Valley of "audiophile" ethernet cables

Hi mate

I genuinely love the passion shown by you and others, on this topic. But this topic is not one of life and death, fortunately.

I quoted your comment that hinted that the little transformers in ethernet isolate against everything - if that’s not what you were hinting then I sincerely apologize in advance.

My linked posts are just to someone who is highly qualified and respected, who is doing his own personal research at the moment and finding that things like leakage currents “sail right through” those little transformers.

I just love learning more about this hobby that we all love. Maybe things aren’t as we’ve all assumed all along - like this topic of the transformers in etherner cables. So I follow peoples like John’s personal research really closely. And for fun.

I just like the learning and the science. And I wish everyone could disagree in a more friendly manner, like me and my mate @xxx have disagreed often but still joke here and there because this is all about enjoying music at the end - not life and death. It’s supposed to be fun. (PS Slim: I haven’t forgotten, you haven’t left the dark side just yet… I told you it wouldn’t easy and you’d always come back :smile:)

The topic of blind tests and whether we’d hear a difference is something else altogether and I’m not sure why I’ve been singled out here. I was one of the first in this thread to hint that BJC’s (their Belden 10GX series cables in particular) are all I really need.

1 Like

Ya I wish the points could be made in a friendlier manner. Frankie had the answer all this time:

We are talking about copper interconnects. When you bring up leakage currents in this context my mind immediately goes to the thought “is the poster stating that one copper RJE is going to filter out those leakage currents more than some other in spec copper based RJE”.

John thinks:

  1. Most boutique Ethernet cable manufacturers are clueless (Arstechnica sending out the $350 one meter AQ Vodka out for CAT6A validation were it passed only in the margin, so not a clear pass, bears this out).

  2. Ethernet cables aren’t going to make any audible difference

  3. If they do, and he’s skeptical about this, it’s by sheer randomness.

Worried about leakage currents? Go wireless. It’s affordable, easy to implement, and sounds great.

My first set of speakers were DIY Karlson Horns in 87. There is a lot of outright crookedness in this hobby / industry.

I’m a CCNP, MCSE, 3COM, Banyan Vines, the list goes on. I’m a hobbyist builder and speaker designer. This is my line in the sand. It wasn’t HDMI, it wasn’t USB, but it is Ethernet as I simply am a material expert from punch down to core routing and switching. I’ve implemented AVB and Dante on Cisco/Intel.

I’ll stand toe to toe with any one that thinks RJE’s that meet the spec matter SQ wise. Period.


Hey @dabassgoesboomboom -

Didja see this new shiny object? :laughing:

1 Like

I did Slim! I got the Uptone version, Iso Regen. Very similar but the iFi doesn’t need an external power supply which is cool. I’ve given my Iso Regen to my old man (my dad in case anyone else is reading this) because it’s cured the pops and crackles with native DSD256 from his Pi3, like we discussed in the iDAC2 thread.

Go on Slim, hit the order button. I know you want to :blush:

Tried an audioquest forest between router and roon core pc. Absolutely no difference whatsoever compared to the standard cat5 it replaced. Totally not worth the 75 euro’s. Ethernet cable subject closed for me.

1 Like

The “little transformers” on Ethernet cables are intended to block common mode RF. They don’t galvanically isolate the equipment.

"Ethernet ports are transformer coupled. "

I am certain this is wrong! You’re saying each of the 4 TP’s is transformer isolated at every port? Crack open a typical cheap 8 port switch, I don’t think you’ll find 32 transformers.

Actually they are. But they are in DIL packages with each package representing a transformer with multiple windings. Other methods can be applied also like the isolation chips so popular in USB isolation gear but the principle is the same.


Yes they are. You’ll find Dual and Quadport packages. Many rated to 1500V

It’s what you think is certain, that simply isn’t.

I doubt you would lose $10,000 except by some awful statistical fluke, an elaborate cheating scheme (that amount of money would certainly make it worthwhile for someone to come up with one, if he could) or divine retribution for deadly sin #4, hubris. But your attitude is certainly not going to make any converts, or lure any takers.

A smaller amount, so as not to discourage someone from risking losing a fortune by shear bad luck, and less arrogance, may get results.

As a recovering audiophile I’ve got a couple of ‘audiophile’ ethernet cables knocking about so I thought I’d pass on my own observations of trying out three different ethernet cables, this afternoon, directly into my KEF LS50W’s (courtesy of Roon’s latest update). The big shootout:

generic 3m Cat6 cable that came with the KEFs (black, suitably dull for a stock cable)

a 1.5m DIY cable from some guy in Germany that I bought off e-bay for £40 (looks pretty, like a snake, with those enormous RJ45 connectors at either end like the AQ Vodka)

0.5m AudioQuest Cinnamon cable that my mate gave me in return for a couple of pints (gold-plated, red/black weave, and with… errr… directional arrows telling me which way to plug the cable in… but isn’t it bi-directional???)

Here’s my view after listening to several tracks I know well… the 3m cable is a noticeably harsher in the highs than the AQ Cinnamon… Nah, just kidding. There’s no difference at all. Seriously. None.


It’s not hubris. It’s confidence in knowing how this stuff works. It’s an informed position. Here is hubris:

" Me: Ethernet ports are transformer coupled. "

You: I am certain this is wrong!"

Don’t try to gloss over this. I’m an azzhole for pointing out people that think they are in the know when they aren’t?


I don’t feel like being tactful with a bunch of know it all’s that:

  1. Don’t really know anything
  2. Can’t find their way to making a slam dunk $10,000 for what is most likely 30-60 minutes worth of effort.

Not trying to start a war here, but I got myself some Wireworld Starlight Ethernet cable, and I like the results compared to the generic Cat6 I had. My perception is that improves the detail and the resolution on my setup (headphones).


Most of WW dealers allow for in home trial, so no risk for you guys to try. If it does not work / improve things on your setup, return it. You’re out only the shipping. PM me for a good WW dealer that gives a very nice discount.

I am now looking at ethernet switches that are discussed extensively over the CA forums.

Again, just my opinion. Please no judging. We all “vote” with our money.

I understand your point. There are good technical papers like Seimons ‘The Antenna Myth’, T.I’s ’ Reducing Radiated Emissions of 10/100 LAN Cabling’. Here’s a real myth buster: UTP Cabling and the Effects of EMI

I’ve also done some proof of concept videos. Here is one demoing a $350 3 meter WireWorld Starlight RJE and 98 meter CAT5e at $90:

The synopsis is that audio continues to play while I swap cables. Even with no connection we hear music. Did the SQ change? Get better, worse?

While many would think that’s a simple question and easy to answer, I still have yet to see a subjective audiophile answer this.


I found this useful. Covers ground loops, antenna effect, cable termination, and baked potato effect (I had never heard of this one).

I’ve been using Cat 6A S/STP I got for around $5-10. Each one came with it’s own measurements. I haven’t noticed any problems in comparison to Cat 5e and Cat 6 UTP I had been using previously.

1 Like

Good read and backs up the other papers I linked.

Some take away from your link:

A shield reduces the effects of induced
noise for signals > 30 MHz. GREATER than 30 MEGAhertz.


Apply 60 Hz AC current
directly onto shield of
F/UTP cable

– Not an induced current
– a placed current
• Test signal integrity on
balanced twisted-pairs
for presence of 60Hz
• No influence
on signal
integrity at

1 Like

You don’t really make your point by linking photos of PC boards with black boxes on them. Why don’t you instead link to a data sheet on one of the devices in question?

You see the repeated series of lego’ish dimension building blocks? Those are Ethernet Magnetic Transformers.

Let me google that for you:

You’ve also been told by Rik that Ethernet is indeed isolated.

1 Like

At least they’re visible (with visible part numbers for googling) - another alternative is the MagJack - the transformer is part of the RJ45 socket so not a visible seperate device.

1 Like