Interesting.
Like you I just use their Cat 6 myself but I might give Belden a buzz (pun intended) to ask how their floating shield design can pickup interference.
I have an open mind so not ruling it out or second guessing you at all. If I get an explanation (that I can understand !! lol ) I’ll let ya know.
No worries Sean…if I’m in left field let me know. I was going to spring for the Cat 6A cables and BJC pushed me back toward the cheaper Cat 6.
Ground loops are not irrelevant when you have any analog or mixed device (DAC) as part of the loop. In such cases it becomes immensely relevant.
I’ve also seen transmission errors over HDMI because of ground currents. Problem was cable network shield not being the same potential as electricity safety earth (three pin IEC power connector). When computer was connected to a TV with HDMI this problem appeared. Desktop computers are earthed through the chassis to which HDMI connector body is connected. While TV’s are typically floating (two pin power connectors). Since the antenna/cable network coaxial shield is connected to the same ground potential as HDMI connector body, this creates link through HDMI cable’s shield to the computer and to the earth. Causing dirty current flowing through and creating transmission errors on the HDMI…
Regarding AES, I’ve also seen jitter problems because the XLR cable was connecting grounds of the two devices. Modifying the cable to have shield connected only at the transmitting end fixed the problem. Other than ground, AES is using transformer isolated differential signaling. By default, coaxial S/PDIF is supposed to have also transformer isolation and the shield connected only to the transformer, not to the device ground, but there are many devices that do this wrong…
One reason for Ethernet having the isolation is that ground currents and large voltage potentials can be dangerous and also destroy equipment. That is why buildings and such are typically wired with UTP because then you don’t have this problem. No ground connection, only transformer isolated differential signals.
Internet works primarily on fiber, not Ethernet. Long distances of electrical wires have various problems, from induced high voltages to high transmission losses.
Yes, can have a shield which is used an RF filter in areas where there’s likely a LOT of electrical interference, in practice even shield cables are not grounded in the vast majority of installations. So buy either and dont worry about it, it wont make a jot of difference unless youre running your cables alongside a load of 3 phase armoured cables…
So effectively youre agreeing, ground loops have no bearing on ethernet… Internet works with a combination of fibre and ethernet. Correct or not?
Nobody has hinted otherwise in this thread…
The first sentence by Jussi here:
"Ground loops are not irrelevant when you have any analog or mixed device (DAC) as part of the loop. In such cases it becomes immensely relevant."
may be highly relevant to the “sound quality” mentioned in your thread title…
Yes they do have. But usually you would be using UTP cables which don’t have ground connections unless you use PoE.
Internet is about different OSI layer than Ethernet. Sure you can put internet over Ethernet, but the two have no direct binding to each other.
But not when connected via ethernet… Sorry. Its just not relevant.
If you think about “sound quality” of digital transmission links, you cannot only stare at the bits flowing through. You need to look at the entire electrical system where the digital link is part of.
We don’t have digital data feed to our brains, at some point the digital data needs to be converted to analog. At that point, any pollution from the entire electrical system will have impact on the result.
Even if the bits are not affected and there are no digital errors, ground loop affects analog circuitry and therefore SQ.
FYI:
One way to eliminate the noise from the network is to use optical networking cable. That is what we are doing on our new opticalRendu. The opticalRendu is a Roon Ready Ethernet to USB bridge but the Ethernet port is optical. This keeps the noise from the network.
Irrelevant to the digital data flow unless ground noise shifts reference, yes. Effectively if ground loops impacted digital flow integrity we would have lots of issue out of optical fiber connectors.
But the end of the story is the analog part… So, flowing HF noise spreading through the otherwise fairly silent power and ground lines of a well-built DAC and reaching the conversion and final analog elements, can anyone really claim ground loops are never an issue ?
And of course the toslink connectors are not so good either - optical coupling issues, etc… They tend to sound inferior to good AES, right ?
Andrew, I don’t mean to be rude, but a few years ago, everybody was raving over the MicroRendu (“revelation”, “removed a veil”, “opened up the sound stage”), and then everybody raved over the super capacitor LPS-1 power supply for the mR (“revelation”…), and I bought them. About $1,300, as I recall?
And recently, in a debate in these pages, somebody was arguing for the benefits of these isolation devices, but not with the MicroRendu, pshaw, only the UltraRendu brings a real improvement (“revelation”…), but now Audiophile Style (the site formerly known as Computer Audiophile but “Style” sounds better than “Computer”) is gushing over the OpticalRendu (“revelation”…).
That upgrade path is about $5,000, right? Skipping the Signature, which by itself is $3,000+.
And when I mention that with a modern DAC and a well-designed Nucleus I don’t notice any difference from a direct-USB connection, people say they won’t trust my hearing, or anybody else’s either, you have to listen for yourself. And when somebody brings up measurements that fail to notice any difference, people say that measurements bear no relationships to real-world experience, why do you even bother, just listen. And listen for yourself, don’t trust anybody else. So you have to buy one of everything and listen.
I agree we each have a responsibility to stimulate the economy. But jeez…
This thread is about isolation noise from the network. Network noise is real and using optical networking eliminates network noise.
It’s great you have your system to the point where it sounds the way you want. Many others are on a quest for better sound.
Margo can. All that matters is that the internet works.
According to Margo, the analogue output of a DAC doesn’t matter, i.e. is “irrelevant”… even regarding “sound quality” he mentions in his thread title… despite some expert advice above…
You actually don’t “have to buy” anything…
Do you feel pressured to buy one by the article on CA Forum? Is that the reason for the “angrily shakes fist at the world” reply to Andrew? 
Statements like this introduce confusion, and they are also irrespectful of the many persons that spend a lot of efforts to improve the analog output of a DAC. If it comes from personal experience it is from a pretty limited one, to say the least!
The only way by which this statement makes sense is that there are far more imperfections in the speakers and the room settings and acoustics, and sometimes in the rest of the electronics. But is is our ear that matters, and a poor or poorly insulated analog stage can ruin the result whatever he format. And a system is a chain of elements that should convey and amplify a signal without altering it, but no part of the system downstream will reinvent a signal that is lost on the source side. And the analog end of a digital is a place where issues that don’t affect the digital part might have a lot of influence.
In a typical DAC implementation SPDIF is slaved to the source clock. Toslink conversion introduces more jitter, that’s why it was common for audiophiles to find that coaxial sounded better than Toslink. AES typically sounds better than coaxial. (With packetized audio from Ethernet, the DAC is not slaved to the source clock.)
Which sounds better become a comparison of what is less harmful in a particular setup - jitter, noise, ground loop, RFI/EMI, etc.
Sean, both you and @agillis think this about me, and my system, and what I buy.
It is not.
I am concerned (as I have written many times) that we give bad advice to the users. To newcomers, specifically — people with lots of experience get to fend for themselves.
Everybody in the press, and the web sites, and of course every vendor, and everybody on this site except me and @danny, are telling people that they have a serious problem with noise transmission over USB. And when I question this belief, ask for evidence of the existence of the problem, and evidence that the purported solutions in fact solve the purported problem, the answer is “it is patently obvious”. Measurements are ignored. And the vendors provide no measurements. The result is that people come in here and ask for advice on how to solve the problem, without knowing if they have a problem. I think that is very dishonest.
Nobody in this thread is questioning whether there is a USB problem.
Nobody is open to the idea that the problem may depend on the noise level of the source, and the sensitivity of the DAC.
But all right.
I’ll stop.
Nobody appreciates my advice.
Emperor, clothes, so forth.