ChrisBell did you try Melco as a server to your exiting equipment,give a change,I think you better try.
Ethernet leads do seem to affect the player they are connected to this is not what SiS is debating, but this is not effecting the data in anyway that is delivered as it was from the server. If it did it would never get to the streamer as it would have failed the checksums. Any noise generated by ethernet and shaped by these cables is only a side effect of the cable and the ethernet ports loading on the analogue stage its not in the digital stage so is not buffered at all as you cant buffer analogue noise.
1 Like
Crystal, without the ethernet cable there is no buffer. So you cannot dissociate both. The buffer is perhaps not the direct consequence of the data flowing in the ethernet cable. However i can’t understand how you can dissociate both.
A digital buffer cannot store analogue noise. The only thing in the buffer is the contents of the TCP packets transmitted over the wire.
1 Like
i can just say : without ethernet cable, no buffer.
1 Like
So where does the music come from when you unplug the network cable ?
The buffer exists independent of the Ethernet cable.
The data in the buffer depend on being transferred through the Ethernet cable: the contents of the buffer are the digital data only (and not any effects of the RFI).
The effects of RFI noise shaping only exist while the Ethernet cable is plugged in.
from the buffer, but this buffer can’t be done before without the ethernet cable. And also without the router, power cable…
1 Like
the data in the buffer is transmitted by the ethernet cable. It’s what you wrote. So without ethernet cable, no buffer.
Just so m.Coq, however the question being asked is, does the sound quality change when the ethernet cable is disconnected & the renderer is playing from the data packets remaining in the buffer .
the problem is that this data packet comes from the ethernet cable. If you disconnect it, the sound quality is still, in some extent, the consequence of this ethernet cable.
Perhaps the better test would be to compare the sound with Wifi vs wired with the ethernet cable ?
That is not true at all.
The data themselves (i.e. the data that end up in the buffer) are transmitted 100% faithfully by the Ethernet cable (assuming that it’s functioning correctly); those data in the buffer are absolutely identical no matter whether the cable is a £3 Cat5e Ethernet patch lead or a £4000 Chord Music so called “Streaming Cable” (in fact a certified Cat5e Ethernet patch lead may actually be more reliable in transmitting the data). Once in the buffer those data have no dependence on the Ethernet cable that transmitted them.
The only thing that the Ethernet cable can affect is the noise that’s introduced into the system after the data are stored in the buffer: specifically this typically affects the DAC and analogue electronics after the data are transferred into the DAC or affect the DAC itself during the loading of the DAC.
The effect of the Ethernet cable can have no influence on the data in the buffer.
3 Likes
Sorry FR, you keep missing the point… we are not talking about data and buffers… TCP and the media server will ensure the media in the streamers application memory buffers is valid whatever valid Ethernet cable you use.
We are talking about the ANALOGUE electronics side effects of RF type modulations from the physical network driver coupling into the streamer ground plane and/or other electronics. This will typically vary on the complex load the NIC (network physical driver) sees. This load will vary somewhat between different types of Ethernet cable.
Therefore as many now have asked to those who hear a difference with different Ethernet leads on their new Naim streamers … how does the sound change when you disconnect the cable… use a helper to do it so you are not aware…
I can’t hear a change… but in my setup I use an off board DAC.
But so far on this forum not one single person has been able to describe how the sound changes which I do find curious.
1 Like
The (Chord Company) C-stream uses twisted pairs, albeit with high-quality oxygen free copper conductors. The Indigo and Sarum Tuned ARAY Streaming cables both feature very high performance sub miniature coaxial cables for the critical signal paths. In the Indigo cable the conductors are silver-plated and PTFE insulated and the shielding is a dual layer, high-density braid. The Sarum, like our other Sarum cables, features micro polished silver-plated conductors, foamed PTFE and a highly effective, heavy gauge foil and high-density braid shield. Both the Indigo Tuned ARAY and the Sarum Tuned ARAY streaming cables also obviously feature Tuned ARAY conductor configuration.
We also experimented with the connection between router/switch and hard drive; again we heard big differences.
Are our cables directional in streaming systems?
We definitely think so, hence the marked heat shrinks. The construction of Tuned ARAY interconnects is such that although they’ll carry a signal when connected in the opposite direction, the placement of the ARAY in the signal path greatly influences the final perceived sound quality.
I attended the Chord Company demonstration referred to elsewhere in their marketing. I could not hear any difference in sound quality. So all that means is these cables are not for me.
Looking at the cables the Indigo one would appear the most promising as it has a good plastic connector. Using a metal connector when the aim is to reduce the effect of spurious analogue electrical noise does not make sense to me.
Chord Company are careful not to say its streaming cables are Cat anything. It implies they are not compliant by design. That seems fair to me.
However, the main thing is Chord Company invites you to try before you buy. So you can make up your own mind. This approach always seem reputable and fair to me.
We can argue the technical merits forever and a day. I don’t think it will achieve much. I think my own views are clear that fully qualified Cat 5e patch lead is as good as gets from switch to streamer. Cat 6a seems ideal for a home LAN as it can go at 10 Gb/s, which is nice for connections between devices that support it (most don’t, but newer NASs and Macs do).
Here FWIW is the comparative noise levels between Ethernet cables.
It’s down at -115 dB so I’m not going to worry about it.
I don’t think anyone is talking about technical merits - in a duplex ethernet lead for example there is no directionality at the MII level (digital layer)… however at the MDI layer (analogue layer) there are electronic analogue loading and noise effects - and both ends of the ethernet physical layer connected by a lead forms a connected analogue system.
No one doubts different leads can potentially change performance and potentially the sound produced by a connected host/streamer by effectively changing the parameters between the MDI interfaces - though for many of the engineers who design the host systems they know this will change from cable to cable, though some like Naim have worked hard in their new architecture to minimise the effect… the important thing is that this change is as likely is as likely on £3.99 cable as you might see on £399 cable…
Now we can argue the merits of pricing and subconscious bias but put that to one side and give it a try - I did and I was pleasantly surprised until I thought about the technical merits and then I was not really surprised at all.
Disconnecting the lead on playout will also demonstrate the difference your lead is providing (on your new Naim streamer)… it really is so straightforward…
For those that don’t want to believe in magic - here is a very good summary of what goes on with the analogue noise effects of connected ethernet segments (Fast Ethernet - as Naim uses) via a MDI and how to design hosts to mitigate the effects of it. The network and data side of things is at the higher MII level.
http://www.ti.com/lit/an/snla107a/snla107a.pdf
I do not understand where some of you want to come up with this technical reasoning,.for me it is completely uninteresting.
Sounds it different with different ehternet-cables,.for me the answer is yes.
As well,.I must of course have an ehternet-cable connected when I shall listen to music… So again what has this for practical terms when listening to music.
•Just read Darke Bears very detailed descriptions,of the significance of different Ethernet cables in his ND555 with two power supply.
He has the best source available,.and he hears a difference between different Ethernet cables.
He also has a very optimally installed music-system.
I recognize myself in Darke Bears way of managing his system,.and ways to test.
As well,.I’ve had the privilege of being able to test lots of different ehternet-cables in my own,.and others music-systems…along with others.
It is ethernet-cables that cost from £4:- to £4300:-.
I/We have always heard the difference on these cables,.sometimes minor differences,sometimes more differences.
A cable must always sit there,.so once again… Explain what you want to accomplish,.with this technical reasoning in practical music-listening.
/Peder🙂
I just did try and explain it, the cable sits between two analogue Medium Dependent Interfaces… and remember there are many facets of technicalities that describe this area and experiences - from Transmission Line theory to Cognitive Bias psychology and that is even before we consider data networking.
If your system is very expensive, you can afford a butler who can put on his white gloves and discretely remove the cable after he has checked that each track has loaded.
But more seriously, removing the cable is just a simple diagnostic tool to indicate whether or not the connected cable is the cause of any perceived difference in the sound. Some of these differences are very small, and it’s hard to tell if they are real or imagined, so trying no cable instead of just different cables might help. Besides, if you don’t like what the cable does, you have other options, such as WiFi or local USB.
or you could vary the cable until you get what you are after… I did this with my old NDX/ NDAC/555PS setup - I had about a dozen different cables of varying make up and cost - in the end i ended up with a Cat6 circa £5 cable from Amazon Marketplace on the end of an 8 port Cisco 2960 that sounded the best for me … but it was all rather subtle
1 Like
Peder, you are listening to the complete system not just the data in the buffer.
Yes as indeed Have I. The cable electrically couples to the analogue electronics affecting the sound of the analogue output stages. (Incidentally for me a particular simple Cat6 UTP sounded better than all the “audiophile” cables that I tried.)
Not strictly true for all the time (you can run tests without the cable connected for a short time, but for practical listening, yes). If you disconnect the Ethernet cable the music continues to play from the buffer for a period of time - disconnecting the cable and listening while the buffer empties is the true test of the Ethernet cable’s effect on the system.
I want to achieve understanding, so I can best design a way to test and compare the effect of Ethernet cables in the most reproducible manner, by more clearly showing the different effect they have on the analogue electronics.
2 Likes