Yes, they do.
How do I know?
I can HEAR them…
2 Likes
I do not want to argue with anyone on this topic. Nor do I want to risk $1,000 to possibly gain $10,000 on the bet that has been offered here. What I DO want to do is share my experience, should it be of value to anyone.
As someone with decades of professional background in networking, until very recently I found myself pretty solidly planted in the “bit are bits” camp. But having come across statements from people that I trust, that they heard a difference, I dug into the subject. I will not recount all that I think I’ve learned, but will just say that it looked like Ethernet cables could make a difference, and that the problem was not bits, but possibly other things: noise on the cable making its way down stream, or the cable imparting noise to analog interconnects.
I ordered a 15’ long Ghent Audio ET03 Cat 6a cable, with the JSSG shield. And per the advice of John Swenson at Computer Audiophile, I ordered a Netgear GS105v5 network switch. Also per John’s advice, I made sure that I plugged the ET03 Ethernet cable (going to my PS Audio DSJ DAC) into a socket that had no cables next to it. And - also per JS at Computer Audiophile - I constructed a DIY ground plug to ground out the barrel of the GS105v5’s power supply plug barrel. And then gave it a listen.
Granted, I can’t tell you if it was the Ethernet cable, the new switch, or the DIY PS ground. But the difference was immediately audible to me. But subtle. Very subtle. Something you have to listen for, or do long term listening. But an improvement, none the less.
Realizing that my reaction was possibly expectation bias induced, I invited over a friend of mine that is a true golden ear, for a few hours of AB testing. This is a guy that can distinguish DSD from PCM (of the same material, same master) in three or four notes. I’ve seen him do it in double blind AB testing. His ears are good.
And his reaction was like mine, but more so. I’ll grant the skeptical, our AB testing was not blind. He always knew what he was listening to. So he could be biased too. But I don’t think so, because he described the differences that he heard, and they were exactly what I had heard. And before today, he was firmly in the “bits are bits” camp. But no more. That said, he also stated his belief that 99% of the people out there would not be able to tell the difference. It’s subtle. Likewise that the vast majority of home audio systems would not be able to provide the resolution necessary to hear the subtle changes anyway.
So, do I think others will have the same results? Since I’m not sure “why” it matters, only that it “can” matter, I have no idea what others will find. But from my personal experience, I wanted to share my new found belief that it is absolutely not all snake oil.
Happy listening. 
7 Likes
It’s possible both could be right – your listening results could be legitimate but bits could still be bits. Many people believe that digital cables don’t affect jitter or other data-related issues, but would still be open to the possibility that electrical noise can affect digital connections at the endpoint.
If things are that subtle, it won’t trouble me as I just don’t listen like that and as you say, my system may well not be revealing enough.
From this I surmise that on installation, you may as well get the best you can but after that, stop worrying and just get into the music. If your dancing, you won’t notice lol (Now that has put an image of me dancing in people’s heads! Ahhhh!!! Apologies…)
Seriously, if it’s that important, go for it but I suspect for most, it’s an academic side show although interesting to many.
3 Likes
I hesitate to say too much. But I should clarify… the effect was subtle, to me, on my system, as it’s configured today, in my listening room, in its current state. My home is on the market, so things have been arranged to look good. Not to sound good. And my normally giant (and wonderful sounding) floor standing speakers have been temporarily replaced by some high quality bookshelf speakers. All acoustic room treatments have been removed. And in that situation it was subtle. Other people - with better setups - might hear a profound change. And someone with a more modest setup may hear nothing at all.
And in my original post above, I wanted to make the point that it was possible to hear a difference, and not dwell on what the difference was. While it affected nearly anything you could say about the music, the most critical change was in improved transients. How quickly the attack was on a bass string plucked, or how sharp the slap of a high hat sounded, and how long the ringing of a cymbal was sustained. So while it was subtle, it had a disproportionate effect on my (and my friend’s) enjoyment of the music. I’m never going back. The music is so much more engaging now.
You’re using a PS Audio DAC right? I suppose PS Audio users can trust this:
https://forum.psaudio.com/t/ethernet-cables-yes-im-going-there-i-was-a-skeptic-and-i-was-wrong/4393/10
I agree with that post, too. Ground loop can certainly affect SQ, even if the digital bits are the same. To get rid of this and other potential noise issues, this is why we adopted fiber optic network in our latest product Lumin X1.
Every DAC that is so poorly build, that it benefits from a ‘better’ ethernet cable, I would give back instantly.
3 Likes
I’ve given up worrying about these things and just enjoy my music. I ditched ethernet for wifi and now use a Unifi Unity AP and its rock solid and sounds the bomb on my Naim Atom, there was no change at all in SQ from switching from hard wired except I now have less wires
3 Likes
That’s right Peter.
For PSA DAC owners, no other words are more trusted.
Thanks for chiming in.
2 Likes
Ted Smith is a brilliant guy and he calls things like he sees them. I would agree his words are trusted (at least by me).
I have installed the exact same setup (NetGear switch with grounded DC + Ghent Audio JSSG cable) and have come to the exact same conclusions.
The cost (less than 100 $ for the cable and switch) is very reasonable. I see no reason not to implement this tweak, and see no reason to spend more on fancy cables and Audiophile switches, which can both demand extremely high prices. I’ll save up for a better DAC instead.
2 Likes
Good to hear. Thank you for sharing that.
I wonder how you guys ground the DC. Using John Swenson’s device like this photo from CA forum? Are there any other simpler alternatives?
I tested it with the bare ground wire looped around the DC jack, but it does not hold well. I was thinking of soldering a thin ring that would fit on the jack instead.
Yep, if 7.5Vdc is enough, just buy the new Uptone branded SMPS which is internally grounded.
7.5Vdc works fine with the NetGear GS105 and GS108 that John S has tested and recommended (until his own switch is released).
Interesting. From your description this seems to be more about isolating noise from the packet generating equipment (the network switch) rather than having anything to do with the actual material/construction of the ethernet cable. Hence, is it fair to say that the ethernet cable itself had little to do with the improvements you heard?
1 Like
I will offer an answer, based on John Swenson’s explanations. The last leg of cable going between the switch and your network player should be isolated so as not to pick up external interférences. For that, he has proposed this JSSG technique of shielding.
Whether it has an audible effect is of course source of debate, and we have given our opinion on this.
What is not up for debate is that we live in “polluted” environments, and that the shielding technique is efficient.
Well, do no want to get into a debate, for as I previously stated, I really don’t know why it matters, or which of the changes I made was responsible for the differences heard. But I do want to make sure that it’s - apparently - not just about reducing the noise in that last Ethernet cable. It can also be about that last Ethernet cable not imparting noise to analog components, interconnects in particular.
In my case my enclosed stand requires that any Ethernet cable coming to my DAC run parallel to analog interconnects for a couple of feet. Those interconnects are Kimber BPJ’s, so not particularly well shielded. And the cable I replaced was a standard CAT 5e, which is unshielded, and known to be one type that can kick up a lot of RFI (or EMI). So replacing that with the shield Cat 6a could have reduced interference that had been previously been bleeding into my analog gear.
Again, that’s just speculation. But it appears to be possible.
“Even a man who is pure in heart and says his prayers by night can become a wolf when wolf’s bane blooms and the autumn moon is bright.” Maria Ouspenskaya as Maleva the Gypsy woman in “The Wolf Man.”
1 Like
That surrounding analog equipment is susceptible to RF the network equipment is putting out? Makes more sense than many of the hypotheses out there. I would still try to do a serious, preferably double blind, audition, before throwing money or much effort into any issue you may think you have…
There is a huge variability in RF susceptibility of audio equipment. So you wouldn’t expect consistency in results of various people swapping cables to address a problem of this nature. A preamp I used to use (a Crown IC150) was dead quiet in the hum and noise department but would regularly pick up radio transmissions from passing mobile radios. These days there isn’t much in the ether that could actually be demodulated by an audio circuit, but it could certainly be driven into nonlinear behavior by a strong field.