I think they make very good products, the context of my comment was in relation to a specific conversation with a Melco sales rep.
I challenged his logic and he wasn’t able to provide what I felt was a credible answer based on my own understanding and experience.
2 Likes
Your comment makes me smile 
It is just some lack of some academic rigor.But anyway who cares, just have a listen to the Linn Organik with optical ethernet, that is all I wanted to say.
1 Like
Sure… I don’t deny that certain ways of doing things might make a device sound preferable for some… but to describe it by stating incorrect technical assertions confuses and mis informs those who are not technically disciplined and can see through the marketing hyperbole… for some reason in the area of consumer Hi-Fi there seems to be a disproportionate amount of this sort of mumbo jumbo… and I find that irksome.
It’s almost as if hood winking is the only way to sell certain “premium” priced consumer products. To me much of this is perilously close to snake oil and charlatanism.
BTW I will say again… I have mentioned over the years on here, including when I predicted fibre Ethernet would become fashionable… there is a big potential down side with Fibre for consumer Hi-Fi use into current all in one streamers … it is sensitive to mechanical/microphonic vibration (this property is very much used in certain industrial and commercial applications) … which means link layer clock speed can be phase modulated by the sound pressure in the room. This means the stability of the link clocking can be governed by the sound. We know that the link layer clock can be important in all in one streamers… hence the rise of the so called audiophile switch … so if using fibre… optimum results may well be achieved with headphones.
Twisted pair doesn’t suffer from microphony to anywhere near the same extent. and I have never seen the recover of audible sound pressure from twisted pair unlike I have with fibre.
So fibre and twisted pair have their respective advantages and disadvantages, there is no magical universal ‘best’… this is why I have suggested future network players should be fitted with SFP module sockets, so the user can select fibre or twisted pair SFPs to suit taste/preference.
4 Likes
Whilst I appreciate the technicality of your point there, I am willing to bet there is no audible difference human hearing can pick up for vibrations in a fibre cable, the whole Ethernet switch/cable BS that audiophiles have grabbed onto is embarrassing for this hobby.
I only wish I had the nads to start my own business selling this crap, I could retire.
8 Likes
We are all in a nut house. 
1 Like
Well I am ‘of service to humanity’ and of course it’s my opinion, that is all anything put on forums are, opinions.
As absolutely no evidence of any change is available (apart from ‘even my wife heard the difference’) we are where we are and those claiming differences only have opinion as well.
6 Likes
I’ll agree with that but you have to agree that perhaps I maintain a healthy amount of cynicism to ensure I don’t get sucked into total and utter nonsense particularly when it relieves your wallet of thousands of pounds, fair?
4 Likes
I think it’s that, or either a system not resolving enough to hear differences. But saying it’s crap is offending for all those who hear easily the differences , with their ears and on his proper system.
2 Likes
Its true playing back my 4000 quid ethernet switch into my 300 quid Pioneer AV amp is disappointing, this must be why.
We have done a number of the billy bingo, I did ‘even the wife heard the difference’, we have also had ‘your ears must not be good enough’ and ‘your system must not be good enough’
Great stuff.
7 Likes
Apologies, why is it offending? I am sure that you know what you know that hearing is subjective.
2 Likes
Not quite… if you remember I was one of the first who discovered switch clock stability did affect SQ on first gen Naim streamers and someother closely situated Hi-Fi equipment equipment… I was hunting down an issue of interference ‘birdies’ on my NAT03… and discovered other beneficial side effects when I significantly mitigated them by changing the Network switch … I reported it here… and others, though to be fair, certainly not all, observed the same.
In this case it was swapping a cheap little consumer Netgear switch with a used Catalyst 2960.
Admittedly one didnt need to spend a lot to eliminate this issue but it was very real, and benefits of link clock phase stability was discovered by accident from a sound performance point of view.
I subsequently had several people including audiophile journalists and others contacting me privately on the matter… others wider than this forum were experiencing similar things… and whether coincidence or not within 18 months or so, ‘audiophile’ switches started to become generally available…
My beef is not about cause and effect… as it is real for many, and others were invited to observe before it became fashionable or costed much to do… and a majority, though I repeat not all, found it beneficial, no my beef is how some of this stuff is marketed and the techno babble justifications from certain quarters.
Fibre can cause the same potential issues… phase distortion through microphony… I however have reduced such things from my audio system now by effectively decoupling transport from DAC , and Naim second generation streamers seem less susceptible to link clock stability issues than earlier products, but it would be churlish to disregard the similarities between network switch link clock stability and fibre microphony phase modulation… albeit the modulation frequencies would be quite different… so I hand the batten to somebody else to explore and measure any consequential sound issues with fibremicrophony on closely coupled DAC transport streamers, just as I initially discovered by accident with low stability switches,
5 Likes
So you didn’t notice the two tests were using different servers then? No wonder the results were repeatably different… Oh no, what have I done? Now we will get a thread discussing which ISP locations provide the largest soundstage, air, clarity, separation and should you change server location depending on their geographic location because their local RF environment varies by time of day impacting jitter and noise floor…
1 Like
That’s something I’ve yet to try. I just slotted the DSM in and used the same copper connection and network setup I was using for my NDX2. I’ve got all the bits to try out the fibre connection, just not the time to have a play around at the moment.
2 Likes
The response was to Gary……
I will not respond and open a hundred time more a discussion on subjective hearing or not. You know already my opinion on that, I am sure.
2 Likes
Have you tried making use of the Linn’s room correction option (aka space optimisation)? It can look pretty complicated upon first look, and hard to get it right the first few times, but it can really enhance the sound of an already good system. I was lucky that my linn retailer (the same one who sold a complete linn system to Jon Ive) spent nearly a whole day setting it up for me. My wife can tell the differences straight away if the space optimisation is on or off.
2 Likes
Frankly, as the Germans so elegantly put it, “es ist mir Wurst” (I couldn’t give a sausage) whether people believe in network optimisations or not. After all its their trade off between
- financial cost
- loss of potential (unrealised) SQ and fidelity
I have been into music my whole life (from radio Caroline on AM radio) to high-end hi-fi installations, with several versions of the LP12, various Naim passive and active systems, various Balanced Audio tube systems and a variety of Speakers, from 2 way Guarneri Homage to AG Trios with 4 subs. I would classify myself as someone who knows hi-fi, knows what typical upgrades bring to the party and can accurately differentiate improvements from no change or downgrade. I have made upgrades I loved, intended upgrades where I listened but wasn’t impressed (eg external earthing systems) so didn’t proceed and changes that brought negative SQ changes (USB reclocker (due to poor PS), Footers…which failed to harmonise with my other support devices).
I discovered the importance of the network quite by accident. I was new to digital streaming, so needed a good network. The then current ISP-based wi-fi was quite poor with not such good coverage so I set out to build a much better network. With several major electronics stores within a reasonable radius of my home I was able to try a large variety of different devices on a 30 day return basis. I tried ethernet, mesh, wi-fi, Internet over Power , wi-fi extenders etc. Most stuff only needed about 24 hours to run in and it all sounded decent….but it also sounded different, with some clearly better than others….Hmmm, so right from the get go I was choosing network components on both their networking capabilities AND on the eventual SQ they produced. I was confident that transmissions were bit perfect as I had no musical or continuity anomalies. The set-up I found best comprised the following:
ISP Modem/router in modem mode
WAN connection to 3 band router (giving dedicated audio 5GHz band)
Wi-Fi to Ethernet bridge then on to the server, an Innuos Zenith MkII SE.
This set up handily improved on a direct router to server Synergistic Research Active SE ethernet cable, presumably due to the galvanic isolation. There was nothing I didn’t like about the sound it produced other than what sounded like a small hump in the bass that for some reason mainly effected specific double bass notes.
Having read some really good reviews from ears I trust, I invested in an AQVox SE switch. Essentially €800 for a modified D-Link switch and tweaked switched mode wall wart power supply.
Installed it made a small but worthwhile difference…something I’d assign say 10% to as a function of perceived change, that is, the AQVox made perceivable, meaningful, positive changes to about 10% of the music, so it stayed but didn’t shake any foundations.
My same virtual networking consultants were now identifying USB reconditioning as the next big SQ influence, so I arranged with my dealer to ship me the much lauded USB retimer/conditioner, with its optional separate power supply. Installation was a mixed bag, I could hear some improvements, but all the magic was gone. The music wasnt ‘lighting up my soul’ and i found myself back reading forums while the music played. Not a good sign at all! 400 hours I waited for the unit to run in before I returned it because the magic never returned. As soon as I removed the unit and its power supply from the mains and reconnected a simple server to DAC ethernet cable, the magic returned, in spades. So I plugged in the USB retimer PS again and the magic again went missing. Hmmm….my system doesn’t seem to like certain SMPSs on the dedicated circuits, which of course took me back to the AQVox and it’s SMPS. AQVox include a note in the box informing you that the SMPS had been optimized and no other power supplies, including expensive LPSs will sound better. At this point my system was sounding magical, but given my experience with the USB retimer, I ordered a Sean Jacobs DC-2 to try. This came with Sean’s own DC and Mains cables, which are excellent. Installation was a shock. The uplift was big and fundamental….everything was cleaner, clearer, purer, with greater speed and vitality and much better dynamics across the entire amplitude spectrum (quiet to loud). This uplift in dynamics had as much affect on classical violins as it did on rock or jazz because it was the music that became more dynamic. Oh and that small hump in the bass was gone! A few weeks later, the system finished running in and I was really impressed with the sound quality I was getting vs all my other attempts over 50 years. So impressed that I returned the DC-2 and started the entire running in process over with a DC-3 + Mundorf caps. Installation was again a very pleasant surprise, the uplift in SQ was expected, what wasn’t expected was the degree by which it was improved.
What was very clear was that the power supply in the network line was imprinting the sound with its identity, which is ‘fast, dynamic, super low-noise, pure’ It was causing the presentation of the music to change.
Essentially, if the power supply imprints the signal, then the imprint has to be within the ‘fabric’ of the bit stream…in otherwords, accuracy of, differences in and stability of the timing of the bits, the quality of the voltage changes within each bit, the amount of VHF noise going with the bitstream and of course the available bandwidth and the amount of network traffic and the noise it creates.
So lets say that like everywhere else in audio, the quality as well as the content of the bitstream matters when the bitstream is evaluated according to the quality of the sound it contributes to producing. Essentially a cleaner, more accurate input always results in an improved output, so changes to the network do not change content, they change the quality of the carrier, making it more accurate, more precise, less noisy, with all the qualities of the upstream stream components, good or bad.
When I purchased my Innuos Zenith SE I read a very good description of Innuos’s engineering goals, which were:
Generate less EMI
Generate less noise
Allow less EMI ingress
Improved cabling and cable joints
Vibration control eg. chassis footers designed to remove internal and isolate external vibration
Excellent ultralow noise power supplies
Ultra accurate, high stability clocks
Low noise components
RAM music buffer
Optimized SW
When the above are applied to a server like the Innuos Zenith MkIII or the Statement, what you get is SoTA sound quality. When you apply the same principles to the server’s network, you get the very same uplift in the music’s presentation, sometimes to amazingly large degrees.
Get your network feed close to perfect in terms of bit timing, structure and associated noise and the rest of your system will reward you with much higher SQ.
I have spent may years upgrading systems and for the most part, an upgrade is assimilated within a few weeks and levels of joy and happiness remain about the same. But when upgrades make the sound more like real musicians playing in a real or engineered acoustic, the increased involvement and believability profoundly change the listeners’ reaction to the music, making it more intense and emotional and enjoyable.
So in the end it matters not a sausage whether you believe in network changes or not….just know that by believing that all network set ups are the same as long as you achieve a bit-perfect data stream, you leave an entire field of potential fundamental improvements in SQ unaddressed. This has no affect on the beautiful music I’m enjoying and the only purpose of this post is to share my experience so not everyone remains in the dark about the positive sonic effects a really well optimised network can deliver.
8 Likes
I’ve not tried SO yet. I’ve downloaded the latest manual on SO so will print it out next time I’m in the office and then give it a try. Still rediscovering old favourites at the moment 
1 Like
I am not sure what is with your streamer or DAC, if it is a Naim DAC, then Naim should say something?
I think it is kind of something that you need to invest a significant amount of money and efforts that make it sound the way you like, and caution - it might not be the true music or sound that you think it is.
My wife and I have been to a number of high-end HIFI exhibitions, and a number of reputable HIFI dealers in the CA, US, and by the end of the day my wife said none of these systems matches the SQ of my system at home in terms of musicality, and these include some systems which cost in excess of $100K, some of which we wish could run fast away, as far as away from them as possible just after 5 mins of listening.
My system is not only musical (this is what we are looking for in the first place, a perfect musical performance if there were one), un-fatigue over a long period of listening, but I can also hear layers, visual images as well as violinists’ breathes, some clothes’ rustling, movement noises in the background.
All these, I just need a simple cisco switch with optical ethernet out.
1 Like
Me too, same thing I am using with my Lumin.
On the weekend it finally cooled down here in Canada enough that I could turn my AC off.
Dead silent noise floor, with music just floating in space.
3 Likes
When i stated at the start of my post that my system was sounding magical, that’s exactly what i meant. An ability to hear deep into the music, with what I thought at the time was state of the art performance. “Nothing I didn’t like”. At that point i really couldn’t see where any further improvements could lie…surely I was revealing everything. I guess i was hearing pretty much what you are hearing now, a perfect digital rendition of the music. But what I didn’t know at the time but later turned out to be the case was that every single further improvement to the network stream brought a further uplift in sound quality, such that magical became more magical, became utterly enthralling, became spell binding, became an utter overload of emotions.
Back in my previous post I mentioned a bunch of engineering criteria that Innuos apply to their servers, which i adopted for my network. In time i managed to apply those criteria to every part of my network chain. I did this because every time i uplifted one or the other criteria, i was rewarded with a compelling uplift in SQ that i judged to be subjectively very well worth the asking price. What I came to realise during the 3 years it took to fully complete my system is that like any other part of hi-fi, the quality of the components used in the network have a major impact on the eventual sound it produces. What I also observed was that an improvement at say the router rippled all the way through the network, meaning that at every component along the way, better in = better out. By the same token, worse in equals worse out. This has a major ramification on the network structure. In order to gain the maximum bang for you buck, the output from any component must be of higher specification compared to the preceding component. For example, if an output has a clocking accuracy of 10ppm, it makes no sense to take that output and feed it into a device with a 100ppm accuracy clock. In order to get the best result, the clock following the 10ppm unit must be an improvement, 10 ppt for example. This is true for clocking, power supplies, cabling, vibration control etc. To get the very best from your network every component must be the same or better in terms of engineering criteria to the foregoing components. If you think about it, it makes no sense to feed a super low noise power supply’s output into a component with a noisier supply. The goal of your network is to both move data and to provide that data in a form that is as close to perfect as possible.
So Y-B, I have no doubt whatsoever that your system sounds wonderful. But imagine if that were just the starting point to something much, much better. After my initial discoveries I have continued refining my network, adding escalating vibration control to each phase, making sure clocks were of consistently improving specs, using the finest power supplies available, utilising Mundorf Silver/Gold cable for all DC wiring, constructing a cabling loom based on Synergistic Research Atmosphere X Ref (and soon Galileo for the final Ethernet and USB). Each step of the way has brought significant improvements, each worth the cost in terms of SQ improvement…and together the sum sounds spectacularly good, way, way better than the quality of sound at the outset, which at the time I considered would be very difficult to improve upon.
The system as it stands now has attributes I have not previously experienced. For example, when I start a track I am immediately engaged, because the recording instantly and profoundly changes the acoustic I am sitting in. The music starts and what’s immediately apparent is a clarity, pureness and accuracy that grabs my consciousness, obliterating any thought. Next the dynamics and speed of the music sound real…notes have a pinpoint source then expand, bloom then decay, often in a particular direction. The amount of information is huge….even ‘silence’ having a particular feel and character. Spatially, the system is completely immersive……to the point I can only listen with eyes closed as the contradiction between eyes and ears is simply too great and very discombobulating, disconcerting and confusing. But most important are 2 things: the majority of improvements are becoming increasingly more difficult to describe in hi-fi terms. Essentially what changes the most is my reaction to the music….how much more subtle rhythm I detect, how much better the music communicates its message in terms of my own strong feelings and emotions and finally how good the music makes me feel.
The second important observation is that the better your system becomes, the larger the perceived leaps in SQ each subsequent upgrade brings.
Finally I would say that so far, I have detected zero in terms of the law of diminishing returns…the opposite in fact. What I would say however is that while the improvements get bigger, they also become more expensive. For me, my personal journey was in extracting the very best possible from my system, so the process was based on my budget and my quality requirements. However, the process of optimising a network is entirely scalable and fits perfectly into a gradual, stepwise program/hobby type environment.
At any level, a really well optimized network will reward (which is exactly why yours works so well), but the key words are ‘at any level’ because network enhancement is a continuum along which sound quality and network quality rise along with price/cost. Just like everything else in hi-fi!
3 Likes