There are no data integrity problems associated with all modern Ethernet Switches - even the cheapest $20 Switches can reproduce data across all ports with 100% accuracy in any home environment
So, the only possible cause for sonic degradation associated with Ethernet streaming is noise of any kind - ie. electrical/electromagnetic common/differential mode, EMI/RFI
Fibre optic isolation guarantees that noise from the Ethernet Switch does not get across to the other end of the fibre.
Those are indisputable real-world objective facts. Thats how easy it is to completely eliminate the Ethernet Switch from any possible impact to sound quality.
I don’t believe for a minute that there are any data integrity issues behind this (unless a fault develops), but what happens if a switch introduces EMI/RFI noise?
As I understand it, Meridian redesigned their ID40 card to work around the issue of noise being introduced - yet I still find that a Network Acoustics filter makes things better on the ID41 card that replaced it
Some great posts here, with thanks to all including recent correspondents here.
I have to say both the quality and tone of discussion here stnds in marked contrast to the abusive tribal exchanges which arise not only in so many other hifi forums but in online discussions on almost any topic. I sometimes find it quite frightening how verbally violent some people can be; I’d like to think this would be different if they met over a beer or a cup of coffee but in many cases I’m not convinced.
100% agreed. I have made it quite clear in every post on this subject that the digital domain works as robustly as it is designed to.
100% agreed. I have made it quite clear in every post on this subject that EMI/RFI are the only issues in this “supply chain” which affect sound quality.
100% agreed.
So…
Many audiophiles choose not to use fibre optic. They stick with copper. They therefore need an alternative solution to addressing (2).
You omit an important step in the chain, which is the media converter required to change the optical signal to an electric one. The only rational reason why so many audiophiles stick with copper vs optical is that optical does not deliver the sonic goods which everything up to the converter promises. There must therefore be issues with some or all media converters which the standards to which they comply do not address sufficiently for audio purposes.
The only rational explanation is that the media converter is adding EMI/RFI noise back in. This could be through some combination of noisy circuitry, poor shielding and/or poor remedial measures eg EMI/RFI absorber.
The Ethernet Switch is an alternative solution. It is not the problem. I’m sure you’re bewildered by how many audiophiles refuse to move to “obviously superior” optical but the real-world objective fact is that they do.
I hate to say it, but they may be in particular areas.
dCS doesn’t believe you need to use high-end clock cables, but using better clock cables very clearly makes an audible difference.
I don’t know why, and don’t care why, but the audible difference is obvious even to non-audiophiles.
Where are you putting the switch in this equation?
If the switch is placed after the fiber to copper conversion, obviously noise is a possible factor.
If the copper to fiber to copper conversion occurs after the switch, it’s hard to see where the switch would make an audible difference… but as you know I would determine that solely by listening.
This also isn’t a perfect situation as its possible noise is being added to the copper connection when the conversion back from fiber to copper occurs. This is why there are a variety of people who have found that different SFPs sound different.
You are absolutely correct in #1 - no one is arguing the data has in any way been altered by any Ethernet switch.
Any equipment connected to the last inches of Ethernet cable going to the dCS device can introduce noise into the system. This could be a switch or just an optical - copper converter. That is why I posted a feature request for an SFP input on future units. That way the electronics (power supply for the SFP converter etc.) can be designed as part of the DAC / upsampler and noise can be tightly controlled.
The way things are today with only Ethernet inputs on dCS gear, there is unfortunately no way to avoid noise on the network connection and one is forced to jump through hoops using linear power supplies and the likes to get close to a good solution.
Here too, care must be taken, because as soon as the signal is converted back to electrical, noise can become a factor.
As I mentioned, on some other boards it’s been found that different SFPs do sound different, not that you need some $1000 audiophile SFP, but rather “name brand” SFPs tend to sound better than generic eBay models.
Why? Because as long as noise levels don’t affect the data stream, the makers of the inexpensive SFPs don’t care, where controlling noise is just something good networking companies like Cisco and Ubiquiti tend to design for.
While this might prove controversial, I have to agree, SFP’s sound different. That said, I wish people would stop thinking it’s because the bits are different, it’s not. That problem is solved.
Data integrity is within the context of EMI/RFI, not in the absence of it. If an Ethernet cable is not properly constructed and not certified, for example if the twisted pair is not properly twisted with one wire being twisted at different angle to the there, then that cable will act as an antenna, and the signal to noise ratio (SNR) goes way down. That’s why it’s important to have properly certified Ethernet cables.
That said, in the home environment, even crap uncertified Ethernet cables will mostly work just fine because the SNR is not bad enough for most Switches/Ethernet ports to deal with it.
Yes, but the amount of noise in a typical home environment is so minuscule, it can’t cause any tangible problems associated with Ethernet streaming unless the Streamer/DAC has really very poorly designed Ethernet port magnetics, and really bad un-certified Ethernet cables are being used.
EMI/RFI is not magic, it’s easily measurable, and Media Converter manufacturer have to meet FCC compliance for EMI/RIFI. A typical Media Converter is many times simpler than any Ethernet Switch - it’s basically 2 Ethernet Ports/Magnetics, and 1 main chip - there’s no way it can generate more noise than the Switch itself. Coupled with a good LPSU, there’s no EMI/RFI generated that could cause any sonic degradation.
$2000-4000 for an Audiophile Ethernet Switch, versus $200 for a Media Converter solution that permanently eliminates the Switch from sonic impact . You do the math
This is precisely an example of where just your subjective opinion is not going to be good enough because it goes against the real-world objective facts of how SFPs operate, their noise compliance, and the layers of abstraction involved.
Thank you as always @Anupc for providing a logical framework for this discussion.
I don’t have the network expertise to debate with you or others on the technical front. However, I will note that unlike dCS, MSB does not recommend their Ethernet network interface (“Network Renderer”) as a first choice, but instead recommends their Pro-USB/fiber connection, which takes USB as input from the music server or computer, and runs ProISL (fiber) out to the DAC, precisely to eliminate network noise, consistent with your explanation above.
MSB ProUSB: “Offers 100% electrical isolation between source and DAC”
Cheers and thanks everyone. I am learning a lot watching/reading this one
This is sidestepping the quite reasonable question. You keep referring to things like data integrity when to my knowledge no-one has suggested here that data integrity is in doubt or that RFI/EMI affect it.
You then talk about poorly designed cables as if, somehow, certified cables are immune to RFI/EMI. Spoiler: they’re not.
Well here’s a thing. I talk about rational, and you talk about magic. I’m not sure why you’re introducing concepts like magic into a rational discussion. No-one is suggesting or claiming that magic is happening here. It’s yet another deflection.
You may be letting your substantial professional experience cloud your thinking here. If standards compliance is all it takes then all media converters do an equally good job, case closed. I’ll ask in a separate thread about SFP modules etc because I’m genuinely interested in the domain and tapping into your knowledge of it, but that might turn out to be a short thread if standards compliance is all we ever need. Maybe the majority of audiophiles, stubbornly sticking with their copper, haven’t even bothered to hear the optical alternative… but somehow, audiophiles generally being both inquisitive and obsessive, I find that hard to believe.
Firstly, you’re clearly out of touch with the audiophile network switch market! How about the Ansuz D-TC, yours for only $16,000. Must be the Tesla coils!
Secondly, if your assertion is correct then no self-respecting audiophile with any system anywhere in the world would ever use a network switch on copper rather than changing to fibre optic. And yet this is clearly not the case. I believe this is because your assertion is incorrect; I do recognise that this is of course an assertion of my own…
“Logical framework?” Interesting euphemism for the “It’s impossible for you to hear that until you prove it to me and give me a technical reason for why” argument.
You’re correct in that MSB does advise using their new interface for just that reason - isolation.
If an “audiophile switch” does indeed eliminate electrical noise as the source of its sonic improvement, it’s actually the more inexpensive solution compared to the MSB solution.
Does a pair of fiber media converters do the same? I haven’t tried it, but as I mentioned above, as soon as the signal gets turned back into electrical, the same concerns regarding noise come into play along with the sonic effects of various SFPs and their “noise compliance” - against what, industry standards? Wow, low-cost products from nameless overseas producers have never done that before.
Of course there is a wide variety of price points for media converters as well, and we already know that the cheap wall-wart power packs they almost invariably use can introduce a fair amount of noise into systems themselves.
So am I spinning my wheels with my current solution?
Cisco switch w/Cisco SFP > Fibre cable > Cisco switch w/Cisco SFP > Ethernet cable to Rossini
Is the second Cisco switch introducing noise thus defeating the gains made by the preceding Fibre connection ?
Should the second Cisco switch be replaced with a Media Converter solution? Recommendations?
Thank you
Hi Bill,
What I meant was that for me, moving the discussion beyond the debate of whether the 0s and 1s are the same (now, finally, apparently settled in the affirmative) to one that focuses on possible sources and noise that can impact the equipment, and potential fixes, represents a step forward in this discussion.
While EMI/RFI can (in extreme cases) cause data errors, an absence of data errors doesn’t mean an absence of noise. Noise which I believe can influence the sound, depending on the design of the DAC.
The designer of the Chord DACs and Uptone Audio equipment, John Swenson, appears to disagree with the assessment that noise from ethernet can’t enter the audio chain and he has written a white paper on this subject, which can be found here.
While it could be argued, as a designer of network switches (too), that he could have some bias, he has clearly thought this topic through.
For our purposes, I thought the geometry of properly certified Cat5e and Cat 6A cables ensure adequate common mode noise rejection. And isn’t ethernet galvanically isolated by design?
It does, but only for the purposes of making sure data can be received correctly at a specific data transmission speed, in a certain type of environment. It doesn’t make the cable immune to noise, however.
Because twisting the wires only reduces induced noise, as the data rate increases, so do the specified number of twists (per metre of cable) and you move from, say, CAT5e to CAT6 (and up).
Ethernet cables are not galvanically isolated by design and they don’t ensure noise rejection… but switches are; “galvanically isolated” may not equal total elimination of noise but it should do a good job of mitigating it. All an audiophile switch designer can do above and beyond this is to minimise the noise entering the case of the switch and minimise the noise generated by the switch circuitry. To use a phrase introduced here by @Anupc, it’s not magic.
