Cables, Bits, and Noise: How Cables Can Make A Sound Difference

"It's not what you look at that matters, it's what you see." Henry David Thoreau

Noise. It's everywhere. When we're talking about hi-fi, noise can present a particularly thorny problem ranging from stuff we can readily hear apart from the music, to noise that is more insidious sneaking into our playback chain and infesting our precious analog signals and our music along with it. Some people like to think that noise is a non-issue, especially when it comes to digital (after all bits are bits), but that's simply not the case since we're talking, ultimately, about the analog world. First a real-world story where we measure the effects of cable related noise.

Back when I reviewed the Pure Vinyl software (see review), Rob Robinson of Channel D delivered the loaner TC Electronic Impact Twin which I used to rip my vinyl using my Rega P3. The Rega has captive single-ended RCA interconnects which necessitated the use of converters, which Channel D also supplied, to connect to the Twin's front-mounted Neutrik combi jack/XLR sockets. Once everything was set up on the hardware side, we launched Pure Vinyl on my MacBook Pro only to find that the noise level of the connection from the TT to my MacBook Pro was unacceptably high.

As we were trouble shooting this problem, Rob picked up the Impact Twin to make sure the connections were all solid and the cables all seated firmly in place. While he did so, Rob noticed that the noise level changed. Aha! Must be a loose connection. After checking and re-checking, the connections were not the problem. Hmm.

We moved the Impact Twin again and saw the noise jump around along with the movement. Then it occurred to Rob that the noise level jumping was directly related to the RCA cables proximity to my La Cie d2 hard drive, which was sitting on my equipment rack in standby mode. The closer the cables from my TT got to the La Cie drive, the more noise crept into my system. With the cables within a few inches of the La Cie drive, the noise level increased by about 20dB. When we removed the La Cie drive, the noise level dropped dramatically.

Interesting.

My best guess is the source of the noise coming from the La Cie was largely due to its switch-mode power supply emitting RF both from the device itself as well as its power cord and the proximity of this noise emission to the turntable cables. This noise made its way into my MacBook Pro through the analog RCAs coming from my Rega 'table making the task at hand, ripping vinyl, impossible due to its high level. Simply moving the La Cie and Rega cables apart solved the problem.

What else emits RF? Computers, laptops, monitors, NAS, routers, switches, and more. WiFi and Bluetooth obviously also emit RF. Power cords are also noise emitters so where possible, keep all of these things away from any cables that attach in any way to your hi-fi. That includes analog cables, USB cables [footnote 1] (that's why they make some USB cables with ferrite beads), and Ethernet cable (which we'll talk about soon). It also makes sense to use shielded cables and ideally balanced cables where applicable. For a more technical look at shielding and balanced cables, refer to Audio Interconnections and Grounding – Dispelling the Myths by Henry Ott.

Here's some good general advice about cables and avoiding noise from recording technology magazine Sound on Sound:

"While we're discussing such 'proximity' interference, remember that the screening of even high-quality audio cables isn't perfect, so keep external analogue and digital ones at least several inches away from mains cables where possible. Also, try to avoid running such cables in parallel with each other, crossing them at right angles if possible and keeping them well away from CRT monitors, wall-wart power supplies, and particularly the switched-mode PSUs often used by laptops and flat-screen monitors. Analogue cables should also be kept away from digital ones where possible."

What About Ethernet Cables?
Extrapolating is tricky business but I'm going to go there anyway. There are two types of Category 5 Ethernet cables; unshielded and shielded. Typically, cheap Category 5 patch cables are unshielded. Back when I reviewed the AudioQuest Ethernet cables, I used an unshielded Category 5 patch cable in my comparisons. As you know if you've read those reviews, I heard a distinct difference between the quality of my music when using different Ethernet cables, the AudioQuest offering a perceived lower noise floor, "There also seemed to be a greater sense of ease as if some underlying noise had been removed." as I initially wrote back in 2012.

The AudioQuest Ethernet cables are solid core Category 7 cables. One difference with Category 7 cables as compared to Cat5/6 is their shielding. From Wikipedia, "...shielding has been added for individual wire pairs and the cable as a whole. Besides the shield, the twisting of the pairs and number of turns per unit length increases RF shielding and protects from crosstalk." It's also worth noting that while all Ethernet cables are designed to reject noise, Cat 6 offers improved noise rejection over Cat 5, and shielded cable like Cat 7 offer even greater rejection. "This means that screened and fully-shielded cabling theoretically offers 100 to 1,000 times the immunity protection from electric field disturbances than UTP [unshielded twisted pair] cabling does!" from a paper by Siemon titled Screened and Shielded Cabling - Noise Immunity, Grounding, and the Antenna Myth.

"...this noise will not affect the transfer of data one iota but it can most certainly be audibly apparent when injected into an audio playback system, i.e. a hi-fi."

Other devices like my multiple NAS, computers, etc. are also emitting RF/EMI that can easily hitch a ride on unshielded Ethernet cables and inject sonically harmful noise directly into the audio path in a networked audio environment. Let's also state the obvious, to hopefully avoid another smattering of outrage, this noise will not affect the transfer of data one iota but it can most certainly be audibly apparent when injected into an audio playback system, i.e. a hi-fi [footnote 2].

Here's an interesting example from reader "CG":

"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 Ohms, 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."

Ethernet Cable Advice
What's a computer audio, networked music loving audiophile to do?

Use good quality shielded Cat 6a or better yet Cat 7 cables. Why did I say "good quality"? Read this report from Blue Jeans cable where they tested 20 Category 6 and 6e Ethernet cables only to find that "four met spec, and of those, two did it by a hair." That's an 80% failure rate. If you take issue with a cable company doing the testing, check out what the Fluke Corporation who make Ethernet test devices had to say. Spoiler Alert: Fluke found the same 80% failure rate in Cat6 cables so this is no...fluke. Cables that fail to meet spec, note that every cable from brick-and-mortar stores used in the Blue Jeans test failed, can also inject unwanted noise and errors into your system. You can also avoid this noise issue completely by using optical isolation, i.e. an Ethernet to Fiber media converter, which is something I may explore in the future.

Here are a few more references addressing these same issues:

"The biggest benefit of Category 6 cabling is the much-improved Signal-to-Noise Ratio (SNR) using the Bandwidth employed by today’s applications and also for future applications. The main result is that Category 6 provides about 12 dB (or 16 times) better Signal-to-Noise Ratio compared to Category 5e over a wide frequency range." from Category 6 vs Category 5e Cabling Systems and Implications for Voice over IP Networks Belden CDT Networking

"Different types of Ethernet cables can vary by as much as 30 dB (or 32X) in common mode rejection ratio, so picking a well designed cable will minimize your bit error rate after installation." from Connectivity: The Last Frontier

And here's another from Electronic Design that gets back to my Cat 5 UTP cable:
"Ethernet’s unshielded twisted pair (UTP) data-transmission cable acts as an antenna. Common-mode noise that leaks to it will show up as conducted or radiated emissions, creating unique electromagnetic interference (EMI) issues."

Are All Cat 6a and Cat 7 Ethernet Cables Created Equal?
In terms of Ethernet in audio applications, I was unable to find any data other than subjective reviews with one unsighted test. If you have no use for these, please skip to the end.

How does the Blue Jean Cat 6 cable compare to the AudioQuest Cat 7 cable? John Darko over on Digital Audio Review compared his Blue Jean Cat 6a cable to the AudioQuest Pearl and Vodka Ethernet cables and in his unsighted listening comparisons (not double-blind), the listener heard a clear difference, with the AudioQuest cables delivering better sounding music.

What about comparing different Category 7 Ethernet cables? While I do not have any other Cat 7 cables, René van Es over at The EAR does and he compared them to the AudioQuest Ethernet cable (read his review) and found that he heard a difference between them, the AudioQuest cables outperforming his Supra CAT-7 cable.

Here's one more example where Computer Audiophile's Chris Connaker talks about his experience with the AudioQuest Vodka Ethernet cables (see his Network Audio Refresher for more great info) and states, "However, I did notice an interesting difference after the complete swap to AQ Vodka cable was finished. My audio system had a lower noise floor."

Silencing Noise Conduits
It's also worth pointing out that the exact equipment we use, including hard drives, NAS, switches, routers, etc. as well as the audio device in question and its ability to handle incoming noise, will all directly relate to how much of an audible difference Ethernet cables will make in our hi-fis. Some manufacturers of hi-fi equipment which attach to a LAN take more care than others when it comes to isolating the Ethernet input. In other words, YMMV so by all means make sure when buying Ethernet cables, or any cables for that matter, you have the option of returning them for a full refund.

Our networked hi-fi's live in increasingly electrically noisy environments. While bits are bits when they remain in the digital world, our analog systems are not immune to noise and cables provide an open invitation for letting sonically deleterious noise in.


Footnote 1. While we're not focusing on USB cables, I thought I'd share this interesting quote.
"The direct-coupled design of USB and Firewire audio interfaces can be susceptible to grounding anomalies that are difficult to discern. This can manifest in a variety of audible ways including hum, artifacts, and noises which can modulate with activity from the mouse, hard drive, wireless network, printer activity, or other functions of the computer." from Apogee Digital Grounding Noise

Footnote 2. Since we've only touched on one basic aspect of a simple noise problem, and we have not delved into how noise contains energy that's spread over a wide range of frequencies and levels, and how different sources of noise have different spectral content, to quote Wikipedia on Audio noise measurement, it's worth exploring this subject in more detail.


Recommended Further Reading
There's no such thing as digital: A conversation with Charles Hansen, Gordon Rankin, and Steve Silberman
Q&A with John Swenson. Part 1: What is Digital?
Q&A with John Swenson. Part 2: Are Bits Just Bits?
Q&A with John Swenson. Part 3: How bit-perfect software can affect sound

COMMENTS
Michael Lavorgna's picture
"Honest disagreement is often a good sign of progress." Mahatma Gandhi
Cables. A very controversial subject, I know. Even so, I am going to ask anyone commenting on this post to keep your comments civil, respectful, and on topic.

Thanks. I look forward to some healthy debate.

ctbarker32's picture

Instead of spilling gallons of digital ink on unsubstantiated theories, why not talk to the inventor of Ethernet - Bob Metcalfe? Maybe he could explain to you how Ethernet error correction works? Or, maybe because he won't support these theories for the Cables Matter Agenda?

Michael Lavorgna's picture
I made a point of relying on data from non-audio companies like Belden, Siemon, etc to support the fact that Ethernet cables are susceptible to noise. I also point to a real, measured instance where noise got into my system via shielded interconnects. Are you disputing these facts?
Maybe he could explain to you how Ethernet error correction works?
I think you are misunderstanding what I said. From the article:
"...this noise will not affect the transfer of data one iota but it can most certainly be audibly apparent when injected into an audio playback system, i.e. a hi-fi."
Error correction has nothing to do with the noise I'm talking about.
Or, maybe because he won't support these theories for the Cables Matter Agenda?
Agenda? My only objective is to try to understand why so many people perceive a difference with cables while others claim there can be no difference.
Michael Lavorgna's picture
If you can prove I have an "agenda", please do so. Since I am certain you can't, I'm going to ask you to refrain from making these kinds of unsubstantiated allegations. If you continue to do so, your account will be blocked.
Agenda: a plan or goal that guides someone's behavior and that is often kept secret
drblank's picture

audio engineer per se. Playing a digital file uses a different protocol than transferring traditional data files. When playing a digital file, timing is highly critical, and if the noise affects timing, that's a bad thing. Also, noise can unfortunately change the electronic pulses that represent 1's and 0's, so it's possible that it can change a 0 to a 1 because of the noise levels.

I did a simple test to see if I could see a difference and I tried a second to cheapest Audioquest USB cable to a El Cheapo no name USB cable and I ran Blackmagic's Speed Test software and the speed of read and write for the Audiquest was about 30% faster. 30% was repeated many times and that's a noticeable improvement to me. Now, this was on my equipment with those two cables, but I'm sure depending on what cables/equipment you have, you may or may not see speed differences. But again, that's just testing data transfer, not audio playback, which uses a different protocol than data transfer.

ctbarker32's picture

I think it would be worthwhile for those that are convinced that digital bits can be changed arbitrarily and that Ethernet cable quality affects perceived analog audio playback, to read this web site for an orderly scientifically repeatable test.

http://archimago.blogspot.com/2015/02/measurements-ethernet-cables-and-a...

I'm not sure why it is acceptable to trash Bob Metcalfe, the co-inventor of Ethernet, without a reprimand from Mr. Lavorgna? I would wager Mr. Metcalfe has forgotten more about the details of Ethernet than anybody including me ever knew in the first place. I suggested Metcalfe since he is a primary source that could help settle the matter. I find it interesting that talking to an obvious primary source expert is ignored and or trashed.

"Playing a digital file uses a different protocol than transferring traditional data files"

I am left breathless with the statement? What on earth does it mean? Do you mean that transferring data files uses TCPIP and audio files use the NetBEUI protocol? Or maybe it's the Netware protocol? Please tell me at what point in the 7 layers of the Network OSI model are these bits are being magically single out for the special treatment?

Mr. Lavorgna gives his bio and says that he has assembled various cables professionally in the 1980's. Did he build Ethernet cables from scratch? I was building Token Ring networks in the late 1980's and then Ethernet networks and hand assembled cables. I guess I was kidding myself when apparently all the bits got through to their destination unscathed?

Additionally, I would like to know how audio is affected when you use only four wires instead of the full eight? For years I only used four wire Ethernet cables and amazingly my audio transfers got through unscathed or changed?

My evidence for the Cables Matter Agenda are based on facts (I know facts today are subject to opinion - My facts Your facts):

1. The revolving header of this very article features the Audioquest Cable company and which is essentially the subject of this article because they push this Ethernet Cable Matters agenda.

2. A former employee and columnist, Steve Mejas (Sic), of Stereophile recently went straight to work for Audioquest after writing positive reviews of Audioquest in Stereophile. I equate this to when a staff member or member of congress goes to work with a K Street lobbyist.

3. Audioquest or some other cable company has often purchased the C4 (Most expensive back cover) page of Stereophile and/or other Audio Rags (Audioquest current C4 of Absolute Sound). The Sterophile 2013 Annual C4 is Crystal Cable. Crystal Cable or Siltech Cable purchased C3 for at least half of the 2012 Stereophile isses. Also, there are no less than three separate Nordost cable Ads on this article page as well.

Michael Lavorgna's picture
...you seem to have missed the most basic point of this article and your "facts" are not in fact, facts.

I specifically stated that the noise I'm talking about will not affect the transfer of data so your comments including, "I guess I was kidding myself when apparently all the bits got through to their destination unscathed?" are irrelevant. As in meaningless. As I also stated in my brief bio notes, the company I owned installed, certified, and tested Ethernet networks. We also dealt with contract work wherein we had to replace faulty Ethernet patch cords on trading floors which were directly causing network issues. Funny, that.

As far as your "facts" go, which are in fact simply conjecture on your part (go ahead and look that word up if you don't believe me), I have said this before but I will say it again - AudioStream's editorial content is in no way influenced by advertising. Advertising is handled by a dedicated group working out of the NY offices.

Yes, Stephen now works for AudioQuest. So what. We also happen to be very good friends so stick that in your pipe and smoke it. No offense intended.

AudioStream is not Stereophile and we are certainly not TAS so there is no relevance as to who advertises in either publication. John Atkinson, Stereophile's Editor, has also gone into excruciating levels of detail in response to the claim that advertisers get some sort of special treatment in their pages, which is utter nonsense. Google the appropriate terms to read his responses for yourself.

With all that said, I do have to wonder why you are here. I suppose you believe you are injecting some "orderly scientifically repeatable" information into this discussion, which I would welcome. Here's the relevant proviso from the link you provided, "within a reasonable ambient electrical noise environment." Since our ambient electrical noise environments vary, as I've shown with the 20dB injection of noise into my system, I remain unconvinced by the rest of the "tests".

Michael Lavorgna's picture
I'm not sure why it is acceptable to trash Bob Metcalfe, the co-inventor of Ethernet, without a reprimand from Mr. Lavorgna?
Here's what "drblank" said:
"Metcalf isn't an audio engineer per se."
I do not see that statement as trashing anyone.

What's funny about your call for a reprimand is you then go on to suggest I'm essentially a paid shill for cable companies, I'm not really sure what your point about Stephen is supposed to mean but you are obviously suggesting some sort of foul play (wrong again Chris and I'll warn you not to even go there), and finally you go on about C4 and C3 ads to support your Cable Matters Agenda pet theory.

As I mentioned to you in a private email, you've had your one and only warning regarding posting libelous conjecture on AudioStream. Do it again, and I will block your account.

drblank's picture

I would get a panel of cable engineers that test and design cables for a living to be part of a discussion panel.

The thing is that some of them have proprietary test equipment or deploy tests that others don't. Plus I don't think there is a single person on the planet that has even tested every single brand/model cable with every single conceivable test. Some companies don't like to tell the public everything because of trade secrets which is understandable. Does Coke give out their formula to Coca Cola so their competitors can make the exact same product? Nope. Not going to happen.

none the less, I think the collective of several different cable engineers in a panel discussion might be helpful and useful because they design/test cables for a living. It would be even better if they showed test measurements to show what differences there are in different quality of cables.

I used to think that Cables didn't matter for many years and it wasn't until I actually heard a difference that I had that way of thinking. Now, I have the mentality of that they CAN make a difference, but don't always, so it's best to at least do some research and narrow down some cables to try out and see if you can do a variety of listening tests to see for yourself. But remember, what you hear on your setup is only what you hear, so blanket statements many times cannot be derived. I'm sure a top mastering engineer in a studio with top equipment obtainable can hear things that a normal Joe can't when we listen to music on a average home stereo.

It takes a long time to develop one's ability to hear and recognize subtle differences in audio, just like it takes a musician or recording engineer to hear differences in equipment and settings they use. Guitar players don't automatically know the difference in sound between two of the same brand/model guitars until they have reached that expertise and listening abilities to distinguish those differences. A novice certainly can't.

drblank's picture

trash Metcalf, i just simply said he's not an audio engineer, he's an EE, so he might not be able to explain streaming audio with different types of cables because they didn't specialize in that area. it's not trash talking them, it's just understanding his area of expertise. I look at EE's as obtaining the basic understanding of electrical engineering, but there are many, many specialities once you obtain an EE.

Would you ask a doctor that specializes in brain surgery a question regarding the digestive tract, or would you ask a specialist in gastroenterology?

When it comes to audio over various types of cables, etc. I would be more inclined to ask someone that is knowledgeable in cable design and understands and is up to date in the area of the latest testing methods of cable design/testing methodology and test equipment.

That's all.

drblank's picture

Some people get jobs with companies because they like the products and want to work with them. What's wrong with that? For example, I use XYZ brand computer and I like their products, so if a job came up to work for that company, I would be more likely to want to work there than a competing brand that I don't like. And? THere's nothing wrong with that, that happens all of the time. You work for companies you like. Do you like the products/services that company you work for makes or would you buy someone else's because they are better? Or do you work for who pays you the most?

I have a friend I met when he was working for an audio dealer. He wasn't looking for a job with a mfg of audio equipment when he first started to work at the stereo dealership. But as time went on, I learned what he likes in terms of sound quality and he simply has his own personal tastes as to what brands/models of product that he likes and he eventually worked for two of them. Does that mean that because he went to work for them years later means that he was pushing their products? I don't think so, he just has his own preferences and because he has the work experience that he can work for these companies should a job opening appear. Does that mean that I have to buy the same equipment he likes or works for? NO. I have to make that decision myself. I will at least listen to the products and get some input as to why he likes them, but that doesn't necessarily mean that I have to.

For some reason, I think you have some amount of angst against Audioquest. Here's a suggestion.

1. Call Audioquest, Wireworld, MIT Cables, Transparent Cables, Nordost, etc. etc. and talk to their head engineer yourself and explain what you've read and if there is anything they can explain that the other information you read doesn't. Sometimes, they might have conducted other tests or used better equipment, that is very possible. Talking to a verity of cable mfg may shed some light on the subject that you just weren't aware of. I know of one company that has some proprietary test equipment and software that's not obtainable outside of their company. So, they might have performed tests to prove things that are applicable to audio cables (analog and/or digital) that everyone outside of their company doesn't have that knowledge or experience. Maybe they can shed some light on cables that you may not be aware of. Maybe they are doing tests other articles you've read didn't. Maybe there are other factors to consider the others didn't. I wouldn't take one or two articles written by people that don't design or test cables for a living to be experts in cable design and testing. I would be just as suspect to their tests as I would anyone else. They might not have the best test conditions, test equipment, or even know what tests to perform other than the generic tests they were taught by someone else that doesn't design/test high end cables.

Take whatever skepticism you have and call a variety of cable engineers and ask them questions and discuss what you know and have read and maybe they will explain something you didn't know about.

Do your own listening tests, but just know that whatever listening tests you perform don't mean that someone else has or will have the same experience. It's only your experience in your environment and conditions.

John Sully's picture

This is just flat out misinformation. Data is transferred over ethernet using one and only one protocol 802.3. It is all the same whether it is a file, video, audio, a website, data transferred from a datalogger, whatever. It is all out on the wire in the same format.

You are just wrong.

drblank's picture

if all you are doing is transferring a file from one place to another, but we are talking about streaming audio in real time.

Here's one site that might explain the different types of streaming audio.

http://www.garymcgath.com/streamingprotocols.html

John Sully's picture

All those other protocols he talks about ride on top of 802.3 and TCP/IP. Have you ever unplugged the ethernet cable from your AVR or DAC while playing music from a local source? How long does it play? For me it will play for 10 or 15 seconds. That is not real time or even near real time, and ethernet, because of it's congestion mediation protocol (CDMA) cannot provide hard real time guarantees.

skateman's picture

On the topic of cable quality affecting noise (eth/usb):

1) Are we talking about poor cable quality that would audibly perturb data transmission (e.g: drops or crackling), or are we talking about cables that have *better* soundstage, emotions, basses, etc .. ?
2) Specifically on the Ethernet cables and playback being affected by cable quality. Are we talking about audio being transferred in a push or pull model?

Steven Plaskin's picture

"1) Are we talking about poor cable quality that would audibly perturb data transmission (e.g: drops or crackling), or are we talking about cables that have *better* soundstage, emotions, basses, etc .. ?"

This was discussed in the comments section of the JCAT Reference USB cable. And yes, USB cables can differ in soundstage, bass, etc. unless the DAC is extremely well isolated from the effects of computers, cables, etc.

My review of the Resonessence Mirus discussed this.

Thanks

skateman's picture

Is this the link you have in mind?

http://www.audiostream.com/content/draft

Steven Plaskin's picture
If you are not interested in what these people have to offer and need a scientific study, please forget my comments.
Michael Lavorgna's picture
To answer your questions:

1) I specifically said, more than once, that I am not talking about noise that will cause data errors. Cables do not have "*better* soundstage, emotions, basses, etc." Try listening to any cable when it is not attached to anything and you'll see what I mean. Obviously cables are part of a system.
2) Ethernet transmission bi-directional. If it wasn't error checking wouldn't work.

skateman's picture

1) I specifically said, more than once, that I am not talking about noise that will cause data errors. Cables do not have "*better* soundstage, emotions, basses, etc." Try listening to any cable when it is not attached to anything and you'll see what I mean. Obviously cables are part of a system.

I see, so if there is no difference in terms of soundstage/emotions/basses between cables, you probably want to take a look at the JCAT reference cable review that was posted to this blog as it claims the opposite.

2) Ethernet transmission bi-directional. If it wasn't error checking wouldn't work.

I'm afraid you missed my question. I was specifically asking whether we were talking about networked audio in a push or pull paradigm.

Michael Lavorgna's picture
...is a failure to communicate.

Every cable is part of a system. Therefore, changing cables can certainly cause changes in the sound of that system. Cables in and of themselves have no sound.

Your second question is not relevant.

derneck's picture

to know your educational / professional background, Michael (obviously other than audio reviewer).

Michael Lavorgna's picture
I've been thinking of writing up a bio of sorts. In the mean time, here's the crib note version of my professional background in reverse chronological order. Overall I spent about 20 years in the information technology and application development fields.

Prior to becoming Editor of AudioStream, I worked for a small NYC-based company that designed and built custom web-based apps for hedge funds as well as a few debit card companies. We also provided remote Unix/Linux Admin for hedge funds. My job encompassed running the finances of the company, performing COO roles, designing the actual web-based interfaces, and representing the company in all contract negotiations. I also wrote most of the proposals and contracts.

I worked for a few years in a two-man company designing and building web-based applications.

For roughly 15 years I was involved in IT consulting. Most recently as the owner, CFO, and COO of a company that serviced financial services firms in NYC. We provided personnel to support everything from the base building cable plant up to and including the WAN. This also included server admins, database admins, project management, etc. We also designed, installed, and certified Ethernet networks for companies in the tri-state area and we also had another division that provided a host of IT services to small and medium sized companies. When I left, we had about 200 employees which included full-time employees with benefits, contractors, and part-time workers.

Before becoming owner, etc, I worked for a number of years as an IT consultant to a few banks and financial firms. This work encompassed everything from installing new trading floor users, including the physical connectivity to the desktop, as well as troubleshooting network-related problems. I also developed a very simple system for documenting and maintaining all of the network connectivity for a large financial services firm.

I also worked for a number of years, before, during and after high school, hand terminating, testing, and installing pretty much any type of cable you can imagine (we're talking about the '80s). I would venture to say that I've terminated, tested, and run more cable than most readers of AudioStream.

Also on the side, I designed and maintained a few websites.

I almost forgot my education! I have a BA in Fine Arts from Bennington College. While living and working in NYC, I spent most of what little off-time I had pursuing a career in painting. The farthest I ever got was to be included in a few editions of a very small, very limited poetry and fine art journal.

derneck's picture

...does work! Though I daresay you're of more developer than IT mindset.

Michael Lavorgna's picture
...ex post facto :-)

I very much enjoyed aspects of both mind sets.

fmak's picture

such as impedance matching and quality of eye pattern are all important.

I have a recent experience of Wireworld Starlight cable whereupon I removed the plasticised end caps to find the usb connectors partially shielded only. Instead of using an upper and lower metal case, the lower metal cavity has been filled with somekind of hot melt(?), thus exposing a perhaps 1.5cm length without shielding.

I made up copper shields on the plugs which can be moved along the cable. Rather surprising has been that, when the plugs are fully shielded, the sound improves, with hf grit noticeably reduced. Sliding the shield along the cable has little effect.

One mod that has a good deal of effect is the fitting of an Elfidelity DDR3 Ram filter onto a spare RAM socket. The filter consists of a set of NEC-Tokin high frequency capacitance filters which suppresses very high frequency noise.

The list of factors that affect audio quality just seem to go on and on.

JohnKing's picture

There are so many potential weak links and unwitting noise carriers/emitters in computer audio that it's sometimes difficult to assess the beneficial impact of changing just one cable-- the benefit of one change can underline or be masked by problems in other parts of the system. In my case with a modest system, I thought the Blue Jeans Cat6a Ethernet cable made things worse (too bright)at first.

But when improvements in power filtration, cables and computer settings start to add up the impact of individual changes can be seen in a more positive light. When I got the computer/NAS/router power bricks and wall warts filtered and somewhat isolated from the linear powered gear I knew the BlueJeans cable was $10 very well spent. I'm now wondering about moving to linear power supplies to replace the switched mode units... and hoping to find some trusted reviews here.

Steven Plaskin's picture

John,

Here is a new product from Shunyata that deals with both digital and analog components:

http://www.shunyata.com/index.php/VENOM-SERIES/venom-mpc12-new.html#!mpc...

JohnKing's picture

Thanks, that looks like a good unit but price is a barrier. I'd be interested to see your comments on the HDPlex linear supply units you show in your system. thanks.

Steven Plaskin's picture
The HDPlex linear power supply has improved the sound in my system driving the hard drives and iFi USBPower. The unit can drive multiple devices of different voltages and is well built for the money.
Steven Plaskin's picture

Here is a new iFi Audio product that might interest you; price?

http://ifi-audio.com/portfolio-view/accessory-ipower/

JohnKing's picture

That looks interesting! I'm already using their iUSB with their Gemini cable and both make a positive difference.

Yes, we need to see the price and some reviews...

Steven Plaskin's picture
It appears the price will be around $50 when available.
BEC's picture

The designers of Audiophile Equipment and many cables are well educated engineers. Paul McGowan of PS Audio, Nelson Pass of Pass Labs,Jack Bybee, a quantum physicist who worked on super silent submarine technology for the US Navy, and many other highly educated people work designing eqipemt and cables that Audiophiles buy. I don't care if Michael has a Degree in EE or Computer Science. Judging by his written communication I woulld never label him as uneducated. I care about what he hears. I have engineering background, and I didn't believe there was a difference in equipment until I heard the difference at a friends house. The difference was undeniable. I was proven wrong, along with five other people, and have been enjoying High End audio ever since.

The human ear and body are not junk, but remarkable feats of engineering. Yes we can hear differences in equipment that the test bench cant measure. There is an old review of Mark Levinson equipment from the 80s done by the late Julian Hirsch. Julian Hirsch was a respected menber of the the Audio Engineering Society and founder of Hirsch Houck labs. He believed that there were no audible differences between well designed amplifiers with low distrotion and noise measurements. This was Julian Hirsch's first encounter with equipment of this caliber. He admitted something that he had previously never done before. He said that the Levinson preamplifier, power amplifier and special cables exhibited a type of "Gold-Plated Sound" which he could not explain based on the normal test bench measurements he obtained. He didn't want to belive it, but it was there. A Mark Levinson preamp and power amplifer, probabably deigned by Engineer Tom Colengelo, or Engineer John Curl, sounded much better than a Pioneer receiver of the day.

I challenge you to go Audition some quality (not Bose) Audio equipment as I did. You might end up hearing that "Gold Plated Sound" Julian Hirsch was talking about.

Michael Lavorgna's picture
Welcome to the club ;-) Yes, there are a lot of very intelligent people behind the products we enjoy.

Cheers.

2_channel_ears's picture

Whenever the subject of cable placement comes up I am always reminded of a most useful Sterophile I read many a moons ago written by the ever witful John-A-Ten Skull.

Having spent too many (sometimes frightful) hours in EMI labs tackling noise issues, the veritable light lit bright right away. I dressed the cables to the "T" per the article. Total cost was a couple of bucks in tie-wraps. Naysayers can say what they will but my system gained deep impactful bass, depth and clarity. Here's the article, highly recommended:

http://www.stereophile.com/finetunes/188/index.html

On the idea of optical isolation, I can vouch this works incredibly well, if you can implement it. Have never tried on Ethernet but used it on signaling lines from a PC for an extremely sensitive (and finicky) RF receiver test.

dallasjustice's picture

In footnote 2 you say you wish to delve into the noise issue further. I believe the theory you and many others like John Swenson have put forth is that RFI noise can ride along the digital cable. This noise the somehow negatively affects the clock or DAC. If this is the hypothesis, we could expect to see some analog measurement from DAC's output which could demonstrate this effect. I've read lots of stories and listening tests but I've never seen any repeatable analog measurement which could prove or disprove this theory using a proper asynchronous data transmission or ethernet which can check the data at both ends.

Often times folks advocate their perspective based on logic, appeals to authority and other war stories. I don't find these forms of proof persuasive. I'm still undecided on this but I know it's truly foolish to assert there is definitive proof of these effects. It doesn't exist. Maybe there's "some" evidence but nothing certain. I hope there's room for alternative perspectives here.
Michael.

Michael Lavorgna's picture
...I have yet to put forth.

As far as Ethernet error checking goes, this has nothing to do with what I'm talking about here.

Reed's picture

I have settled on studio grade cabling. I have Mogami Studio Gold balanced cables and Blue Jeans LC-1 cables. I have had a ton of cables in the past. Some examples are Audiovox, several Nordost, Cardas, Chord, Audioquest, etc. I finally settled on studio grade type cables. My philosophy is to follow the path of the recording industry. When I first put them in my system, they sounded a bit uninteresting, but after several weeks they sound smooth and detailed. When I switch back Any of those other cables now, they sound bright and thin. It's not that they broke in, but I think rather that I adjusted. The brain and ear are very complex...and not fixed in their ability. Like I stated in a prior posting, I try to match my system to what I hear at the live venue. They get me closest to that.

bscharp's picture

After reading your article, I googled "Computer RF noise" and read some interesting info. The amateur radio guys have been fighting to keep computer RF out of their equipment and have some good suggestions. Other articles discuss " ferrite core filters...small rings that can be placed around a cable and filter out noise. Perhaps a ferrite core filter would remove RF noise from an ethernet cable.

bobflood's picture

I bought a bunch in various sizes (from an amateur radio supply house) and placed one at the input end of every cable in my system and they did make a big difference in the noise level. Prior to this, I could always hear a low level hiss (for lack of a better term) when I placed my ear close to my speakers without music playing. After putting these on, the noise is gone. Many of the cable makers use them as a permanent built in part of the cable, especially on power cables.

hltf's picture

Michael:
After reading your writeup about Ethernet cables I decided to try removing my ethernet cable (I think it is Cat 5, some years old) and played back a playlist without any internet connection at all. I believe the sound improved - less fuzz/buzz on edges and indistinctness perhaps is the way that comes to mind to describe the difference. However, because I remote connect to my MacMini using Apple Remote on my Macbook laptop, I do ordinarily need an internet connection active on the MacMini. I am going to experiment with running my MacMini on WiFi so I can get a remote connection into it through my home network on my MacBook. On first listen, with WiFi it seems that the benefit of eliminating my Cat 5 ethernet cable is retained. But there are a couple of variables, so I guess over time I should be able to decide for sure. Just curious though, what are your thoughts about the two options - using WiFi for a remote connection (but not to access a music library, that is on an external drive connected nicely with a cable) vs using ethernet for the same purpose? I wonder if I laid that out clearly enough?
Thank you

hltf's picture

Michael:
My header inaccurately describes what I am asking about. So to clarify, I do not currently nor do I plan to stream music over WiFi or ethernet. I store my music on an external drive. But I do need an internet connection on my MacMini only to remote into it and operate it as I do not like to use a dedicated monitor or TV which I find noisy. So my question is would you prefer ethernet or WiFi for my purposes?

And here is an additional question on my mind in this regard, if one prefers WiFi for this, would I not be better off with an external drive other than NAS drive, since as I understand it, NAS requires ethernet cable connection, right?

Thank you

Michael Lavorgna's picture
...I would recommend experimenting. In terms of a NAS v. external hard drive, I would say that if you want to share your music library among several devices throughout your home, a NAS is the better solution. On the other hand if you are only using your MacMini, stick with your hard drive. While I have not tested this, some very knowledgeable people recommend not using a USB drive if you are using a USB DAC.

In any case, just make sure to backup.

hltf's picture

Thanks Michael. Makes sense to try myself of course. And I have confirmed fully to my satisfaction that WiFi-ing into my MacMini to remote control it is much better than using the ethernet cable. So your article has been well worth reading for me as it got me thinking about trying this configuration, something I had set aside as an option long ago for some forgotten reason.

In fact, I think going WiFi for the purpose of remote control of the MacMini is sonically equivalent to what I heard without any internet.

And, yes, I do connect to the external hard drive with a FireWire from the MacMini. The USB section is dedicated during listening to the DAC and USB convertor.

So thanks for the help and article. It has proved very helpful.

Michael Lavorgna's picture
and I do mean it gives me great pleasure that this article has helped you get better sound which leads to more musical enjoyment.

Cheers and enjoy.

drblank's picture

Over the weekend, I was trying to see if I could hear a difference between AAC 256kbps on my hard drive vs 320kbps AAC on Spotify streaming.

I also tried 320kbps AAC on Spotify vs AIFF on my hard drive. In one case, I had a new track that I hadn't heard that much, but it was a live big band recording that seemed to be quite natural in the way it was recorded and mastered. I heard a lot of low level detail and could hear the natural room acoustics clearly. What I did is when I ran one of the tests, I wasn't listening to when the full band was playing, I took a very low level passage and I concentrated on that, and I heard a lot more clarity in the AIFF file vs AAC streaming, I also heard more bass, and I heard low level detail better.

So when people do listening tests, there is no standardized method or even sound file to listen to. So, just a recommendation with cables, choose files that have no compression or limiting, and listen to low level passages and see if there are noticeable differences. Since cables is difficult to swap in a AB back and forth manner, it's hard to do this with power cables, since you have to power up/down electronics and that just takes time, and some electronics don't like being powered up/down all of the time, so that could be a factor.

Just understand that listening to pristine audio tracks to begin with can change the outcome of a listening test.

fmak's picture

quite right. Unfortunately many are 'brainwashed' by basic elements of 'bits' and 'noise' that have been stated or learntt, with inadequate understandings of the elemental and global transfer functions within and out of a system.

There is almost no consideration of audio acoustics and noise annoyance versus noise perception.

I am a professional engineer who practised in areas of measurement, control, environmental acoustics, and noise and vibration control.

A good example of ignorance is a person in another forum who connected a fanned lab power supply to a fanned MAC using croc clips without any form of external decoupling, and proclaimed that he couldn't discern any differences in audio replay. The same guy was stacking his MAC on top of his dac, with a large power supply sitting next to it. If anyone had measured or understood the natures of the magnetic and other em fields around the system, it would have been obvious that the setup wasn't even half right. Unfortunately some are convinced by the pretty pictures that the 'ITI guy posted.

bobflood's picture

noise coming from the computer in your system, try this. Get an ethernet isolator (just search the term) and insert it as close to the input of the computer as possible and also get an independent 5V power supply for the USB output (Aqvox has a very reasonable model). This should lower the noise entering the system and emanating from it by a large degree.

skateman's picture

I'm finally confused. If cables don't change the soundstage, are power cables an exception??

Audiostream review of PowerCell 10 UEF:

http://www.audiostream.com/content/synergistic-research-powercell-uef-an...

Michael Lavorgna's picture
Re-read my last comment to your last inane question which will answer this inane question.
skateman's picture

I did re-read this last comment and it just says that my second question is irrelevant, which is making me assume that you have no idea what pull is vs push.

Michael Lavorgna's picture
Here, let me help:
Every cable is part of a system. Therefore, changing cables can certainly cause changes in the sound of that system. Cables in and of themselves have no sound.
That is the relevant part of my last comment, not the fact that your push/pull question is irrelevant.
Chrisg2229's picture

I think you're taking your new "moderator" role too seriously. Before going too much farther with your new unwritten policies, you need to sort out what the rules of engagement are on this site so people don't run afoul of arbitrary decisions in the absence of a posted code of conduct and get banned or, worse, sued!

Christ, even at Head-fi and other sites where it can get nasty, posters don't get warned publicly like you did to ctbarker32 and I've never seen anyone in our hobby that runs a website call out "libel" for something written in the comments section or even in a forum.

As for the facts presented by ctbarker32, do his comments about advertising dollars suggest a conflict of interest? Perhaps, but I do not see anything in print where he wrote that you have personally received in the past (or are currently receiving) payola for saying that cables matter.

As a public service to both you and your readers, for something to be considered libel it must be a written statement which claims to be fact and is not clearly identified as an opinion. Again, nothing was written as fact about you personally nor was it even suggested that you personally are "essentially a paid shill for cable companies".

And as for a conflict of interest, I would probably use the term "conflict of objectivity".

The definition of objective is "not influenced by personal feelings, interpretations, or prejudice; based on facts; unbiased" And, while perhaps not a conflict of interest, all of the companies in the TEN network have the same potential conflict of "objectivity".

Why?
First there are the financial relationships - TEN network's magazines/websites receive advertising dollars from the firms whose products their employees review. Secondly, there are the personal relationships - many reviewers and manufacturers reps/owners are friends (for example, your friendship with Stephen Meijas an employee of Audioquest) or have known each other for years, etc.

So even in the land of subjective reviews, there needs to be some objectivity.

And when it comes to something as contentious as "do cables matter?", in the absence of using the scientific method to prove if they do or don't matter, it's not hard for the skeptics to think that the professional believers (reviewers) are influenced (unknowingly or otherwise) by these financial and personal relationships.

That's why facts matter, even in a subjective review. In the case of the noise in an analog cable, we've all heard it and I bet it's measurable. Is the same true in a USB or Ethernet cable? Don't know...haven't heard it myself and I have yet to see any measurements. I do use Blue Jean Cables BTW and I also use a GISO DS on my ARC Ref DAC; can't be too careful you know.

To sum up, if you feel so strongly that you need to have the ability to ban people for comments, (1) the Audiostream Code of Conduct needs to be written & posted (2) once posted, people need to agree to it when they register and those of us that are already registered need to agree to it, probably by re-registering for this site and (3) you need to get someone else to be a moderator since it is hard for anyone to be objective when another person is commenting/challenging/criticizing the author of a story as we've seen here.

Once that's done, ban away! Until then, it's kind of hard to play by the rules if you don't know what the rules are.

Michael Lavorgna's picture
Be respectful, civil, and don't accuse me of shit you can't prove. I remind offenders of these simple rules and if they cannot abide by them, I block their account. Simple.
fmak's picture

I agree.

However, what about the postings of plain untruths to anyone with experience and exposure to high quality high end audio? These can seemingly be 'expert' comments that lead many others to believe in approaches that cost them a lot of money and effort but to no end?

Michael Lavorgna's picture
People are free to disagree and point to errors or misinformation as they see fit. I've never banned anyone for correcting me or anyone else.

Disagreement can certainly be civil and respectful but often times it isn't, at least on the Internet.

llminsk's picture

Have been reading this blog from the day one and finally have registered to comment on this post (well, inspired by Monthy Pyton post as well, thanks Michael :-)

My dear audiophile colleagues, do you know how it feels for a person with some experience in ultra weak signal detection in extra noisy environment to read about ethernet EM crosstalk onto audio circuits?

Like a VOGON POETRY - deadly painful.

Just imagine someone who just spent amount equal to average US household annual income buying super-duper balanced interconnects to achieve ultimate common mode rejection ratio (well most probably this person doesn't operate language like that) just to discover that there is one more source of electromagnetic noise in his/her system - bloody ethernet cable (which probably violates all the physics law - otherwise why it is not dumped by above mentioned balanced connections?) And now he/she is in need to replace it with another miracle "infinite category" Ethernet cable from Nord Quest Monstous Transparency company. Yet for some more 10% of average annual income.

OK! Our cost-no-object love-to-the-sound deserves that gift!!!

But wait! What about your mobile phone? Have you dropped it no less than 10 miles away from your sweet spot? NO!? But don't you know that mobile phone is the device INTENDED to emit radio waves? This would spoil all that brilliant noise reduction heroism you've posted so far.

Now NQMT company is coming to the rescue!!! They offer special "Phone Housing Audio Purifying" box. And don't think that this is a mere rusted Faraday cage. No! It is cryogenically treated monocrystal formed smart filtrating cabinet. For how much? Well, price is not announced yet, but it is expected to be in the range of 5% of average annual income.

ENJOY YOUR EM SILENCE!

PS ready to collect royalty for this business idea

Michael Lavorgna's picture
Your sarcasm has me a bit confused. Are you suggesting that there's no need for Cat 6, Cat 7, or STP Ethernet cable?

Since you have "some experience in ultra weak signal detection in extra noisy environment" you are in a position to offer real useful information, which I'd be very interested in hearing about, especially in terms of the subject at hand, which I'll remind you is about more than Ethernet cables and EM.

Frankly sarcasm suggests that you may not have anything of value to offer.

llminsk's picture

There were a lot of reasonable professional opinions expressed in your blog already. I do not position myself above those mature authors.

Cat 6 is questionable for audio application, it is intended for high speed data connection. Cat 7 is surely excessive unless you are going to decorate your rig with the whole 1000ft wire bulk (sorry sarcasm again) To get rid of sarcasm completely we need to switch to engineering calculations and use the language of formulas not opinions. And that's what one can find in professional literature.

And I mean EM influence of an Ethernet cable onto audio circuits. Speculations regarding data transfer over Ethernet are irrelevant for home audio. UPNP AV and similar proprietary (like squeezebox) LAN media streaming protocols have nothing to do with Internet media streaming and technically are not streaming but data packet transmissions with all integrity checking in place. As well as in file transmission if your computer reads file from network storage while playing music.

Balances interconnects could be required if you have more than 10ft length for line level signals (keeping aside the input stage ideas like dual-differential topology - questionable by itself for audio grade applications) And in fact depends on EM pollution level. It is funny to have WiFi router nearby but fight Ethernet cable emission!

Where balanced connection truly wins is phono cartridge output. But how often could one see balanced wiring here? Even for premium brands. But line level outputs would be balanced for sure! And how often in-tonearm-wiring (if one gets there :-) makes an engineer bleeding in fact!!! Let's suggest sister blog Analog Planet top post a questionnaire.

And finally. As you excellently put in another post : "The moral of this story is to simply enjoy both for what they are."

Let's read "both" as "any Ethernet category" here.

Michael Lavorgna's picture
If I'm reading what you're saying correctly, Cat 5 (UTP or STP?) may very well provide enough protection from any kind of noise for use in home audio. I'm not so sure I'd agree with this since it completely depends on the environment and Cat 6 and 7 Ethernet cables simply provide better protection from noise. Since you can buy Cat 6a and Cat 7 cables for a few dollars more than Cat 5, why not?

In terms of balanced cables, the benefits of running balanced offers additional benefits above and beyond the cables ability to reject noise. So as I said, running balanced where you can strikes me as good, general advice. As we saw with the Rega shielded ICs, which are only a meter long, cable length had nothing to do with picking up noise radiating from my hard drive. It's also interesting to note that this noise made its way through the Impact Twin A/D, through the FireWire cable, and into my MacBook Pro.

As far as "any Ethernet category" goes, I'm going to stick to my advice which is if you have a networked audio environment, try different Ethernet cables. Just make sure you can return them if you find they do not offer an improvement.

llminsk's picture

Well, now I am heading for "completely non-sarcastic reply" badge from you. "slightly humorous" is ok :-)

To accomplish that I have to confess that I run Cat 6 cables across my house. But I did so to be future-proof and it is about "household backbone". In any case STP is preferable over UTP and higher category is better. If we talk about "materially" reasonable increase in pricing - YEPP - use the best available.

Let me please remind that Ethernet specification requires the connection to be transformer coupled (for several reasons) which provides very high noise rejection together with twisted wires.

All this buzz with shielding is mostly targeting INTER-wire crosstalk and cross-cables crosstalk provided we have really LONG parallel cables and really HIGH transmission rates (both well beyond audio universe).

Consideration of the Ethernet wire as receiving antenna is a ... oh I am struggling for non-sarcastic badge, so I'd rather stop here!

Michael Lavorgna's picture
But I won't make a Cat 6 joke at your expense ;-)
Let me please remind that Ethernet specification requires the connection to be transformer coupled (for several reasons) which provides very high noise rejection together with twisted wires.
Yes and the higher the Cat number, the better the impedance matching and the better the EMI performance of the cable.
In any case STP is preferable over UTP and higher category is better. If we talk about "materially" reasonable increase in pricing - YEPP - use the best available.
Exactly. I'll leave what represents a materially reasonable increase in pricing and "the best available" up to the buyer.
Tom Rb's picture

Michael, your theory just isn't electronically sound (heh) and it's simply not how digital audio systems work.

The signals in an ethernet cable are analogue. Therefore, they are subject to noise. No one has (that I've seen) denied this on your site. However the problem lies in the fact that analogue signal never enters your audio system - it goes no-where near, electronically speaking.

When data crosses from your source to your DAC, the following happens:

A network card receives data from the system. It converts this into an analogue signal and outputs it through the cable.
The receiver interprets the analogue signal and converts it back into data. CRC-checks ensure that the interpretation is mathematically identical to what was sent.

Everything after that is digital. It probably (depending on the playback device) simply transfers the whole file in the first 2-3 seconds of hitting play and runs it from a local storage buffer, or buffers a certain amount into RAM before beginning playback, meaning the cable itself is only used a fraction of the time anyway.
In any system, you're going to have a CPU, RAM and storage medium of some sort doing the actual data reading and interpretation. These are needfully electronically isolated from the ethernet port (and anything else) and are responsible for reading the data, and passing it to the DAC. The DAC is then where we move back into the analogue world. The entire output from the DAC is going to be based on the CRC-checked data fed to it by the CPU/OS, and nothing else. The quality of the DAC itself, and nothing else, is gonig to determine how closely that output matches the intended output.

There simply isn't a route from ethernet cable to DAC output, which would be required for any electronic frequencies to pass into the output. The 'route' has two layers (at least) of isolation. First when the signal is interpreted back into the digital realm, and then when it is read as data from the storage medium.

Ultimately what you postulate is original research, since your therum is contrary to the significant amount of human knowledge we have on electronics. People rage at you because people who come up with theories and willfully refuse any evidence to the contrary, or to prove it in any way, shape or form, to be fraudsters and charlatans. You're very likely neither, but you can see where people's frustrations lie when you tell them they should respect your original theory, but can't or won't do anything to prove it's viable in any way.

There's a good reason why people can trivially pass non-blind listening tests, but no one has ever managed to do a double-blind test successfully, (and there's motive - not only the infamy of proving it can be done, but a prize has been offered for about 7 years which stands at $1,000,000USD .. and that's for analogue cables, digital would be several orders of magnitude harder) and there's nothing technical about it. It's just cognitive biases at work.

Michael Lavorgna's picture
...in the There's no such thing as digital article where you'll find people disagreeing with this:
The signals in an ethernet cable are analogue. Therefore, they are subject to noise. No one has (that I've seen) denied this on your site.
Not to mention the forum comments elsewhere. To your points on the substance of my article...

I am talking about noise, not data, and how cables can allow noise to enter our audio chain. This has nothing to do with, well, everything you talk about as far as I can tell.

I shared an instance of how noise can get into a shielded cable, travel through an A/D converter, over a FireWire cable, and end up directly affecting the noise level in my MacBook Pro. While the data will be just fine up to the MacBook, the output of the audio signal from the MacBook Pro will not be fine.

I also include an example from "CG" in the article that you may want re-read as it directly addresses what we're talking about for Ethernet.

People rage at you because people who come up with theories and willfully refuse any evidence to the contrary, or to prove it in any way, shape or form, to be fraudsters and charlatans.
To date, the only objection I have gotten to this article, as well as everything I've written about Ethernet cable, boils down to bits are bits. Since I'm not talking about the corruption of data, I fail to see the relevance of this argument. This objection seems to be rooted in a belief system, not science. That belief being something along the lines of Ethernet is for all intents and purposes, perfect. Which raises the obvious question of why error correction schemes were needed in the first place.

And frankly the level of outrage implies that there's much more to this "rage" than a disagreement over whether noise can a) get into an Ethernet cable (it can), and b) can this noise get into the audio chain (I say it can). I have no idea how that simple idea can cause rage. Perhaps you can share some insight here.

As far as I'm concerned, there is nothing in what I have said that contradicts any existing theories.

There's a good reason why people can trivially pass non-blind listening tests, but no one has ever managed to do a double-blind test successfully, (and there's motive - not only the infamy of proving it can be done, but a prize has been offered for about 7 years which stands at $1,000,000USD .. and that's for analogue cables, digital would be several orders of magnitude harder) and there's nothing technical about it. It's just cognitive biases at work.
In my opinion, you weaken your position by falling back on a theory, "It's just cognitive biases at work" that cannot be proven. I've written about my feelings concerning blind testing which I believe are not relevant to our appreciation of listening to music on a hi-fi since they create an unnatural listening condition.
skateman's picture

I am talking about noise, not data, and how cables can allow noise to enter our audio chain. This has nothing to do with, well, everything you talk about as far as I can tell.

Fair enough, but could you please clarify your position regarding the claims made in several reviews available on your website, such as:

The JCAT Reference USB Cable:
http://www.audiostream.com/content/jcat-reference-usb-cable

The Synergistic Research PowerCell 10 UEF and the FEQ-PowerCell Equalizer:
http://www.audiostream.com/content/synergistic-research-powercell-uef-an...

+ And many, *many* more

I am reading unsubstantiated allegations of soundstage improvements by using higher end cables here, it would be interesting to inform us readers what your take is on this.

Thanks!

Michael Lavorgna's picture
Do me a favor "skateman". Print this out, stick it on your monitor, and when you think of asking this same question yet again, just re-read this response which I've now shared with you for the third time:
Every cable is part of a system. Therefore, changing cables can certainly cause changes in the sound of that system. Cables in and of themselves have no sound.
As far as "unsubstantiated allegations of soundstage improvements by using higher end cables", this isn't a courtroom, "skateman". It's an audio site where we talk about listening impressions. If you find that this approach does not meet your interests, then I'd suggest you simply stop reading.
skateman's picture

Every cable is part of a system. Therefore, changing cables can certainly cause changes in the sound of that system. Cables in and of themselves have no sound.

My apologies if I asked the same question yet again, but now that I have read your answer for the third time, I still think it is not answering my question.

e.g: it's not because something is part of a system that it can affect sound.

Michael Lavorgna's picture
The other thing you may have to do, beyond reading my answer for the third time, is think.
skateman's picture

The other thing you may have to do, beyond reading my answer for the third time, is think.

We can agree to disagree using rhetoric as gentlemen instead of personal attacks can't we?

Michael Lavorgna's picture
...in our thought-provoking time is that we are still not thinking." Martin Heidegger

I've always loved that quote as well as the essay it's from.

We've gone back and forth with you asking questions and me trying to answer them. From my point of view, this has proven futile. If you'd like to continue this conversation, it's time for you to put forth what you seem to have been skating around, but never say outright.

In others words, if you have a point you'd like to make, then make it. Otherwise this conversation is over.

Tom R_'s picture

"you weaken your position by falling back on a theory"

Are you being serious? Cognitive biases in research have long (long, long) been recognised as a very real thing. Known simply as the placebo effect, it can be measured in nearly every trial where you measure biological feedback, even on Cancer, epilepsy and other areas where you couldn't possibly hope to see the mechanism. This is why, it is absolutely essential double-blind trials are undertaken to ensure there's an actual effect happening.

*You* are the one with an unproven theory: that the placebo effect can't happen if you think you're an audiophile, and that every observation must be 100% true. Compounding that, it is your original theory that the nature of double-blind trials themselves hinder the outcome, despite 200 years of research that says it does not.

Homoeopaths pull the same flim-flam reversal; they claim double-blind trials can't disprove their fake medicine, yet talk in circles when asked why actual medicine can.
Many trials have pandered to audiophiles claiming reasons why they can't double-blind, such as taking place in their own home, etc. Unsurprisingly, none have confounded expectations and been passed.

There's a reason why you can, no matter how "clinical", double-blind things like:

- different CD masters
- bitrates on lossy formats
- 24/96 vs 16/44 CDs
- DACs
- Speakers
- Headphones
- Amps
.. which are all worth spending money on, and can't DB thing like:

- Power cables
- Power conditioners
- Speaker cables
- Ethernet and other digital connects (USB, HDMI, etc)
- Magic boxes that filter cosmic rays
- Directional cables
- Magic magnets
- Sticking your wires on expensive blocks
.. which are products sold to people who don't know any better.

I wouldn't change audiophiles for the world though, not really. There's nothing that fills me with more schadenfreude than seeing pages and pages of comments of people who spend loads of money on something, told the world how amazing it is, only for it to turn out to be a regular USB cable in a hosepipe, or any of the other amazing scams that have passed audiophile critics' desks.

To some other points:

"That belief being something along the lines of Ethernet is for all intents and purposes, perfect. Which raises the obvious question of why error correction schemes were needed in the first place."

Well, it's the error correction that ensures it's perfect. Why is this bad? Nothing like 'begging the question' though eh?

"b) can this noise get into the audio chain (I say it can)"
It just can't. Tiny voltages just won't pass through a system in that way, if they did every part of your system would simply fry because, in effect, what you're describing is an electrical short-circuit. You're saying even small currents have a pathway from the ethernet cable, through the network card, through the mainboard, through the CPU, through the RAM, through some real-time abstraction layers, is stored somehow, and then passed through into the DAC, which then incorporates it into the output.
If tiny levels of RF were passed through the system in this manner the CPU would simply not function.
Luckily, electrical engineers know what they're doing and each part of the system is isolated. Microcircuits literally wouldn't work if this were not the case. I don't know how you can just chose to ignore the entire design of everything in the chain when you reach your conclusion.

What you heard, was the placebo effect telling you it was better. This is why, when you know which is which, you can sit there and swap them, talking about "night and day" differences that are "immediately apparent", but as soon as someone offers to swap them for you, you suddenly can't, even when offered $1mil dollars to do so.

Audiophiles are fun people :D

CG's picture

"It just can't. Tiny voltages just won't pass through a system in that way, if they did every part of your system would simply fry because, in effect, what you're describing is an electrical short-circuit. You're saying even small currents have a pathway from the ethernet cable, through the network card, through the mainboard, through the CPU, through the RAM, through some real-time abstraction layers, is stored somehow, and then passed through into the DAC, which then incorporates it into the output."

Except for the abstraction layer bit, you've pretty well described the system in detail.

(Abstraction layers have to do with the content of the data, not any electrical connection. You could send nonsensical data or bit perfect data and the noise would still be conducted)

It's not exactly a short circuit, since the impedance is not zero or close to it. That doesn't stop the current. This is all well described by Ohm's Law as well as Thevenin's Theorem.

"If tiny levels of RF were passed through the system in this manner the CPU would simply not function.
Luckily, electrical engineers know what they're doing and each part of the system is isolated. Microcircuits literally wouldn't work if this were not the case."

If the noise currents are large enough, that is exactly what happens!

Unless the design was executed poorly, the noise levels are low enough that the digital circuitry performs well enough. That's one of the advantages of digital circuitry - the noise thresholds are higher than in analog circuits. Still, designers, both at the pc board level and at the chip level have to deal with such phenomena as "ground bounce" that is really digital noise. (https://www.fairchildsemi.com/application-notes/AN/AN-640.pdf)

"I don't know how you can just chose to ignore the entire design of everything in the chain when you reach your conclusion."

Again, that is exactly the point in these discussions!

The analog circuitry is part of a multitude of current loops, either explicitly obvious or subtle. Like it or not, it's the entire system that matters. Analog circuitry will amplify whatever noise currents pass through to varying degrees.

Finally...

"Well, it's the error correction that ensures it's perfect."

That ensures that the data comes through intact, even in the obviously noisy environment. Why else would there be error correction? To correct for errors caused by noise! The error correction does nothing to get rid of noise currents - it's a data processing approach that corrects for their presence. Correcting bit content by using redundant data or by resending defective packets does not get rid of the noise. How could it? It just changes the data stream content passed to the next stage.

What needs to be done in a mixed signal system is to effectively isolate noise currents to keep them away from the sensitive analog sections. How that's done is too involved to discuss here. The challenge in an audio system is that in most cases it all is a pretty much random collection of equipment with varying interfaces all connected through a mishmash of wiring including the AC mains.

CG's picture

One more thing...

"What you heard, was the placebo effect telling you it was better."

You may be right. The placebo effect is very real. No doubt about it.

If an interested observer is predisposed to want to hear a difference, there's a strong likelihood that he or she will.

But, it works both ways.

If an interested observer is predisposed to want to not hear a difference, there's a strong likelihood that he or she will not.

What you really need is disinterested observers with no predisposition of any kind. Unfortunately, when you use these observers you also discover that the reason they are disinterested is that they don't care. When you do that, you settle on results that often is more indicative of people's lack of interest than in finding valid data.

Double blind testing seems like a great idea for this, but it has its own challenges, too. It's very hard to perform, for one thing. Medical studies cost a ton of money, even if you factor out all regulation stuff. And, they are just as susceptible to predisposed observers as well.

But, forgetting that part, what is done in medical studies is long term observation. They don't just give somebody a blood pressure pill, measure their blood pressure, write down the collected results, and send the subject home. That's obviously dumb since most medication takes a while for the body to respond. Beyond that, it's not exactly a secret that when people are being measured for blood pressure, they often are nervous which in turn raises their blood pressure. So, you need to perform more long term screening where that nervousness over the results is lessened.

Is that done with audio tests? Not that I've ever heard of. People are pressured, subtly or on their own, to perform right then and there. Does that not affect the outcome?

I bet a million dollars on the line would add to that pressure. That's why it's a sucker's bet. Especially since the person or organization putting up the money has a strong interest in keeping that money.

Michael Lavorgna's picture
;-)

While I look forward to your response to CG's comments, I wanted to offer an additional correction.

*You* are the one with an unproven theory: that the placebo effect can't happen if you think you're an audiophile, and that every observation must be 100% true. Compounding that, it is your original theory that the nature of double-blind trials themselves hinder the outcome, despite 200 years of research that says it does not.
I have never said any such thing. What I have said regarding blind tests for audio is this:
"...taking part in any kind of blind listening test necessarily creates an unnatural condition, one that we never encounter when listening to music for pleasure. I am not saying that science doesn't matter, I am not saying that measurements of hi-fi equipment have no value, or that blind testing is not a valid approach to remove perceptual bias. What I am saying is that listening tests of hi-fi gear at best tell you about the listening capabilities of the people taking the test under those specific testing conditions. That's about it. At their worst, listening tests mask real sonic differences."
While you seem to think that smugness is a trait worth showing off, I'd suggest you'd be better off making an effort to learn about something you clearly do not understand.
skateman's picture

Everything after that is digital. It probably (depending on the playback device) simply transfers the whole file in the first 2-3 seconds of hitting play and runs it from a local storage buffer, or buffers a certain amount into RAM before beginning playback, meaning the cable itself is only used a fraction of the time anyway.

This is where I was trying to get Michael with my 'inane' question about push vs pull but he didn't really get it :/

Michael Lavorgna's picture
The point I was making is that its irrelevant.
skateman's picture

So if I understand you correctly (finally), cable quality affects the sound just by virtue of being plugged to the system. This phenomenon, I imagine, also affects the sound when playing files stored on the local hard drive.

is that right?

Michael Lavorgna's picture
.
fritzg's picture

I was one who pushed on the notion that cables impact soundstage, etc. on the JCAT review. I get your response here, Michael, and it makes sense.

Quote: Every cable is part of a system. Therefore, changing cables can certainly cause changes in the sound of that system. Cables in and of themselves have no sound.

I'd like to suggest that future reviews of cables and maybe other peripherals contain direct comparisons re the soundstage between two or more cables when swapped out so folks don't get the impression the reviewer believes the cable itself has those sound qualities.

Michael Lavorgna's picture
I have found that cables and their affects on a system can vary depending on the system and environment. We can say the same thing about speakers, amplifiers, preamps, and so on, more or less. The easiest example being an amplifier will "sound different" depending on the speakers it's driving. Yet we typically don't seem to mind attributing things like an improved soundstage to an amplifier. Interesting.

Every once in a while we review something that really stands out to us. I'd use the Auralic Vega as an example of DAC that really impressed me, and continues to do so. If you read that review, you'll see I do not offer the caveat of "listening for yourself" as I do with many other DACs because I feel the Vega's strengths will be apparent in most any system. Sure that's a guess, but I've heard the Vega in a number of very different systems and I was very pleased with what I heard.

My point being, and sorry for the long-winded response, we can feel the same way about a cable's performance. I would say that Steve felt very strongly about the JCAT cable's strengths so much so that he felt as if they would be easily heard in most systems. Therefore, the things he heard like an improved soundstage, would also be heard by others in their systems.

audioacct's picture

Perhaps, Michael, you should take a dose of your own medicine and a) not troll for negative comments with pointless articles like this, and b) not escalate conflict with commenters. It takes two to tango, and you're obviously quite up for a vigorous dance.

Michael Lavorgna's picture
;-)

I am awaiting a response from "skateman" to see exactly what he's driving at with his numerous questions and at times pseudo legalese language. If I am wrong with what appears to be obvious, I will apologize. Otherwise the claim that he is not thinking about what I've written is accurate.

As far as trolling goes, I'll have to disagree with you. This article is hardly pointless, it is in fact kinda pointed in that it obviously irks people who only want to talk about data. It also offers an answer to a hotly debated question.

If you've read the comments here, someone has already used this information, tried a few things in their system, and arrived at what they feel to be better sound from their hi-fi. As far as I'm concerned, that makes this article not only not pointless but certainly much more helpful than people who cannot think outside of the bits are bits box.

skateman's picture

I am awaiting a response from "skateman" to see exactly what he's driving at with his numerous questions and at times pseudo legalese language.

I'm not driving at anything, just trying to honestly find out whether you are in disagreement with other reviews of this website assessing that cable quality affects the soundstage of music, improve the basses, etc

I only provided those two links as a reference of allegations of soundstage improvement without making any comment.

Michael Lavorgna's picture
That you do not accept my answer is not my problem. Simply repeating the same question over and over and expecting a different answer is something you may want to give some thought to.

I also provided the answer to your push/pull question in my reviews of the Ethernet cables where you can find all of the details of the system context that you need to answer this question for yourself. Even though as I've pointed out this question is irrelevant.

skateman's picture

I also provided the answer to your push/pull question in my reviews of the Ethernet cables where you can find all of the details of the system context that you need to answer this question for yourself.

Absolutely not, I have just re-read your original vodka cable review, this is the one review that went viral. Here's a sample:

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.

At no point I see any clear precision regarding the way you have been doing your test. If you played back those files stored on your NAS using AFS/NFS/Samba you actually pulled them, if they were however streamed using something like airplay, they were pushed.

If you witnessed the phenomenon of audio improvement using the former method (pull),
considering the fact that unless you have a rather slow network, the full content of the file will be transmitted entirely to your macbook in a second or two, this means that the audio improvement phenomenon will be strictly similar if you play this file locally right? That would tend to prove that the phenomenon is due to the fact being plugged in. It should therefor be possible to measure a change of sound by simply unplugging the cable on and off while playing a local file.

Let me just quote you again:

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.

:)

CG's picture

"That would tend to prove that the phenomenon is due to the fact being plugged in. It should therefor be possible to measure a change of sound by simply unplugging the cable on and off while playing a local file."

Bingo! Ever try that?

The same idea could be applied to the charger of your laptop while playing music. (Assuming you use a laptop.) Ever try that?

skateman's picture

The same idea could be applied to the charger of your laptop while playing music. (Assuming you use a laptop.) Ever try that?

Unless the laptop power-brick is plugged to PowerCell 10 UEF:

http://www.audiostream.com/content/synergistic-research-powercell-uef-an...

CG's picture

You have any idea how these power filters work? In general, I mean. (I have no idea whether this device works as advertised, aside from reading the review here.)

I sense that I am being baited into a supposed trap, so no need to respond.

For anyone else: http://www.analog.com/media/en/training-seminars/tutorials/MT-095.pdf

Michael Lavorgna's picture
In terms of the noise I'm talking about, your push/pull question has no relevance. And yes you are right in that I did not specify in that review how I was irrelevantly pulling. My apologies. While you are correct insofar as removing the Ethernet cable would certainly eliminate any noise coming from the cable, that's obviously not a solution to the problem. I do enjoy it when you quote me to me.

Pages

X