All my streaming is from a NAS. The NAS and the streamer are connected to a switch, in this case the EE8. The EE8 then connects to a little Cisco, which in turn connects to the router, a Google Nest wifi Pro.
It’s my understanding that the files go from the NAS to the streamer via the switch, without going anywhere near the router. Is that understanding correct? If it is, does that mean that in my setup the router cannot affect the sound?
This is obviously different when streaming from Qobuz or Tidal, or when using iRadio, as the files must go through the router.
I’d have thought that the goal of streamer design must be to make it immune to external influences such as routers, cables switches etc. of course, makes of LPSs etc have a vested interest in explaining why these things are necessary.
You may have a point in what you write,.BUT, you should always try in your own system.
This thread is to raise awareness of this,.as well as that I myself want to capture experiences from those who have possibly explored the possible importance of the router for the sound quality.
I know,.and am in contact with lots of like-minded people in this interest, even fanatically interested traders, importers, distributors.
AND,.no one has reflected on the possible importance of the router for the soundquality, hence this thread as I know that many incredibly knowledgeable people here have explored other areas within streaming etc…
Only one of my high-end friends can’t tell the difference between really good switches or ethernet-cables.
Mind you,.only one regardless of which music-system.
My friend has Linn’s absolute latest and absolute best active system,.it apparently has a design that minimizes the impact of switches and ethernet cables etc, exactly what you are on to Innocent_Bystander…
We are investigating this now so I can’t elaborate on this any further at this time.
Below a picture of the military-spec 4-stage grinding fibercable I mentioned earlier.
Many people are open-minded and some sit on the fence wondering if it is snake oil or not. “Military spec”, “NASA spec”, “aircraft spec”, “Mysteron spec” do tend to push fence-sitters away from perhaps trying something.
And of course the military are well known for releasing specs of all of their equipment…
My friend is an entrepreneur in the hifi area,.and is well known as one of Sweden’s most knowledgeable when it comes to the implementation of fiber vs soundquality.
He researches and develops himself…
BUT,.in this with fiber, he is in a very close collaboration with Europe’s most well-deserved research laboratory, and its best research engineers.
This has taken him over three years to develop and arrive at.
He is a dear fanatical friend in hifi, whom I got to know in 1995 when he was an exhibitor at the first Hifi-Show I arranged.
I am only reporting what is happening in the development area,.if some want to limit themselves with names like “snake-oil” etc…
Yes,.I don’t care about that at all, it’s their problem not mine/ours .
I might be totally wrong here but my understanding is that the router forms the ‘hub’ of the local network and as such everything on the local network goes through it. It is the control point for everything on the network as it were. Streamers are configured to be connected to a local network via the router - that is how it all works by convention. As I say that’s my understanding - but I’m very far from being an IT expert so it could be total c**p.
One thing that I do know. My Melco has an ethernet port and a player port. If using the ethernet port then it connects to the LAN via the router. If using a seperate streamer then one has the option to connect directly to that via the player port in which case the router is bypassed.
Even if your files don’t go via the router then my experience is that yes it can still affect the sound. It is connected to your LAN albeit via switches and so it is still able to pollute it with noise.
My Melco is connected to the router via ethernet and an EE8 switch. However, this is for remote control only from the app on an iPad and for information downloading when ripping. The music stored on it goes via USB straight to my Chord Qutest DAC - so no music transmitted via ethernet at all. ( we don’t stream from remote servers). However, despite this and despite it being perhaps counter-intuitive, putting a LPS on the router has a significamt impact on performance - as did the EE8 switch.
Melco do say that optimum performance will be had by disconnecting from the router. Inconvenient - as one then loses remote control. But it does imply that performance is affected by noise from the router despite no music going through it.
PS. One thing might be worth trying. Switch off just your router. If you can still stream your files via your Nova then I guess that answers your question! My guess is that your Nova will simply say ‘no network found’ or similar.
You may be partially confusing the control side with the data side, though elsewhere in your post seems not…
You can have a network with a separate switch (or several), to which the music store and streamer are attached, the wireless router/internet access point connected to that switch (whether ir not via other switches), but none or only a different part of the local network passing through the router. Easily checked by switching off the router, when music should keep playing (and other wired things still wotk like print from computers etc). Different if you choose to use the router as a hub through which the audio data are routed, but I think that is generally not considered best practice.
On the other hand, other than if you use
a streamer’s infrared remote control, music browsing and selection etc are most commonly done via apps on tablets, phones etc, all linked via Wi-Fi ultimately to the streamer/player. That side of things goes through the wireless router, but quite separate from the music data.
So are you saying that if you have a switch connected to the router, but that a NAS and streamer are connected to the switch only, then no music files pass through the router by virtue of having the switch there?
True, but whatever noise they generate, it can be reduced.
I have each SMPS unit clamped with frequency specific ferrites on both AC & DC sides, plus a RA “silencer”. This suppresses noise/ripple as measured by my Fluke o’scope.
SMPS switching noise is typically in the 50Khz-3MHz range.
Although this noise is transmitted on the AC line as well as DC (plus RF), I am more concered with the DC side polluting the switch & wireless hub, and ferrite is the obvious (easy) solution.
Problem is the ferrite alloy mix that works at such low (50k-3MHz) frequencies is very limited and also not so effective as ferrite can be with the higher freq mixes.
I use Fair-rite (brand) ‘75 mix’ (150kHz-10MHz [Max 2MHz]) & with multiple passes of the cable through the the ferrite clamp this both multiplies up the impedance & lowers the effective peak frequency.
I also have a ‘75’ on the AC side together with ‘H30’ general purpose mix for other non-specific higher frequency noise.
Interesting Mike. It seems a very elaborate way to reduce SMPS noise. Presumably you object to LPS’s on sound quality grounds?
The problem is knowing which SMPS’s are good enough. My Melco came with a ‘medical grade’ SMPS but the Plixir LPS (@£500 and which Melco endorse) brought a very significant improvement.
In your experience would adding ferrites etc. to the supplied SMPS be likely to better the performance of the Plixir?
It’s not elaborate at all, remarkably simple IMO.
I have no objections to LPS, it’s just that a well spec’d SMPS can be +/- as good, plus in a smaller (more manageable) package.
Adding ferrite to SMPS does not do anything for performance, it helps suppress switching noise. Weather the noise is audible or not is the question.
Hi …
I don’t know how much I can
tell you without talking more
with my innovative friend.
But below is a small description
from them themselves (the lab).
A Continuous Commitment to Innovation.
We focuses on the research and development, design, and sales of core equipment, general accessories, and system solutions in the field of network communication.
Our world-class R&D center houses research, design, analysis testing capabilities.
We standardize the design and development of communication products and solutions with market demand in mind.
At the same time, our technical experts, sales representatives and partners gain an in-depth understanding of customer needs and feedback, constantly driving our dedicated R&D teams to develop standardized, high-performance and high-quality products to ensure our products are developed for success in real-world applications.
That is the point. How does one know if a SMPS has a good spec?
From a company like Chord Electronics with the Qutest DAC or the Chord Company with the EE8 switch I am happy to accept that what they supply is perfectly OK.
But these are specialist audio companies, concerned of course with sound quality. They will not supply SMPS’s that spoil their products performance. Which is not to say that improvements can’t be had with ‘aftermarket’ supplies but that’s another story and not a route which I’m in any way keen to pursue for various reasons.
The concern for me is more with SMPS’s supplied with non-specialist equipment. In my case that means my BT router and ONT box. I’ve no way of knowing how good or bad these are, except empirically. I’ve already replaced the router SMPS with a MCRU LPS and it’s much better. Perhaps a good quality SMPS would have been better still or at least as good - but how do I know which one? At least a good LPS for audio use is almost guaranteed to sound better than a general use cheap SMPS from BT.
I replaced the medical grade SMPS supplied with my Melco with the Plixir LPS on the strength of various reviews that claimed it gave a very worthwhile improvement and also because Melco endorse it. Again maybe a very good quality SMPS would have been better - but how on earth do I find that? There is little or no interest amongst the reviewing fraternity in SMPS’s other than to say that they all bad news.
Hi …
No,.I don’t believe otherwise ,.I Suspect,that the router may have some significance based on what I wrote in my posts in the thread.
That’s why I opened the thread to possibly get other people’s experiences.
It’s also all about how careful you were when installing and optimizing your music-system.
If,.for example, you have set “Live or Hot” incorrectly from a powersupply to a device, the possibilities of hearing any differences are drastically reduced.
There are many posts in the thread now,.but no one has directly compared different routers under equivalent conditions.
I only tend to believe if the brand publishing so that I can read the spec, then to see the noise/ripple on my ‘scope.
Hence I have more confidence in my FRIWO units. They’re not perfect but so much better and electrically more robust than the Netgear and Cisco units I’ve measured.
I’ll respond to you with the same question…
Why do you think the Chord supplied SMPS is good, because they say so …. or that they advise only to use the supplied unit?
Do you know the BT unit is bad? they too recommend to only use the supplied SMPS.
I suspect product protection plays a good part in this.
…
Below a picture of the military-spec 4-stage grinding fibercable I mentioned earlier.
…
.