I’ve watched this thread with the usual audiophile scepticism. Mine is based on some experience. I have gone down the media converter route. Then I tried a Gigafoil. Both were retired for lack of a meaningful contribution to my audio experience. I then bought one of the first EtherRegen’s. I used the fibre input and suffered through excessive overheating. I added two heat sinks on each side with thermal tape. Better but not great performance. I suffered regular drop offs with Roon. I tried a Bonn and that was inconsequential in its performance contribution. I tried a Waversa Hub3.0 managed switch and it sounded better and was reliable. In the end that’s what I was hoping for. Reliable good sound.
I have tried a few switches over the years, on both my rossini and now vivaldi.
I have new fibre running into the house now and at first i didn’t, but will explain as i go.
So at the beginning i used a cisco switch sitting right next to the router, this had a few different things also plugged into it, like tv’s, etc. Was off the normal phone line into house.
This i found did indeed work well, not only gave me more connections it seams to make the sound better into at the time my naim streamer. When i changed up to the rossini, i never tried it without, as needed it for the other connections, so this is where i come in really.
Next as i had the N10 melco now, i decided to try the melco S100, this for me squashed the dynamics slightly, and made things sound worse. I even tried a better power supply on it, but it didn’t change the way i felt about it, so i sent it back.
Next i tried was the EtherRegen, this i found did inpove the sound slightly, i was using an upgrade on the power supply as i had one.but as others have mentioned it runs very hot, after about 6 months, i decided to un plug it and see how it went, i didnt bother putting it back, as decided i could like with how it sounded with out, and lose the heat, and the power draw this has.
Then i got the vivaldi and upsampler, no switch apart from the cisco still. But after hearing good things about the phoenixnet, i decided to try one, as i saw a secondhand one for sale at about half price. Got to say its certainly does something, i have tried it with and without many times but i prefer it in the system. It just makes things sound nicer, not really sure how to explain it, bit like a clock.
But the phoenixnet is certainly the better one i have tried
5 Likes
Where do you have the PhoenixNET installed? Probably best way of explaining is length of cable from P’net to router and length of cable from P’net to streamer. Thanks
don’t want to steal this…but would a Vivaldi Upsampler gives a benefit to Rossini for network duties (vs internal streaming)?
No problen with posting here but I’m not sure you’ll get the attention you deserve in this particular thread! Might be worth searching for something closer or starting a new thread as it’s a good question.
All the best.
2 Likes
[quote=“T38.45, post:165, topic:3450”]
Quite right. This item is very off topic as this thread deals with ethernet switches. To get a proper response which includes people who may pass over an ethernet switch item try re-posting your question as a fresh topic.
I tested Melco S100, Cisco 2960, d-link, Ansuz and Innuos. My fav were Ansuz followed by Innuos. Innuos took some time to improves it SQ (for me almost one week). I will try the Ansuz again, seems they changed their HW internally so Mosaic detects upnp servics now (but I will implement it in my setup and see if it really works…)
Yes, thanks for the list of switches you tried but I was asking about where Dunc installed them.
How long is the cable from the router to the switch and how long is the cable from there to the streamer?
Let me clarify. The way an audiophile switch (actually any ethernet switch used to improve SQ) works means it is meant to be installed just before the streamer. This way it kills as much noise as possible, including noise on the (usually longer) cable from the router to the switch.
So if, for example, you used to have a 5m cable from router to streamer, you should hear a bigger difference with this setup
Router - 4m - switch - 1m - streamer
than with this setup
Router - 1m - switch - 4m - streamer
Phoenixnet is sitting on my rack.
Then i have the phoenixnet cable running between it and the melco N10.
Then from the player port on the melco, i use a 1.5m long audioquest diamond to connect to the vivaldi upsampler
1 Like
Thanks Dunc. That sounds right.
You’d be amazed how many people spend money on an audiophile switch and then put it nearer to their router than to their streamer.
I did also try running the melco through the PhoenixNet to the vivaldi, but i didn’t like it as much.
Of course I put my switch closer to my router than my Upsampler. That way the +40 foot run of fiber optic cable coming out of the switch prevents any potential nasties from impacting the signal. That cable terminates less than a meter from the Upsampler into an optical module that converts the signal back to ethernet for my Vivaldi Upsampler.
2 Likes
Have you tried putting your switch after the optical module?
The fibreoptic cable won’t transport any nasties anyway so a switch before it won’t really be doing much.
At the other end, some optical converters are noisy in their own right. If you use a switch for sound quality improvement purposes, rather than as a port replicator, you’ll get the biggest impact by having it at the very last point in the chain before the streamer, whatever the cable is.
1 Like
It’s St. Patrick’s Day here in the USA, so there’s lots of conviviality, drinking, laughter, etc. In that spirit of joviality and friendliness, I do have to say Nigel that your “audiophile switch” is sure doing some heavy lifting in your audio chain. The notion that the switch, after a fiber isolation stage, can improve SQ is . . . intriguing. What is it doing, and how is it doing that? 
Spooky coincidence, it’s St Patrick’s day here too! 
Really? It’s doing what all audiophile network switches do and stopping any accumulated RFI/EMI noise getting to the DAC (it’s not doing anything digital per se).
If you have a nice special low noise optical-to-digital converter then that might well do the trick, but most such converters are not as effective in practice as in theory due to the noise generated by the circuitry doing the conversion back to digital. Installed after an optical-digital converter, the only noise a switch will have to mitigate is the noise generated by the converter circuitry of course.
If one installs an audiophile switch at the router end of things, it can stop most/all of the noise from other (non-audio) devices getting into the chain. But if I had just one switch in my system, it would be the last link in the chain. And is.
Cheers!
That’s not true at all.
While it’s true that any Ethernet Switch with a long unshielded Ethernet cable is likely to pick up EMI/RFI, and so should be close to the DAC with a short Ethernet cable. However, virtually any Media Converter, with a proper low-noise Power Supply, and any normal to-specification optical SFP - CISPR 22/FCC Part 15 compliance (look it up) - close to the DAC, with a short unshielded Ethernet Cable, will be the least EMI/RFI polluting solution, period.
3 Likes
Obviously “period” is not exactly an invitation to reflect together on the mechanisms at play and explore this truth thing… but let me build on your response.
I presume you see no place anywhere in such a system (including at the router end) for an audiophile ethernet switch then, because the optically transported digital signal is as purely digital as it can be. Correct?
Agreed 100% about the importance of the low noise power supply. I said “most converters” because one should be as picky when choosing one as anywhere else in the system.
Have a great weekend
Well Nigel, the reason for “period” is that it’s not just some random opinion, it’s measured fact.
A typical Media Converter is far simpler in construction than any Audiophile Ethernet Switch. Typically based on a single Ethernet chip with ancillary power-related componentry around it. All the conversion heavy lifting is done by the SFPs.
And SFPs are very precise components; they conform to very strict specifications (in order to be compatible across many different platforms). Those includes EMI/RFI specifications, Thermal and Power-draw specifications, not to mention physical and mechanical specifications.
Within the last decade or so, SFP’s have come down in costs so significantly and preform so precisely to specifications that Audio vendors are now starting to incorporate Optical SFP transceiver cages directly into their platforms; Linn, Lumin, Melco, MSB, and I expect we’ll see more in the coming years. No audio vendor in their right mind is going to do that if SFPs generate more EMI/RFI than a standard Ethernet Switch.
So, to suggest that a Media Convert/SFP generates more EMI/RFI than an Audiophile Ethernet switch is, frankly, just the usual ignorant audiophile B.S.
3 Likes
Really indeed.
I’ve owned/tried several “audiophile” switches. Not a one did a thing in my system. I had fun playing with them. And I had fun cooking eggs on both my EtherREGENs—just kidding, but they do get hot, and occasionally go into thermal shutdown—but not a single switch I tried made any audio difference. Maybe I’ve already got good noise isolation in my system, and there isn’t much to do. Maybe. Or maybe Ethernet combined with good Ethernet circuitry just performs very well on dCS equipment. [There’s probably a reason it’s their preferred input.]
Now, I didn’t go to the lengths that ASR did (see caveat below) in this review of the EtherREGEN, but I was reasonably thorough. The ASR review is worth reading, because one of the things they specifically measured was noise and noise reduction. The TL;DR: the EtherREGEN had no effect on noise (or anything else). So, unless you’ve got some measurements that prove that wrong, I would suggest it’s safe to say not all audiophile switches reduce noise. Granted, I haven’t tried them all, and I certainly haven’t read reviews of all of them. But I’ve been using network audio since 2013. I’ve had an opportunity to hear a few. All I’m asking is if there is any actual evidence, not even proof, just evidence that audiophile switch noise reduction is anything more than marketing hype or audiophile speculation. If there is, I would love to see it. Seems to me that, for anyone with the skill and equipment, it should be easy to measure.
P.S. Note the usual disclaimers: if you hear it, it’s real in your system. I’m not dissing anyone’s perceptions. You’re free to spend your money any way you want. I know some people refuse to consider anything from ASR; I occasionally find myself in disagreement, but I usually find it educational.
3 Likes