AudioQuest Vodka Ethernet Cable and Diamond Ethernet Cable review: regular old Cat. 5 Ethernet cable
 | ||
|
||
COMMENTS
I've been into computer networking AND played music files (mostly beeps at the time) since 1984.
I congratulate you on the review and I am happy to see that personal integrity and competence of an audio reviewer shone brightly and unambigously through once again.
Is it possible that the difference is not that the network data are being carried any more accurately, but rather that there is less electromagnetic radiation being emitted for the other components to pick up?
...is not for the Ethernet cable to resolve. It's for the designer (or integrator) of the Ethernet module to resolve.
I just noticed this:
what AudioQuest calls a "Noise-Dissipation System" which according to AudioQuest, "...prevents a significant amount of RFI from reaching the equipment’s ground plane."
So I was not so much off-base.
As long as everybody wants to use low cost computer equipment to stream the data to their DACs, you have to live with the compromises native to that equipment design. The best you can do is to correct or compensate for those compromises. Cables and add-on widgets are fair game.
One area that I think can confuse people is the usual measurement information posted by manufacturers and reviewers. In order to try to pick out teeny weeny low level distortion or other desired tones, it's common practice to average a LOT of analyzer traces to get the noise level down. That's fine, but it doesn't really allow you to catch occasional, or even occasional but repetitive noise bursts - that's another kind of measurement option. These get averaged and might bring up the average observed noise floor by a few tenths of a dB or several dB, depending on the amplitude of the noise burst and its duration. So, if you have a burst that's 20 dB above the nominal average noise floor that occurs every 60 packets, the noise floor might move a dB on the analyzer. Who cares about that?
But, what is the sensitivity to the ear for something like this? Does everybody have the same sensitivity? Does other equipment in the playback chain change the character and relative significance of less repetitive events? Beats me, but it is an aspect of objective measurements that gets ignored, at best, or beaten into your head as being stupid and irrelevant.
See Bruce A. Brown of Puget Sound Studios comment in post #12.
http://www.whatsbestforum.com/showthread.php?11880-HI-End-Ethernet-Cables&p=221602#post221602
The only thing easier than hearing a difference is not hearing a difference.
No, it means these cables are perposterous. Major credibility hit to the site.
I know you resisted and procrastinated...should have put if off until never.
I see you did not use CAT7 cable in the mix. That is a big shame. Because it would have embarassed AQ.
http://compnetworking.about.com/b/2010/07/26/what-is-cat7.htm
http://www.amazon.co.uk/s/?ie=UTF8&keywords=ethernet+cable+cat7&tag=goog...
They are essentially offerring Cat7 cables with fancy jackets and microns of silver coating..what a joke...LOL..the joke is on all of you.
...that I used Cat 5 for the comparison instead of Cat 7?
You are joking right?
.
Category 7 cable (Cat7), (ISO/IEC 11801:2002 category 7/class F), is a cable standard for Ethernet and other interconnect technologies. CAT 7 is backwards compatible with traditional Cat5 and Cat6 Ethernet. Cat7 features even more strict specifications for crosstalk and system noise than Cat 5/6. Shielding has been added for individual wire pairs on the Category 7 cable.
Cat7 has been designed as a standard for Gigabit Ethernet over 100m of copper cabling The cable contains four twisted copper wire pairs, just like the earlier standards. Cat7 can be terminated either with 8P8C compatible GG45 electrical connectors which incorporate the 8P8C standard or with TERA connectors. When combined with GG45 or TERA connectors, Cat7 cable is rated for transmission frequencies of up to 600 MHz. Xmultiple's UltraJAX connectors are designed specifically for these high speeds with not contact pins and a printed circuit board with contact pads integrated into the RJ45 style housing.
You are suggesting that Cat 7's construction affects sound quality as compared to Cat 5. You may want to chat with some of the other people commenting here, like "Khablam" below, and see why your bits are different from his bits.
I am NOT suggesting my bits are any different than any one else!!!
I AM saying that Audioquest are Charlatans who are selling "pro grade" Ethernet cable used all over Europe and parts of America (Cat7) with fancy jackets and cool names and are charging 100x more. A fifty foot run of terminated high grade CAT7 for mission critical applications will cost you $30 or less.
THAT is what I am saying. And to be honest, I have found you to be sensible and earthbound, but Audioquest is pulling the wool over everyone's eyes here.
You notice that I never said an Ethernet cablle cannot make a difference.
...pulling the wool over everyone's ears ;-)
And don't forget the fancy connectors and fancy shielding.
You notice that I never said an Ethernet cablle cannot make a difference.
That was my point. Some people argue here and elsewhere that an Ethernet cable (to spec...) cannot make a difference. You are saying they can make a difference. So you are at odds with some of the bits are bits adherents.
You are correct sir.....
Audioquest publishes NO specifications for their Ethernet cables. Shameful.
And..that is because they mirror Cat 7..the shielding, bandwidth, termination.
Lots of Chutzpah here.
Well, I have no idea at all if these cables are good, bad, or just the same. I don't even have a way to audition such cables since I don't have an Ethernet connection to my audio file server.
But, I can say that I can envision several system related areas where these kinds of cables could make the system sound different from other cables. In fact, that could be true even if the data bits are perfectly transmitted.
Huh?
Example: The ground connections might have lower impedance in the audio band. That could reduce the noise voltages at the receive end - AKA DAC - that could be passed directly on to the preamp or amp.
Ethernet, and most of these other data connection standards, were designed for adequate data transmission capabilities. I really don't think that any thought was applied to how these connections might affect the performance of an audio system. That just wasn't on the radar at any stage. So, a lot of collateral effects come with using these standards. If you don't like it, you can either decide that it's good enough for your use or do something about it. Special cables might be an answer. Again, I have no idea if these cables perform that function.
Now, you could ask why audio gear isn't made totally impervious to these effects. One answer is that a lot of designers don't believe it's an issue or don't know how to deal with it. Another might be that these audio companies know their business and recognize that the customers won't pay for it. Yet another is that so much of the equipment is not designed for use in audio systems that it's futile. Or, none of the above.
1) TCP connections to NAS systems have checksum error correction. Bad checksum? The data gets retransmitted. No error!
2) There *is* no 'ground', the signals are floating/differential. No 'ground' connection between systems.
Now that we've settled on these facts, how in helk is one cable supposed to sound better than another, when the underlying data is the same?
Ahh, where did I say - ever, anywhere - that there is no error correction in Ethernet systems? And, just how is that relevant at all for this discussion?
And... The signalling may be differential. That certainly helps. But, just how much capacitance is there between the primary and secondary of that transformer? How much leakage flux in the transformers? The isolation is really good to prevent electrical shock to users, but not terrific with regard to high frequency isolation. There is a ground connection, too, unless you are powering the whole thing by separate battery supplies. Not to mention any shielding that is grounded as well - even if not at DC.
Those are facts. Don't believe me? Fine. Go see what guys like Henry Ott, Ralph Morrison, and Howard Johnson have to say about this. Read what they say about common mode noise and noise currents. None of them are audio guys, at least as far as I know.
I just don't get why so many folks think that the data integrity is the only thing that matters in a mixed signal system. At some point, there's analog circuitry, including conversion clocks. This is often what is affected. Some time spent with an oscilloscope or spectrum analyzer and maybe a current probe shows this demonstrably. I'm sure there's lots of examples all over the Internet that aren't even related to digital audio, all done by professionals and professional organizations.
If you don't want to believe that, fine by me, If your goal is to shut me up, so that you feel good about your perception of the world, fine, too. That doesn't change the physics.
Here is the thing - If you are getting any problems, its at your DAC. Your DAC should be able to isolate any extra noise (the almost nil amount) coming through the system, just like every NIC does when trying to get a signal through a dirty line.
The people you linked to, while specialists in their fields, are not specialists in networking. That is where this comes into play. Ethernet barely generates any kind of field that would mess with audio, and that is why it is so susceptible to outside interference. That is also why there is strong error checking in the TCP protocol.
Besides, if this were the case, this 48 port switch I have sitting next to me on my desk with 32+ ports used should have such an obscene amount of cross-talk that nothing should work...
...oh wait.
You are comparing apples and armpits.
What do you consider to be a reasonable and acceptable noise current on an Ethernet router? One micro ampere? That's pretty small, right?
Across the Ethernet termination impedance of 100 Ohmso, that's 0.1 millivolts. Compared to the Ethernet signal level of 5 volts peak-to-peak, that's relatively nothing for a properly working digital receiver.
What is the current output of a typical DAC chip? Somewhere around 2-5 millamperes, peak. So, with 1 micro ampere of noise current getting into the analog portion of the DAC, that gives a 65-70 dB SNR, assuming the DAC is fed a 0 dBfs signal. That's around 11 bits of resolution. How great is that? That's also assuming that noise current is truly random noise and doesn't have any sort of coherent spectral characteristics. It also assumes that the noise doesn't affect the conversion clocking system at all.
This is the difference between networking guys and the guys who design networking equipment. The latter worry about this stuff so that the former can plug equipment in and it will work. The networking guys have other things to do, like software issues and unhappy users. But, make no mistake, this all comes to electrical engineering and beyond to physics. Even if it isn't obvious.
I have no more to offer. That means no more replies from me. Get your fun elsewhere.
Perhaps a blind test would have been more appropriate and would have been at proving whether you can and really did hear any difference?!
Is the reviewer aware that digital data carried over a TCP/IP stream is subject to a parity check, and ergo, by design, it is literally impossible for the cable to alter the stream? If the cable did alter a bit, the parity check picks it up and re-requests it, leading to a bit-perfect representation on the other end. Additionally, any breakdowns that went uncorrected (say, from a severely failing cable) would result in complete breaks in the transit, making all the signal fail.
For instance, when I copy 'War and Peace.txt' over an ethernet cable I get 'War and Peace.txt.' on the other end. Every letter is in the same place on both devices. If a failure does occur (say, a bad sector of the HDD made the file marginally corrupt on writing) then the failure would be obvious - whole sections of the book would become garbage. Random error doesn't produce random words any more than random error can "increase clarity".
It doesn't subtlely re-write itself, so that the prose carries a little less emotion in certain sections, or events don't play out a little differently. To suggest that a completely random corruption could do so would be absurd, which is basically how articles like this look.
I strongly urge anyone spending this amount of money on digital cables to instead spend the money on a basic networking course, so that you can come back to the page here and join in telling people why 010100 will always arrive as 010100 and any time 010100 arrives will sound identical to any other time 010100 arrives.
But my job entails listening to things and reporting on what I hear. Sometimes this doesn't match up with what we think we know, as I said in the review.
But my job entails listening to things and reporting on what I hear. Sometimes this doesn't match up with what we think we know, as I said in the review
Look, the only logical conclusion when we know A=A and you're saying you think A!=A is to conclude your ability to measure A is flawed. You want A<A so you hear A<A even though A=A. This is the flawed nature of human perception.
This isn't some far out position - digital data = digital data. The mathematical proof that 010101010 = 010101010 ... is right there. Like equals like. It really could not be any more simple. The word "sponge" is always s p o n g e. There's no way I can transmit the word "sponge" such that the other user sees it as a different concept than "sponge."
Your decoder sees 010101010 whether that travelled through $1000 cable or $5 cable. Data parity ensures that 010101010 being sent always arrives as 010101010.
Your computer then reads 010101010 and converts that into a particular part of the sound. Every time it receives 010101010 is makes the identical sound. To do otherwise is simply faulty. If 010101010 is sent and 010101011 arrives, the parity check rejects the packet and 010101010 is resent. Assuming this fails (and, it really never does) 010101011 is not *nearly* the same thing as 010101010 and the decoder will produce random sound, or drop the frame.
Data is binary. 0=0 and 1=1. Even if the signal is altered in the cable, the reciever is making a determination. If 1.2v is sent but 1.154 volts is recieved, that makes a difference in analoge, it is a different amount of power. In the digital domain, the processor is viewing them both as the same.
Your article here does nothing but harm the perception of your ilk as people who have no idea what they're talking about.
And I stated my position in the review. That you don't agree with the results of my listening to these cables doesn't change that one bit.
that makes absolutely no sense.
what you claim to be hearing is physically impossible based on how digital data is transmitted. You are listening to the exact same bit-perfect stream twice and saying one of them sounds better.
Even claiming they sound DIFFERENT is ludicrous.
If you are hearing any sound at all, the cable is working.
This exact same argument that idiots made about monster cables... look up those amazon reviews.
Simply make a recording of each, and null the signals. Or inspect the files. You'll see they're identical. It is impossible for them not to be.
Assuming the bits are perfectly transmitted, factors such as ground plane noise could still effect the sound; cables like these might make a difference.
No.
That has nothing to do with the cables. Either the DAC is getting bit-perfect data from your other device, or it isn't.
A $2000 benchmark DAC doesn't get "better sound" if the USB cable connecting it to your computer is more or less expensive.
Khablam, I signed up to comment here only to thank you for being a rare voice of reason against what may be the most absurd audiophile snakeoil I've seen (that wasn't a parody). It's immensely frustrating to watch a farce like this (and the ensuing ignorance) play out. Our author, Mr. Lavorgna, is literally insane; he's convinced himself of an effect that's not only utterly illogical it's completely impossible, and—rather than retract this nonsense and apologize for not conducting a legitimate test—he's only dug his heels in deeper as more light is cast upon how digital systems actually work in theory and practice.
Mr. Lavorgna, if your subjective analysis—your impression—of data handled by a component that must not in any way whatsoever alter the transmission concludes there exist differences over multiple trials, then you're selling your employer a faulty service. You experienced confirmation bias, you didn't control all variables, your methodology was flawed, or a combination of the above.
While I appreciate the seemingly overwhelming logic that would suggest the former is the case, I am for better or worse bound by what I experience.
That's what confirmation bias is. That someone else believes what you also believe reinforces your belief - Epistemology 101.
Yet, in that comment, you make reference to others apparently sharing a similar experience as if that were evidence that your claim has merit. Confirmation bias and circular reasoning. Well done. Logical fallacies 101.
No, I'm saying that: a) if you think two bi-directional cable carrying data and using checksum mechanisms to maintain data integrity are distinguishable in sound quality, you can have absolutely no understanding of the mechanism of data transfer; b) that there is even a glimmer of relevance what someone else also thinks they too can hear, given (a).
There is much to distinguish between science and pseudo-scientific guff. You talk all of the latter, and it is backed by none of the former.
So what you're saying is that my error is that I presumed that you wrote your post to be trusted, that you intended to present an authoritative and objective post. In short, it is MY mistake that I confused your "review" for a REVIEW in the normal sense.
So is yours an advertorial masquerading as a review, then? Did the manufacturer pay for this?
That your opinion matters nothing, that you are either delusional or a charlatan, and that people are paying you for nothing.
... that doesn't change that everybody now thinks you are crazy.
OK... I also strongly urge anyone spending any amount of money on any kind of cable - digital or otherwise - to first spend the money on a basic EE course that describes Kirchoff's Law and the use of Nodal Analysis, as well as on a copy of "Noise Reduction Techniques in Electronic Systems" by Henry Ott. Then you can come back to the page here and join in telling people why just getting 010100 to arrive as 010100 is a necessary but insufficient condition for proper mixed signal system operation.
Or perhaps, it's just me. I don't find any musical pleasure at all in looking at a bunch of ones and zeros in a long stream. I much prefer sensing the air moved by an electromechanical transducer that is controlled by an analog electrical signal. Not as good as live musicians, but life is full of compromise.
Do people rail against these products because of the price or is it something else? (I don't like the price, either...)
...that listening to music on a hi-fi required so much damned book work.
Price is relative and if you met some of my relatives you'd know you shouldn't be bothered.
What? You actually think this has anything at all to do with your ears? Silly rabbit...
Books aren't necessary, if you believe your ears. That should be all that matters.
Otrherwise, they are helpful if you are trying to engineer something. Or, if you are trying to challenge other beliefs that may cause people to question their ears.
In some circles that's considered heresy!
Indeed. One has to wonder how these folks choose favorite foods. Or mates. Or, even, music...
first spend the money on a basic EE course that describes Kirchoff's Law and the use of Nodal Analysis, as well as on a copy of "Noise Reduction Techniques in Electronic Systems" by Henry Ott.
Hi, I actually have an EECS degree. The point of digital systems like modern computers is that if you have two signals that correspond to the bit pattern 001, even if one of them is 0.02V 0.03V 0.98V and the other is 0.1V 0.07V 0.89V, the system will tend to 'restore' them back towards the 'nominal' values of 0V and 1V (or whatever values the system uses). Signals closer to 0V get pushed towards 0V, and signals closer to 1V get pushed towards 1V.
just getting 010100 to arrive as 010100 is a necessary but insufficient condition for proper mixed signal system operation.
Nonsense. The network card will feed 010100 to the CPU, the CPU will run it through the FLAC decoder or whatever, and the FLAC decoder will output voltages to the audio out component. Every part of the system from the NAS to audio out is digital, and so cable quality should only make a difference if you're seeing packets drop or whatever. Once the signal becomes analog, of course, you will get different results with different cables, but the signal certainly isn't analog over the Ethernet cable.
Oh, where do I start? Perhaps John Swenson's earlier articles on this web site might be a beginning.
The real crux is that all these various audio and computer gadgets are connected together electrically (unless you use optical fiber, but that's not what these cables are). As such, electrical noise, whether caused by digital signals, analog signals, or even tribolectric properties of the cable itself, can and will propagate right through the entire system right until the speaker terminals unless something prevents that. This noise can and will affect the sonic performance even if the bits are 100% perfect. Exactly how is too long to explain here, but I'm certain Analog Devices has papers about power supply and common mode noise on their web site.
The bits themselves are only tangential to this problem. (Bit patterns can and do affect the levels and the spectral content.)
So, if you have an electrically noisy Ethernet system attached to your (electrically noisy) computer, which in turn is connected to your DAC, both through the explicit connections as well as through any AC mains connections, you get noise at the output! That is, unless you do something to break the noise current loops or dissipate the noise. Where else does the noise current go? Eaten by electromagnetic zombies? Krell monsters?
Again, I don't have any idea if these cables do that job well - that's why we read Michael's reports, isn't it?
This noise can and will affect the sonic performance even if the bits are 100% perfect
This is not true. If noise in the cable exists in sufficient quantity it will alter the bits, and the end result will be digital soup, complete garbage.
If we can verify that the data as sent is identical to the data recieved, then we can conclude the noise did not change the data. This is the digital world - if 1 is read as 1 everytime then it is impossible for change to occur.
Huh? The noise just magically goes away before it gets to the analog circuitry? How does that work?
Khablam, I believe that your mind is entirely made up on this subject. The networking analogy is true, but highly incomplete. Digital transmission in an audio system is not just a matter of getting the bits right. Since every audio system, at least as far as I know, uses loudpseaker transducers that are controlled by analog signal currents, you HAVE to be concerned about distortion and noise. There is no way around that. The components are connected together electrically. Any undesired currents will pass between the components, unless steps are taken to prevent that. No amount of signal modulation abstraction - digital transmission - gets around that. This is not some goofy "audiophool" philosophy - it is conventional physics, understood since the 19th century.
I will make no pretense about whether these cables help reduce the passage of unwanted noise currents. My only insight into this is what Michael has reported. But, to suggest that none of this is relevant may be popular dogma, but doesn't stand up well to actual engineering and scientific analysis.
Huh? The noise just magically goes away before it gets to the analog circuitry? How does that work?
The point of a digital circuit is that it decreases noise. Look at this simple model of a NOT gate. You can see that the large input voltage range of 0 to V_IL is 'compressed' into a small range of V_OH to V_MAX, and similarly for V_IH to V_MAX being 'compressed' to 0 to V_OL (where I assume that V_MAX is the maximum input/output voltage). So if you have a .2V input noise fluctuation, it might turn into .1V, or .05V, or whatever, as long as your input stays within the 'allowable' ranges.
you start by using some logic.
if i have a file that is 0101
and it doesn't make it to my DAC as 0101.
...then there is a problem.
I have built DAC boxes. And the problem was never that the DAC was not getting the data from my source accurately. That isn't even a problem.
The problem comes from quality of playback based on the DAC's operations on that data, jitter, and the transmission of it (noise matters from the DAC outward, not prior to the dac).
I urge anyone in this thread defending these bogus cables "because their ears" to spend $30 on ebay and build/obtain a DAC kit.
It is repeatably provable that the data from your source makes it to the DAC. You could even build your own junky usb cable out of scrap wire and get bit-perfect...
Where did I say that the bits are corrupted?
Where did anyone say you said that?
This is step 2 of the "bad argument" where it becomes about semantically dissecting what was said.
you stated:
Quote: As such, electrical noise, whether caused by digital signals, analog signals, or even tribolectric properties of the cable itself, can and will propagate right through the entire system right until the speaker terminals unless something prevents that. This noise can and will affect the sonic performance even if the bits are 100% perfect.
This is a paradox as relevant to the discussion of these bogus cables.
100% bitperfect means 100% signal maintenance up until it begins being processed by the DAC.
Which in turn means that everything up to the DAC is irrelevant to the problem, including the cable.
So saying 100% bitperfect BUT WITH NOISE is pointless.
Let me ask this another way... If what you say is true, just why is there a practical distance limit on Ethernet or any digital connection system? The signal will be there at the end, right? All 1's and 0's, right?
I'm sure you've see and measured eye diagrams. What does the noise do?
Anyway, all the digital processing circuits are ultimately connected to the DAC which is connected to the analog circuitry. That's the point right?
I can bet that none of this sways you one iota, so there's no point in annoying you any further.
Quote: Let me ask this another way... If what you say is true, just why is there a practical distance limit on Ethernet or any digital connection system? The signal will be there at the end, right? All 1's and 0's, right?
Because of Ohm's law. Every conductor has resistive losses, how much varies depending on conductor material and gauge of conductor. So over a long enough distance the voltage drop becomes large enough to make that "1" not a logical 1 but fall in grey area, and the receiving part isn't sure what it is dealing with. It might be a 1 or it might be a 0, but that's the beauty of digital, its a lot less susceptible to that decay than an analog signal. It either is or it is not, none of this (sort of) that we get in the analog world.
This introduces corruption as some of the bits aren't quite up to par to transmit their desired value. If we lengthen the cable far enough then the voltage drop effectively reduces the voltage difference to nil. Then you are effectively getting no data.
Uhh, then why not just use an amplifier to get the signals to the required level?
If noise does not matter, this should not be an issue. It is either there or not, and an amplifier sound fix this. Right? (There really are ways around this, but that's a different point.)
And, btw, from the "digital is really a very non-linear analog signal" side of the world - the ignorant and dumb side, if you prefer - copper losses (the Ohm's Law part of the equation) is only but one factor in why these Ethernet and other signals are attenuated over distance.
Thanks for this, I am doing a paper for my Physics class (electricity and magnetism) and I will use this and credit you.
Kevin
I get what you are trying to convey, but there is a problem with that idea.
In ethernet networking ALL devices have isolation transformers. These transformers are typically integrated right into the ethernet jack itself, though some devices will keep the transformers (often referred to as "magnetics") external to the port.
These are REQUIRED to have a good data transmission, and to protect the network equipment from the very same ground loops that could cause havoc with our audio equipment. Occasionally there is a news story about someone messing up and forgetting to add the transformers (see the launch of the original raspberry pi).
So really the noise is a non issue up until you get to your DAC.
Interestingly enough it is an exploitation of these isolation transformers that allows POE (power over ethernet) where 2 pairs are used for a voltage source and ground. Because of the isolation transformers, data is properly carried over the lines despite the voltage potential between pairs.
Yes, there are transformers, as was already mentioned. Typically, these have lousy isolation for common mode currents. They do what they are intended for, though. That is, to make the digital part of the system work adequately and to provide isolation to prevent electrical shock due to the potential differences at different parts of a building. (Which should be a clue, btw...)
And, once again, I AM NOT DISPUTING THAT THE DATA ARRIVES AT THE RECEIVE END!
Jeeez
What I am saying is the noise currents, which the Ethernet transceiver can fairly well ignore, at least in a typical audio system, still exist. They can and probably will find their way into the conversion clock and the analog circuitry. These currents have no intelligence and can't tell where they're supposed to go, and whether they are digital noise or analog noise - a real identity crisis. You either need to completely isolate the various parts of the circuit - which includes coupling into and out of the AC mains through the power supplies - or, you need to eliminate the noise in the first place.
As you say, the noise (probably) is a non-issue until you get to your DAC. Very true, although I'd add that it matters downstream of the DAC, too. But, whether the connection is obvious or not, there is an electrical connection throughout the various parts of the system. Go look at Henry Ott's book or web site if you want more on that.
Well, you (guys) must be having some fun taunting people over here and showing off your (supposed) expertise. I doubt that if I sat you in front of a working system with all the test equipment available that I'd sway your opinion one bit. I have no problem with that. I'll just say that you are entitled to your opinion and we'll just agree to disagree. If you want to go back to reddit and gloat over your "victory", go right ahead. I'm glad I could be such entertainment.
.
Khablam's post is, IMO, just a repeat of the old objecivist refrain that it's digital and bits are bits and therefore the ethernet cable cannot make any difference. Electricity is essentially analog, however. I believe that CG hits the nail on the head. The different abilities of CAT7 (or CAT6A) to reject electrical noise entering the system as compared to CAT5A can affect sound quality. Similarly, not all manufacturers take the same approach to shielding/noise rejection within a particular CAT designation.
Yes, my position that when something is identical when received as when it was sent, that it is the same, is some nonsense position that I have taken. *sigh*
I really wonder how you people think you're reading the words on the screen. They were sent by TCP/IP and the identical words appear on your screen as when sent. Internet communication is proof-positive that verified data can be sent over poor cabling as long as it is sent digitally and parity checked.
That you post this is impossible whilst using such technology is a strange irony.
"old objecivist refrain that its digital and bits" is 100% true...
I find it funny that people discussing things on a bitperfect internet seem to think otherwise.
CAT7 and CAT5 cannot "affect sound quality."
Want proof? Play any audio file over the internet.
That's passing through how many different ethernet cables to output on your computer? Dozens?
Yet, apparently magically, the file is exactly the same file as it was while residing on the server! WOW!
A CAT7 comparison with the AQ would be the most apples-to-apples. Assuming the digital data is the same, then this would at least equalize the RF/EMI emission argument...
For this kind of money - even assuming some incremental improvement for some as to yet unknown phenomena - one could most certainly get better bang-for-the-buck elsewhere.
A pair of Auditorium23 speaker cables or a Cat7? A pair of Shindo interconnects or a Cat7 Cable? Or a lot of music! Are we talking improvements comprable to a Pass Class-A Integrated amp for this amount of coin???
I truly beleive this is the sort of slippery-slope voodoo that has taken hi-fi to absurd levels and illegitimized the industry to the detrminent of just about every independent stereo store, and print publication. No amount of full-page AQ ad rev can justify this stuff. Could I be wrong? I hope so. Lavorgna is a good guy IMO, but I'd just run regular Cat7 and buy other hi-fi toys... ;)
Great post junker..the voice of reason...
No amount of full-page AQ ad rev can justify this stuff. Could I be wrong? I hope so.
The ad revenue comment was expected sooner or later but yes, you are wrong.
Lavorgna is a good guy IMO, but I'd just run regular Cat7 and buy other hi-fi toys... ;)
Thanks. I already own all of the A23 speaker cable and Shindo interconnects I need ;-)
" Are we talking improvements comprable to a Pass Class-A Integrated amp for this amount of coin???"
I'm curious... What makes you think that the mark-ups - I presume that is the point of contention - is different for these cables than for a Pass amplifier? Given the economy of scale for hi-fi gear, or really lack of scale, it could be that the equipment and facilities needed for making cables are far more expensive than what is needed to build Pass quality gear.
No, I don't know the answer myself.
If that is not what you are saying, I apologize for misinterpreting your comments.
My point was just that some of the longer lengths of AQ Diamond cost the same or even more than a Pass INT-30A (Michael has one on his "Main" system) and wanted to ask in the scheme of possible contributors to overall sound quality which would have more impact?
Pass INT-30A and Cable Unlimited / Rosewill Cat 7 or cheaper integrated and the best AQ Diamond. Thanks.
I suspect that some of the more vigorous reactions to these AudioQuest Ethernet cables are budget-driven. In other words, people look at them as you're suggesting - I have $X to spend so I can either get these cables or _____.
While this may be the case for some people, I'm certain there are people whose systems and associated budgets easily allow for the addition of the Diamond Ethernet cable in any length they need. And a $5,000 price tag could very well be viewed as a reasonable amount of money to spend for the associated sonic benefits. Its all relative to the system context, imo. And I'm not making any value judgments ;-)
In the end, how much someone chooses to spend on their hobby is not a matter for public debate.
Let's get real.
These cables are for the 1% based on cost per meter and the typical lenghts needs for any home network. As a matter of fact it is for the delusional and the rich.
You really stepped into a Hornests nest here. By not installing pro grade Cat7 cable you have nullified the entire review. Sorry.
I checked the link concerning Bruce B. He also apparantly says he did measurements that verified a null result between AQ and generic Ethernet cable. One look at his reference gear and it is clear he knows what he is doing.
I just don't see the benefit of reviewing something so few can afford on a computer audio site.
I don't think that every installation will necessarily need long runs of this "expensive" cable.
But the 1% are entitled to a review even if you can't see the need for it.
Put aside my comment about the 1%, which I admit was gratiutous, and lets focus on the logic of your answer.
Networked audio by definition requires long cable runs, except locally from the Ethernet jack or router to component.
I would need about 200 feet for my set up.
Even a modest 25 foot or 50 foot run would is absurdly expensive. Audiostream got what it wanted..site traffic and buzz.
...are less than 1M from my switch. So a .75M Ethernet cable works for my setup.
Anyone is the real-world has a finite budget with which to play (no indefinitely loaned demo units or industry "buddy" prices), and it is this budget with which they are depending on the "trusted-source" AudioStream for sage, actionabale advice.
I even reproduced AudioQuest's entire price sheet for their Ethernet cable line in this review.
Also from the review, "If you have your music library sitting on a network, connected with Ethernet cables, and you'd like to improve the sound of your music coming through your hi-fi, I'd suggest you give the AudioQuest Ethernet cables a listen. Whichever ones you feel make the most sense price-wise."
Whether you agree with it or not, that's pretty straight forward advice.
The either/or is
"either you understand basic electronic engineering"
OR
"you make posts defending digital cables as having an effect on quality."
Again, this is an old discussion when monster was trying to sell $150 HDMI cables. They don't work that way.
The more vigorous reactions come from people who are informed in an area seeing marketing BS being propagated by those who are uninformed (though they may be experts in some other area) and taking repeatable, quantifiable experimentation into the realm of the "subjective."
It isn't really up for debate. If you knew how digital files are processed by a DAC your review would simply be a lot different.
No other way to put it... would a goofy analogy help?
The internet would not work if ethernet cables were allowed to turn 0 into 1 and vice versa?
Agreed. But should you be promoting something that has absolutely no value over a standard CAT7 cable? Sure, its not in my range that I'd pay for my system. That, and I have a CCNA, studying for CCNP, and have experience working in data centers so I understand what actual interference over ethernet would do. If such interference would happen that would cause data transmission problems, the connection would start to read unstable, and return to a handshake (SYNACK) to verify that the connection is there there.
The parity check system built into the TCP protocol makes it so that everything has to be bit perfect on both ends. Now, if you were doing something stupid like trying to stream UDP for audio on a low latency, high bandwidth network (i.e. a home network), then you could possibly have a need for this. But streaming UDP for something that can use TCP in such a setting is pointless.
These will greatly enhance the qualities of the Ethernet cables. http://www.francktchang.com/acousticresonators.html
I have tried many of these cables. The best way to provide "clean" data to your computer (define that anyway you want) is to purchase a Acoustive Revive RLI-1 Lan Isolator.
For 250 bucks it was one of the best upgrades I have ever purchased.
Interesting! Their web site shows actual noise measurements, and the technology they espouse for their filter is hardly voodoo - it's right out of the previously mentioned Ott book.
While I have notice some large differences in USB cables ( my first was the Audioquest Diamond but have since moved on), I experienced no such experience with ethernet cables. I have recommended this product to many who were thinking about rewiring their networks, etc for big bucks and have saved them some big bucks as THE most important aspect to clean input into your server, is "the final common denominator" which the RLI-1 is.
I love it.
As I have said on my other sites, these kind of "upgrades" when you least expect it for realatively little money are my favorites.
Thanks for the recommendation. Personally, my solution is to not connect to a network. Unless you count sneakernet - can't beat the electrical isolation of that. That is just my personal choice.
Your experience just goes to show that there's a lot more to the overall system than initially meets the eye. Or, that you're delusional and crazy. But, even if that's so, it works for you! This is a hobby, after all.
there is a lot more to the overall system than initially meets the eye and I am indeed delusional and crazy and make no apologies for being so.
LATEST VIDEO
| Hardware Software | |
Greatest Bits Favorite Bits | |
GB Components List
|
|
Music Computer Audio 101 | |
Resources Blogs | |
Show Reports | |
Community Site Info |
© 2018 AudioStream
