"Audiophile Grade" Ethernet Switches - The new generation

[baconbrain]

It's another JS +1 from me.  The link above is for the same post that I linked to in the "clocking" thread.  I think it good that Antoine has provided a link here too, it is very much on topic in this thread.  In fact, I think it offers a possible explanation for Guillaume's "cognitive dissonance" experience when adding FMC's and an optical bridge.  Impossible to prove of course, but it does seem possible to me.

I would lean a bit more in the direction of a termination issue with one (or both) of the media converters when it comes to the dissonance issue. I am quite certain that my similar experience using ethernet was caused by an improper shielding / grounding constellation. Just my objective thoughts ...

[Johnnydev]

It's another JS +1 from me.  The link above is for the same post that I linked to in the "clocking" thread.  I think it good that Antoine has provided a link here too, it is very much on topic in this thread.  In fact, I think it offers a possible explanation for Guillaume's "cognitive dissonance" experience when adding FMC's and an optical bridge.  Impossible to prove of course, but it does seem possible to me.

I would lean a bit more in the direction of a termination issue with one (or both) of the media converters when it comes to the dissonance issue. I am quite certain that my similar experience using ethernet was caused by an improper shielding / grounding constellation. Just my objective thoughts ...

Hi baconbrain,

I am quite certain that my similar experience using ethernet was caused by an improper shielding / grounding constellation.

Can you tell us some more about your grounding/shielding experience

[Antoine]

Me too! They’re very informative and often make a lot of sense.

It's another JS +1 from me.  The link above is for the same post that I linked to in the "clocking" thread.  I think it good that Antoine has provided a link here too, it is very much on topic in this thread.  In fact, I think it offers a possible explanation for Guillaume's "cognitive dissonance" experience when adding FMC's and an optical bridge.  Impossible to prove of course, but it does seem possible to me.

Thanks. I read that. I think I’ve got power sorted. Both my tplink converters are powered by a single battery, so mains noise is not an issue. The Zen mk3 server has lps and Shunyata, next stop is Devialet.

So the linked article talks about noise from poor reclocking. When arguing with the Sonore guy elsewhere he said, and I quote:

You keep talking about “reclocking”. Let’s be clear we are not reclocking anything. These devices have oscillators in them because they are required for proper operation. All the gear you own has oscillators in them for one reason or another, but don’t intermingle them. A DAC might reclock after the USB receiver or Ethernet receiver to reduce their inherent jitter even though both processes might be asynchronous. The audio oscillators in your DAC are different than the data oscillators in network gear and computers and perform different tasks all together. The femto Aries appears to have femto oscillators on the audio outputs. This has nothing to do with this conversation.

The reason for systemOptique is simply to pair compatible network gear with our hardware. With so much network diversity on the market we needed to assure compatibility for our customers. Our optical products each come with an SFP module. The optical modules you are referring to are extra units for those who need them. There is no irony in any of this.

So this designer John S is going on about reclocking and the owner says there is no reclocking. I’m confused!

Current boxes - SOTM powering two tplink converters.

If that was JesusR you’ve had the discussion with, I’ve noticed he’s always very careful not to make claims about technical improvements and their possible effect on sound quality. It’s as if he’s trying to avoid discussions that may arise from that.

John used reclocking between “” in the post I and Confused linked to. There’s been many discussions about when to speak about reclocking or not. A digital circuit that processes incoming data uses a local clock to send it out again. Reclocking in the past was often used to describe the process of disregarding the embedded clock signal inside a S/PDIF stream in effect rebuilding a new S/PDIF stream embedding the clock signal using the own local clock.

Of course John described it much better than I can though he did not elaborate on why he put the reclocking word in between “”. I think what I wrote is the reason.

I’d trust John’s views over JesusR’s in these matters anytime. John is the designer and specialist engineer here.

The relevant part of John’s post here:
“The contamination is not complete, every time the signal gets "reclocked" by a much better clock the resulting signal carries an attenuated version of the first clock layered on top of the fingerprint of the second clock. The word "reclocked" just means the signal is regenerated by a circuit fed a different clock. It may be a better or a worse clock, reclocking doesn't always make things better!”

[Blackmorec]

I’m going to be repeating some stuff here but I think its worth it in the context of this discussion about switches. But first onto the subject of cognitive dissonance. This really rang a bell for me, given that I recently installed a USB reclocking device, ran it in for >400 hours then removed it because it had robbed my system of some of its magic.  In my case, cognitive dissonance could be summarised as the conscious improvement of hi-fi related values like frequency extension, image size and focus vs the poorer rendition of what my subconscious brain recognises as musicians singing and playing instruments. In summary I got more of the hi-fi values but my subconscious was less convinced that it was created by real musicians. My system became more hi-fi-like and less natural. It was at this point I realised that I couldn’t just blindly follow other’s advice without carefully listening to the results. The 30 day trial period suddenly gained a whole new importance.
So back to switches. About 18 months ago the flavour of the moment was the AQVox Switch and specifically their SE version. I purchased a unit to go between my router and my Innuos Zenith SE.  I liked the result, with a significant improvement in both hi-fi related qualities and musicality.  At that point I considered my system to be very close to well balanced and musical. I felt that my room had a small problem around 120Hz (mid-tone of a double bass), but otherwise I was extremely happy with how it all sounded.
One of the after-affects of installing the USB re-clocker was the discovery that my system really did not like wall-wart type SMPSs. The AQVox SE is powered with just such a module, which I’d plugged into my non-hi-fi household mains. In their literature, AQVox state that their PS is fully optimised and they could not find anything to better it, including their own, much more expensive LPS. I decided to investigate this claim. Given my surprise and delight at the outstanding SQ of the Innuos Zenith SE, I decided to contact Sean Jacobs and ordered one of his CHC DC2 LPSs. When the unit arrived, I placed it on a small Atacama mini-stand and hooked it up to the AQVox.  How to summarise the result....let’s cut straight to the fully run-in state. From a hi-fi perspective,  music has more body, the frequency extremes are more extended, imaging is more focused and defined and the soundstage has gained in layering but also has a more believable wholeness....the system now projects a ‘whole room acoustic’ ....in terms of subconscious qualities, there’s more of a feeling of voicing coming from humans instead of being slightly disembodied, brass coming from horns, guitars with nylon or steel strings, small or large wooden bodies, fingers sliding on strings etc. The pace of the music increased as did the sense of rhythmic interplay between musicians.  Frankly if I’d changed the AQVox for another, more accomplished switch I’d be entirely thrilled with the result. I was shocked that a power supply on a switch could make such a fundamental difference and of course, soon after I’d heard the impact of the DC2 I realised I should have gone for a DC3 with optional Mundorf caps. Following up with Sean Jacobs, I arranged to trade in the DC2 for a DC3 and a couple of months later I experienced another jump in performance similar to the one I’d had before. Remarkable. 
In light of the above I elected to upgrade router and ethernet cables and to add DC3s to modem, router and extender (which required extensive modification).  With these mods in place 2 things happened....the anomaly in the bass disappeared and music took on a ‘live’ feeling in terms of listener engagement, performance venue creation and rhythmic interaction. The only time there was any ‘cognitive dissonance’ along this journey was immediately following installation of consumer grade networking products which can sound pretty weird in the first 24 hours but soon settle in to provide a stable,natural sounding performance.  
In summary, while I found the switch performance to be most helpful in improving my system’s performance, at least in my case, the power supplies used to drive the whole network seemed to make a far more fundamental and valuable contribution.

[GuillaumeB]

Fascinating post Blackmorec! 

Guillaume

[baconbrain]

In summary, while I found the switch performance to be most helpful in improving my system’s performance, at least in my case, the power supplies used to drive the whole network seemed to make a far more fundamental and valuable contribution.

I have heard quite a few positive comments about the DC3 and am considering contacting them about a custom build.

[Blackmorec]

In summary, while I found the switch performance to be most helpful in improving my system’s performance, at least in my case, the power supplies used to drive the whole network seemed to make a far more fundamental and valuable contribution.

I have heard quite a few positive comments about the DC3 and am considering contacting them about a custom build.

Quite a fair portion of the Innuos Zenith SE’s and Statement’s realistic sound is down to the SJ power supplies. With the DC3 you’ll hear the same qualities of pace, rhythm and timing (fabulously intense rhythmic interactions), clarity, purity, beautifully refined and accurate dynamics that give things like brass and piano that intensity that makes them sound very realistic and believable.

[thumb5]

...
Quite a fair portion of the Innuos Zenith SE’s and Statement’s realistic sound is down the SJ power supplies.
...

You state that with some certainty. Might I ask how you know that to be the case?

[Blackmorec]

...
Quite a fair portion of the Innuos Zenith SE’s and Statement’s realistic sound is down the SJ power supplies.
...

You state that with some certainty.  Might I ask how you know that to be the case?

Sure....adding a DC3 LPS to the network brings more of the typical Innuos goodness....specifically the qualities I mentioned in my previous posts

[ssfas]

I'm not a technical person and don't understand much of the stuff said above. It fries my brain. That said, I have an Asus router with a LPS, optical bridge using tplink, iFi power on the first one and Sotm battery on the second one, a Zen Mk3 server with LPS to the ethernet output and Blue Jeans CAT6 that just arrived throughout. It all sounds lovely.