Nordost announces QNET switch

[kennyb123]

I don’t agree. I think the fiber thing is different, but not better. We tried it here with the X1, and my reaction was “meh”. Different, not better. We went back to Ethernet as we thought it was more organic.

I think you're mostly right. Fiber can vary greatly depending on the SFPs and FMCs used. One of the biggest surprises to me in audio was that I could actually hear differences in SFPs and FMCs. I must have tried at least a half dozen SFPs until I finally landed on a discontinued model of Finisar SFPs recommended by Emile of Taiko. My downstream FMC had been the EtherRegen, but again to my surprise, even the FMC upstream of the fiber span mattered. A Sonore opticalModule turned out to be a noticeable improvement over a Startech FMC that I first started with.

I like what I hear from my span of fiber because it seems to do no harm while noticeably lowering the noise floor. It's a big part of the reason I am feeling pretty settled as far as my network goes.

Net net, I can certainly understand why companies like Nordost and Innuos wouldn't want to add fiber capabilities to their switches. It's not always better. But with a little work it can be, so if I ever do purchase another switch I think I'd want it to have fiber capability.

 

[kennyb123]

Sorry but SFP from point to point eliminates noise potential compared to Ethernet

Some SFPs actually add their own noise. Same for the FMCs.

 

[jmusica]

Never considered that the SFPs could have characteristics that contribute to SQ. Interesting.

Are you aware of anywhere that discusses this/has some recommendations?

And yeah, I don't really think that fiber vs copper is really much different than many other tweak-ish alternatives - they change the sound and some may like one way more than the other - little of this, as has been stated elsewhere, is objective when it comes to preference.

So much energy and money is spent on tweaks to "reduce jitter" "lower noise" "resist emi/rmf etc", from magic discs sitting on top of things to wooden doo-dads sitting in a corner to (expensive) things that pick cables up off the floor or carbon fiber outlet covers. Why so resistant to something that actually is grounded in principle in doing those things?

In three systems so far I've preferred fiber to copper - my perception is a lower noise floor/blacker background/more clarity, and better soundstage width/depth/height. Which brings me back to a point I made earlier and others echoed - why would a manufacturer not include this very standard tech as an option on a new product? Give us the choice, to pick the network topo with the characteristics we prefer whether that's SQ or physical installation. Imo they were just saving some $ on engineering and manufacture. For such an expensive product that strikes me as a customer-unfriendly decision.

 

[Puma Cat]

Never considered that the SFPs could have characteristics that contribute to SQ. Interesting.

Are you aware of anywhere that discusses this/has some recommendations?

Lot of discussion regarding audio qualities of SFPs to be found at formerly named computer-audiophile based site.

And yeah, I don't really think that fiber vs copper is really much different than many other tweak-ish alternatives - they change the sound and some may like one way more than the other - little of this, as has been stated elsewhere, is objective when it comes to preference.

That's not accurate. The variability between different copper Ethernet cables is WAY more than with the various fiber-based solutions. This is because by nature, optical fiber is less susceptible to the influence of noise factors than impact audio quality than copper Ethernet.

And, optical fiber, when utilized for longer runs, is not susceptible to the passage of high-source impedance leakage current. Nor is it susceptible to RFI/EMI.

Oh, and copper Ethernet is also way susceptible to common-mode noise, as well. This is why I recommend using Shunyata Alpha or better Ethernet cables when/where using a copper Ethernet cable.

The only "error mode" that it passes the same as copper Ethernet, is phase noise from el-cheapo consumer-grade clocks in cheap-*ss consumer-grade networking devices.

So much energy and money is spent on tweaks to "reduce jitter" "lower noise" "resist emi/rmf etc", from magic discs sitting on top of things to wooden doo-dads sitting in a corner to (expensive) things that pick cables up off the floor or carbon fiber outlet covers. Why so resistant to something that actually is grounded in principle in doing those things?

Cable elevators actually work because they mitigate the impact of floor-borne vibration imparting triboelectic noise into the dieletric of speaker cables, which causes audible "blur and slur" in the presentation. The key thing is you need to use a cable elevator that also dissipates the static E-field that arises when you lift a cable off the floor. This static field can be measured. The static field also imparts a noise component to the EM wave that propogates down the speaker cable that carries the signal. Remember: the signal is NOT carried by electrons "flowing" in the cable conductor, like water in hose, it is carried by the EM wave that propogates along the cable. And, the VP (velocity of propagation) of that EM wave is frequency-dependent.

In three systems so far I've preferred fiber to copper - my perception is a lower noise floor/blacker background/more clarity, and better soundstage width/depth/height. Which brings me back to a point I made earlier and others echoed - why would a manufacturer not include this very standard tech as an option on a new product? Give us the choice, to pick the network topo with the characteristics we prefer whether that's SQ or physical installation. Imo they were just saving some $ on engineering and manufacture. For such an expensive product that strikes me as a customer-unfriendly decision.

Because most manufacturers are behind the eight-ball with respect to this, that's why. Look at how long it took them to understand that power cords make a BIG difference.

 

[kennyb123]

Never considered that the SFPs could have characteristics that contribute to SQ. Interesting.

Are you aware of anywhere that discusses this/has some recommendations?

Search for the word Finisar at Audiophile Style. You should see posts from me where I share my thoughts. You’ll also find the key threads where this stuff is discussed.

And yeah, I don't really think that fiber vs copper is really much different than many other tweak-ish alternatives - they change the sound and some may like one way more than the other - little of this, as has been stated elsewhere, is objective when it comes to preference.

So much energy and money is spent on tweaks to "reduce jitter" "lower noise" "resist emi/rmf etc", from magic discs sitting on top of things to wooden doo-dads sitting in a corner to (expensive) things that pick cables up off the floor or carbon fiber outlet covers. Why so resistant to something that actually is grounded in principle in doing those things?

Fiber gets way too much credit by some because they refuse to acknowledge that perfection is simply not possible. Everything comes with a series of tradeoffs. Running a span of fiber doesn’t perfect isolate.

(The worst is what I see on some of the Chord DAC forums where there are some who believe use of a toslink cable will eliminate the need for a better server.)

In three systems so far I've preferred fiber to copper - my perception is a lower noise floor/blacker background/more clarity, and better soundstage width/depth/height. Which brings me back to a point I made earlier and others echoed - why would a manufacturer not include this very standard tech as an option on a new product? Give us the choice, to pick the network topo with the characteristics we prefer whether that's SQ or physical installation. Imo they were just saving some $ on engineering and manufacture. For such an expensive product that strikes me as a customer-unfriendly decision.

A friend who owns a PhoenixNET says that what he hears from it is indistinguishable from not having the network plugged in at all. Maybe Innuos didn’t see further benefit from adding fiber capability.

Another reason not to add fiber is that it will increase the burden on customer support (or on their dealers as I believe both Innuos and Nordost look for dealers to be the first support tier). Have a look at Sonore’s or Uptone’s forums to see how many questions they get about fiber networking. They also get questions about the best choice in copper ethernet but that’s far simpler to get going.

Nordost actually sells their own Ethernet cables so it would seem kind if ridiculous for them to add fiber capabilities to their switch. They would of course want their customers to use their own cables with their switch.

 

[kennyb123]

Cable elevators actually work because they mitigate the impact of floor-borne vibration imparting triboelectic noise into the dieletric of speaker cables, which causes audible "blur and slur" in the presentation. The key thing is you need to use a cable elevator that also dissipates the static E-field that arises when you lift a cable off the floor. This static field can be measured. The static field also imparts a noise component to the EM wave that propogates down the speaker cable that carries the signal. Remember: the signal is NOT carried by electrons "flowing" in the cable conductor, like water in hose, it is carried by the EM wave that propogates along the cable. And, the VP (velocity of propagation) of that EM wave is frequency-dependent.

I use the original Shunyata Dark Field Elevators. They aren't as effective as their current elevators, which I believe you are using. They aren't as adept at mitigating floor-borne vibration but they do help to dissipate the static field. I didn't hear an improvement immediately upon placing my cables on them but after pulling them out to loan to a friend, there definitely was an added "blur and slur".

My previous MIT speaker cable were from their Oracle series, so the networks were encased in metal boxes. Placing those boxes on Stillpoints helped to remove a bit more of the "blur and slur".

I wasn't sure I fully understood why both these measures helped, but Stephen your excellent post sure makes a lot of sense.

 

[Puma Cat]

I use the original Shunyata Dark Field Elevators. They aren't as effective as their current elevators, which I believe you are using. They aren't as adept at mitigating floor-borne vibration but they do help to dissipate the static field. I didn't hear an improvement immediately upon placing my cables on them but after pulling them out to loan to a friend, there definitely was an added "blur and slur".

My previous MIT speaker cable were from their Oracle series, so the networks were encased in metal boxes. Placing those boxes on Stillpoints helped to remove a bit more of the "blur and slur".

I wasn't sure I fully understood why both these measures helped, but Stephen your excellent post sure makes a lot of sense.

Yep, the newer ones are more effective, I think Kenny, because the polymer strap the cable is rested on provides more isolation from mechanically-transferred vibration; the white piece of plank foam in the middle of the original design of elevators may impart a bit more vibration-sourced energy into the cable. And, the crystalline materials in the base are more effective at dissipating static fields than the foam used in the original design. I've found that when Shunyata makes a new design change, it's for the better with respect to functionality/features.

Cheers, mate.