What Ethernet cable are you using into you streamer?

May I suggest that this thread is moving way off topic. It is supposed to be about what ethernet cables people are using. It is now dealing with auditioning and break in regarding all sorts of components and/or manufacturer/distributor/retailer’s trading policies.

This is by some considerable margin the longest thread ever on this forum having had 280 responses so far since December 2019 ( many of which IMO are effectively just rehashes of points that have been posted earlier on, sometimes several times). It is now either going off topic or becoming excessively repetitious.

May I suggest that it is no longer serving the purpose for which it was intended? My feeling is that it is about time it was closed. Any newcomer is unlikely to read back over 280 posts to see if their question or point has already been made. It almost certainly will have been so the chance of yet further repetition is increased.

If anyone wants to then please start a new thread about burn-in or dealers’ loan policies.


Agreed. Forums and threads are hard to shepherd, much as Zoom calls without a clear agenda. In closing nothing beats try and buy, as Bill describes. Especially with dealers who have the capital or manufacturers who are willing do so, or folks like The Cable Company with their lending library. As for Ethernet cables I’ll stick by the Shunyata Omega … until something better comes along.


The ‘Ethernet Switch and Cables Mania’ thread on the Naim forum refused to die at >7K before it was closed by the moderator :roll_eyes:

I won’t be alone in hoping the religious debates and point scoring over there doesn’t reach this place :grin:


So you are in fact saying that those that hear clear differences between Ethernet cables are at least somewhat dillusional. :wink: Still, in my very high-end system anyone with a functional pair of ears clearly hears the differences I am hearing - blind. In my case running between Melco S100 switch and Aurender W20.

I did like the Shunyata Sigma as others did. Then upgraded to CrystalConnect Absolute Dream and recently upgraded to its successor CrystalConnect Monet, which really is a crazy good upgrade considering the fairly low price uplift compared to its predecessor. Interesting fact: CrystalConnect chooses to limit their Ethernet cables to 100Mb (not supporting 1Gb), because they feel this simply sounds better. Clearly this has nothing to do with bits not reaching their destination. It’s all about noise, folks.

Regarding the lack of ‘objective measurements’: as stated, Rossini (hell, even Bartók) measures 99,9% the same as Vivaldi. I chose the latter. Guess why…? :grin:


Your point has been discussed over and over and over again. Anup alone has posted answers relevant to your question 23 times. And that’s just this year !

I am sorry Vincent but if you read the whole thread rather than just the last posting you would quickly realise that you have just added yet another repetitive turn to this seemingly endlessly circular discussion. Please see my comments just 3 posts above.

MODERATOR: I know that you want this forum to be lightly moderated but this thread has run out of original content. If not closed it could carry on infinitely. Neither the objectivists who believe that the design of the cables and how networks work make any claims of subjective cable differences impossible nor those that believe that audible differences do exist ( but cannot offer any explanation) are ever going to change their position. Ever. So round and round we go endlessly because nobody can find anything novel to add.

1 Like

Aaargh again! Wow, mate! Hold yer horses…

You were complaining about posts being off topic and rightfully so. Mine certainly was on topic. Whether you agree or not with my view (my second only on this whole forum…) certainly should not be a reason to close the topic. Should it…?

I’ve read most of the comments - see my likes to many posts.

Thanks for the warm welcome.

Vincent sorry if I sound more negative than is my usual stance but IMO this is just getting out of hand . Not your fault but shared with the other 282 contributors (including me). Incidentally I was not just complaining about going off topic but:

I really do appreciate your being here and look forward to reading your future contributions on other topics but I have already mentioned the problem that occurs with such a cumbersome and repetitive thread as this with relative newcomers:

I appreciate we can offer O.P. opinion to help him make a choice, but given the subject is so contentious, there must be a way we can test and measure the performance of these sorts of products in an audio setting.

Totally appreciate the IMO factor in terms of preference and that’s fair enough but either there is more noise or there is less, this aspect is not subjective.

Would this work, make a test track with a frequency outside of the human hearing 25khz say, play it over ethernet and everything I would hear would be noise from the various components, making it easier to identify a culprit or better performing component?

Or is this a silly idea and unwanted noise only happens when something happens if you know what I mean?

(At the risk of this thread going on for a while longer :laughing:)

Graham, finally, someone asking the right question! It’s not a silly idea, and it’s not that difficult to make objective measurements actually.

In fact, just over 18 months ago, (when COVID19 emerged and holiday plans got scuttled), I found myself with time to seriously poke around to investigate a separate issue - that there was in fact no sound quality difference between streaming via Roon/RAAT vs. Mosaic/UPnP - and by extension, that Ethernet Cables/Switches make no difference either.

Reading this thread, the Ethernet Cable believers seem to imagine that the Ethernet cable connects virtually directly to the DAC output stage, hence any “noise” on the Ethernet makes its way to the output stage thus affecting the sound quality.

Nothing could be further from the truth.

On dCS platforms, there are a number of isolated stages between the Ethernet input port to the final output. Without going into a terrible level of technical detail, hopefully the diagram below shows what I mean. Obviously, it’s slightly different on a Rossini or Bartok (but not that different actually).

Here’s the bottom line; People can claim to hear whatever they like, but objectively speaking, Ethernet cables do not make any difference to the data bits nor to the noise levels at the output on a dCS system. (No doubt they’ll now claim I’m measuring the wrong things :rofl:)


So, if I understand well the schema, after the signal is out of the Upsampler, it is still data travelling on the AES cables.
Then, as it is data, the same data coming in via the ethernet cable, then the quality of the AES cable is not important, and obviously it is impossible to hear a difference between a very cheap AES cable and an “audiophile” AES cable. After all, data is data, am I right ?


Well all i know, is when i swapped one ethernet cable for another one i instantly heard that they
didn’t sound the same.


I spoke with a high-end manufacturer (a friend of mine) here in The Netherlands. His opinion is that it’s not about the data not reaching their destination, but about the transfer of HF common mode noise that interferes with (a.o.) the DAC process in the smallest but audible details. His devices (Grimm) are claimed to be more or less immune to this. However, even then in tests Ethernet cable differences were audible.
Anupc: By the way, I don’t connect Ethernet to Vivaldi (it simply sounds worse then) but only to the Aurender W20.


I am currently trying to figure out which network cable sounds best in my setup. And unbelievably what differences there are, in terms of performance.
Not just the network cable, but
The switch is also a very important link in the chain.
A switch with a good power supply, for the lowest possible, audibly lower noise floor.


What is the feedback from all of you on this?

I would propose with humility we open a separate topic ‘Debate Floor’.

We could begin by listing and summarizing the different views expressed and argued for in past debates (maybe more generalized topics than cable X vs. cable Y). That way we have a central spot to freshen up and keep us from repeating exchanges that happened before.

Then we could debate at heart’s content - always respecting that views are debatable and all participants are in their right minds. This is not to say that we can know nothing and everything is relative. Debates can be won or lost, but that is not the point. The point is mutual learning. We could for example debate the audibility of unmeasurable phenomena. As we all know it’s an age old debate in hifi and also here. Maybe we can generate fresh insight. Maybe not. The role of our consciousnesses in this might be a contender for further exploration.

Another idea for a debate would be what we optimize our systems for. What we look for when listening to the music presentation of our system. What are advantages and disadvantages of using ‘live’ as a yardstick? Which other yardsticks are useful? How does the music presentation make me feel ideally with different genres / artists? Lots to usefully talk about…

Also, with a dedicated debate floor we could keep other threads focussed on the system question at hand. Within the debate floor we can moderate ourselves when we see all arguments have been exchanged and only truly new insights are called for.

Looking forward to your feedback…

1 Like

Trust your ears! The science has yet to catch up with what the brain and its sense “devices” pick up and inform you.


Maybe the science has yet to catch up because we’re missing a few elements of the scientific method:

…like a hypothesis as to why some people hear a difference and some don’t.

…or a hypothesis as to how that difference might have manifested itself.

I’m all for improving things, even if the improvements are all in my head, but until there are some hypotheses to be tested I’ll spend my time and cashish on other things. More power to those who hear improvements, though. I wish I did too.

@whitecube — I like the idea of the other topic. It could mean we could go in prepared: post-cocktail, with opinions at the ready, and — hopefully! — brains open.

Though by the scientific method many wines would taste the same even though at least some oenophiles can tell the difference in the same way some audiophiles with some systems can hear the difference in cables others with other systems cannot.

(There is also lively debate in the wine community whether you actually can tell a difference as well.)

The “trouble” if you will with the scientific method is the hypothesis has to have an unbiased method of testing it and it can give false positives. By the scientific method all expensive cables sound great UNTIL you find a clunker.

We know that skilled sensing can often detect things in a way science doesn’t yet understand. There is ample evidence that dogs’ olfactory systems can pick up on cancer in their owners in some cases, but doctors are still at a loss to explain why and we are left with various hypotheses that medical science has been as of yet unable to test.

There is much about the brain/ear interface we can’t explain, one of the biggies how sound, audible and infrasonic, affects our bodies’ sense of space and where a sound comes from, less “where’s Sinatra on this recording” and more “where did that tree snap behind me in the woods come from?”

Anupc gave a great block diagram of how Ethernet data goes from the port on a dCS unit to a DAC, but as I have posited that is data based, like the claims Ethernet switches cannot affect sound, but does not address whether say noise on the ground at the Ethernet jack could conceivably affect the ground in the DAC and whether that could conceivably have an audible effect.

If you go back a few pages and watch the video from AudioQuest’s Garth Powell on power, he shows how what EEs consider to all be “ground” can very audibly affect playback.

So no, there is no coming to a conclusion on the subject easily in this thread, all I know is I hear it. It is not price-based, it is not review-based, and it’s not because my dealer is a nice guy and I want to throw money his way or not.

Perhaps I am being fooled - though if sighted listening is bad, that doesn’t explain why I have borrowed and returned far more demo equipment that was Stereophile Class A than I have purchased because it didn’t make an improvement in MY system or didn’t improve the sound for me.

All I know is that for me, I don’t need to know why something changes the sound, but if it does repeatedly and the change is positive, it goes on my list of positive upgrades and may even become a purchase.

Honestly I don’t stream enough to upgrade my cable, but that doesn’t mean I haven’t heard it and noticed it.

As I stated in my original post, I don’t care if others do or don’t hear it, but it’s not very scientific to say that I absolutely CANNOT.

To bring this back to the original topic, the scientific method would dictate there is research to be done on why those that can repeatedly hear a difference in particular systems do.

It’s not even a matter of double-blind testing here, it’s more a matter of giving someone a mix of cables, asking them which they like best and seeing the results. If it turns out Belden 2413 sounds worse than AudioQuest Carbon, why is that? What happens if you put $23 Telegartner RJ45 connectors on the Belden? What happens if you put a $7 for ten Klein Tools Cat 6a connector from Home Depot on bare AudioQuest Carbon Ethernet cable?

I’m curious but I’m not going to do the experiment; perhaps someone else who cares will.


During the period that I worked in a hi-fi shop, I noticed how different listening impressions customers had about the same product. I thought it was because of the different spaces and equipment the customers had at home. But that changed when I was at an Audioclub for a while. Every month a different set in the same listening room that was rented. The differing opinions on how it sounds and should sound… unbelievable. Fortunately, we all agree that dCS is very good. And that the Vivaldi sounds (measure?) much better than the Rossini…

1 Like

Hi BillK,

First off: I highly respect your careful and balanced insights, which some others here unfortunately seem to lack.

Regarding: ‘Anupc gave a great block diagram of how Ethernet data goes from the port on a dCS unit to a DAC, but as I have posited that is data based, like the claims Ethernet switches cannot affect sound, but does not address whether say noise on the ground at the Ethernet jack could conceivably affect the ground in the DAC and whether that could conceivably have an audible effect.’

According to my EE friend and owner of Grimm Audio it is completely clear that noise on the ground at the Ethernet jack does affect the ground in the DAC. It’s common mode noise and IMO it’s a fairly easy to understand explanation why switches (e.g. try a Melco S100 and be amazed) and Ethernet cables do matter and do give audibly different results - at least to those who care to actually listen. Which is what high-end audio largely is about, I would think. :wink:

1 Like

Hi Ruud,

I am pretty sure that, during your time at the audio shop, you have sold me my first serious turntable; a Denon DP-47f. I still have it (although not using it anymore), but I’m still grateful for your help in introducing me to better sounding audio which has led me to this crazy, never-ending journey to high-end audio!


1 Like