Cakyol, I didn't say there was noise on the switch. There is noise on any signal that is transmitted via an electrical connection. This doesn't mean this will affect the decoding of the digital signals. I completely concur that, with the error correction present with Ethernet protocols, the bits are the bits and they are highly likely to get too the streamer/DAC correctly. However, any electrical noise carried along with the signal has to be filtered out or it will have some effect on the generated audio. Whether it is audible is debatable.
What I was saying is that, if you believe it is audible, it makes a lot more sense to provide the best filtering close to the digital to analog conversion than to try to produce super clean Ethernet output at the switch, since you're likely to pick up more noise on the way between the switch and the steamer/DAC.
Again, I don't believe that noise on the Ethernet connection has any effect on the actual digital data that is being transferred. But there is at least a theoretical possibility that noise on the incoming connection could affect the analog output. It's up to the last device in the digital chain to adequately isolate this noise so that it doesn't adversely affect the output. |
Before I go looking at a switch for issues, try looking at all the cables and connectors first, then if you are using copper for your outside network, change that to fiber. When I did corporate networks in the early 90’s, we would verify/signature every cable we would make with a fluke meter and a good number of them had to be reterminated because of too much noise. Also, most home network cables are terminated using the cheap plastic ends. In my house, I use a 1G fiber to the outside (probably much more important than any switch because the lack of noise), and cat 7 cables to my audio and server components. |
|
A month ago, I purchased a new streamer as I was certain it was the weakest link in my system. I had previously been using a Sonos Connect modded by W4S. The Sonos Connect allows you turn off its internal WiFi module if using ethernet. I found this gave a slight improvement in SQ - and confirmed to me the benefit of reducing electrical noise in your digital chain and components. I use the Netgear Powerline devices to get ethernet to my streamer. Based on the status lights on those devices, I was linking at less than 50mb/s, and sometimes I would completely lose the ability to stream 16 bit FLAC files - dropouts, lost link, etc. That’s only, what, 1.4 mb/s? Very frustrating, but with the Connect, I could just switch back to streaming via wifi. The digital chain is: Router -> Powerline adaptor -> wall sockets -> Powerline adaptor -> streamer -> DAC (spdif) My new streamer does not have WiFi. I had to solve the powerline dropouts, and I also wanted to reduce noise in the chain as much as possible to get the most out of my new streamer (which was immediately a major upgrade in SQ over the Sonos - even with my finicky and poor networking architecture). I changed out my router from the Comcast router to a Motorola. My wifi performance improved, but the ethernet issues with the powerline adaptors did not. I then switched out the stock SMPS on the router with a Sbooster LPS, and replaced the unknown/stock ethernet cable with the Blue Jeans cat 6a. The powerline adaptors were now linking at 100mb/s with no dropouts and I heard improved sound. Steve Nugent from Empirical Audio suggests "fast" LPS for routers, switches and streamers - anything digital. He stated the Sbooster was "fast", so I went with that one. I think this might be related to the power supply "pumping" injecting noise that @atdavid has brought up. I don’t know, but I do know this change helped. I was reading about audio network switches as a upgrade (to be deployed between downstream powerline adaptor and new streamer). I was having thoughts similar to @almarg in that, as he stated better than I could: "differences in waveform characteristics in turn may, IMO, affect the degree to which some of the RF energy present in the signal may bypass, i.e., may find its way around, the ethernet interface circuitry in the receiving component and affect circuitry that is further downstream".
Knowing the powerline adaptors were probably noisy as heck, going the switch route seemed like it would be less effective than isolation. So, I decided to buy the Gigafoil v4 ethernet filter - basically a ethernet-to-fiber-to-ethernet device that removes potential noise on ethernet cable. I’ve powered the Gigafoil with an Sbooster LPS, and connected it to the network with a Blue Jeans cable on input side, and a higher end cable on the output to my streamer. The results? I totally concur with @parsons previous comment: tighter and more realistic base, lower noise floor, relaxed, dynamic and I can listen to it out much louder volumes comfortably. I have a power cable for the new streamer on the way, as well as a new coax cable. Hopefully, those will be that last upgrades for a while. Sorry for the length of this. I wanted to share some recent experience related to this topic. I totally agree with @grannyring : Everything matters in our digital audio front ends. Everything folks. |
@mitch2 To make sure I am clear, should I run a single short Ethernet cable from my router to the switch and then connect the various TV devices, plus my Antipodes DX server, to the switch....is it that simple? Yes correct. It doesn’t matter if the TV is connected to the router or the switch. Just the audio components matter. The reason I suggested a second hand Cisco 2960 switch rather than a cheap new net gear switch is that: 1. When compared to a cheap switch they sound better. I compared the Cisco 2960 to a cheap net gear and the difference was obvious. The Cisco switches are really well made and have top quality parts inside re clocking, power supply and shielding which seems to help compared to a cheap new switch. 2. I have heard that they are comparable to an audio switch but have never heard a dedicated audio switch so have no evidence here. 3. They are cheap on eBay. In the UK a £400 new price 8 port Cisco 2960 switch is £50 on eBay. 4. If it does not work you can resell it for the price you paid. |
Thanks again guys.
Regarding power supplies, I have an HD Plex just sitting around that I can hook up to my router this weekend. The HD Plex I have is totally configurable for different voltages with multiple fixed and variable outputs. I can also verify the output using my volt-ohm meter. I may eventually be able to use it to run both the router and a switch when I get one. |
There are well made switches and there are switches that were made specifically to be sold at Walmart. Once you pass a certain level of build quality, all switches at that level could be considered audiophile quality. It really boils down to what artifacts outside of the data that the switch brings can be delt with by your streamer/system. Cheap power supplies, cheap caps, cheap ICB's, you get the idea. If your streamer buffers, then speed also becomes not so important. Most isp's use a medium grade switch/modem which they try and tell you you have to rent (you don't) and unless the Audiophile switch is coming straight off the ONT at your house, all your data is passing through a medium grade switch at best. At the end of the day if your system sounds better to you after buying a $1500.00 switch, keep it and enjoy it.
|
Let me say first that I'm no expert at all in things digital. But what I know as an indisputable fact is that there is RF noise associated with the interaction between toroidal transformers, amplifiers, other gear and routers/switches/cables. This is easy to observe when I get LOUD obnoxious "motorboating" pulsing depending on placement of my router near my rack. I get very audible noise carried on the physical ethernet cable as well if I'm not careful with placement relative to the rest of my gear. I think none of this has to do with the 1's and 0's being transmitted as part part of the signal and has EVERYTHING to do with shielding and RF/EM interference. Noise can get carried on the physical line (antenna essentially). The materials that are used to shield the cable so it's not acting as a noise conduit is incredibly important. I'm not convinced the "quality" of the bits transferred has anything to do with what my ears are capable of hearing. I think separating the switch from the renderer in the router can reduce RF noise. I used a $20 Netgear switch and observed this myself. I will never pay $2000 for a network switch though! |
Things that may or may not make a difference in what I will hear is that with my Roon endpoint, the Metrum Ambre; "The board is completely and optically decoupled from the rest of the Ambre." This is apparently done using; "ultra-fast industrial optical decoupling" and "The network connection is also galvanically isolated." My understanding is the main Pi board and the outputs board are optically decoupled. They also apparently use two femto precision clocks by Tentlabs. |
|
Ethernet is always galvanically isolated. The connection is through a transformer.
|
@mitch2 Did the posts / responses answer your questions? |
To elaborate onthe post by @atdavid (and which I have not heard in this discussion) is that RJ45 based ethernet is via fully balanced signals travelling on twisted pairs. Thus, just as is the case in balanced (XLR based) analog, any common mode noise should cancel at the destination. There is also no ground . . . So, I am of the mindset that any noise in the DAC is due to defective design on the NIC in that device, nothing else.
Not sure where USB got in the dialog, since that isn't networking . . . . and wifi is out of the context of noise on cables . . . (the focus here seems to wander hopelessly at times . . .). And yes, crap offshore *wireless* devices can spew god only knows what, but that was not the discussion . . . that was *cables* and *switches*.
|
“Looks like a gimmick.”
Judging a book by its cover....thanks for gracing us with your brief presence 😉 |
LOL. There is one born every minute. ;)
|
Here's the deal. Any good service should download their stuff to a good sized fifo. Then you are playing it off your SSD and all this network nonsense goes away.
In Linux that's quite simple, I can make a fifo, any size I want, dump the network stream into it, and play it, but I don't stream. ;)
|
In the past it was the electrical engineer with years of studies behind him and plaques of degrees on his study wall whose domain it was to protect us from ourselves in audiophoolery.
Thanks to computer audio, now a highly certified CCNA can now jump on that high horse too. |
@david_ten Did the posts / responses answer your questions? Yes, I believe so for now.
Thanks again to all for the help. Based on the responses here and my sleuthing around on the internet, it seems a good start would be to purchase a switch and try it in different positions either near my router or near my Antipodes DX server. I also plan to power my router, and possibly the switch too (if it needs power), using my HD Plex linear power supply that is currently not being used. As far as which switch, it seems a
Cisco WS-C2960G-8TC-L Catalyst 2960 8-Port 10/100/1000 Ethernet Switch is pretty well-regarded and would be a good start. It can be purchased for less than $100 used or less than $200 "refurbished." Should I assume I need 10/100/1000 ports? The port configuration (speed?) seems to make a difference in the price, and they are not all the same. Another "audiophile" switch I came across in my searching that is not mentioned in the OP is the Silent Angel Bonn N8 Switch for about $400 new, from Crux. |
#adamaley
We used to make the accredited, especially the M$ network certified guys, work in the back, till we could beat most of the stupid out of them. ;)
|
@mitch2 Great. There have been a number of really helpful posts. Thanks to all for their contributions and support of the community.
Let us know which switch you choose and how it works out.
In the past I used the HDPlex to power my switch and other front end peripherals and the differences vs. the stock wall warts were easily discernible. |
Since it hasn’t been brought up, consider GROUNDING the switch with a RJ45 terminated grounding cable.
Additionally, vibrational / resonance control via footers (etc.) will help. Weights to help stabilize the unit (the consumer 8 Port units are lightweight), especially important if the LAN cables are hefty.
I bring this up because in many systems the switch seems to have the ultimate ugly duckling status. This was the case again (just last evening)...the switch was under the rack on the carpet. Out of sight but likely not out of earshot. : ) |
|
I think that the quality of digital transmission is hard to improve upon. DAC receivers have also in the last few years become excellent at jitter and ground loop isolation. However, the noise emitted by the switch mode power adapters, as well as the Ethernet cables themselves can make its way easily into linear components. So, I suggest using shielded power cables for all components. A linear power supply for digital and inexpensive power strips that have great noise reduction features to keep the network components in their own dirty sandbox. My thoughts and recommendations here: https://inatinear.blogspot.com/2019/04/power-management-for-frugal-audiophiles.html |
oh, and try to route your ethernet cables away from all audio cables, especially power supply and interconnect cables.
|
@atdavid Agreed on the ethernet cable but clearly routers and switches aren't always galvanically isolated. Ethernet is always galvanically isolated. The connection is through a transformer. |
Hiya everybody
I am a music lover (not an audiophile) who has spent a reasonable amount of money to listen to music as close as possible as to having sat in the front row of where the music was recorded
A large part of my budget was spent on a high end DAC to give me 'as close to the original as possible' audio out. According to the manufacturer (I am no expert) a digital source along with a high end DAC will give two massive benefits. It will isolate the audio output from anything that would have been considered 'interference' and 'degradation' in the old analog world, and faithfully reproduces the audio from the digital stream. It does so and it sounds wonderful.
My wired and wireless network, configured and installed by an A/V consultant, is more than capable of handling everything thrown at it including several concurrent digital high definition video streams, so even the very highest quality digital audio stream is a breeze.
I was researching switches when I found your thread which has confused me.
Why would I install something which will change the audio that the studio took so much care to create, and my DAC has taken so much care to faithfully reproduce? It just doesn't make any sense to me. Have I misunderstood what the switch manufacturers, and some of the posters on here, have stated a switch will do to my digital audio? I though a switch was a device that simply connected point a to point b.
I have copied this post to the manufacturer of my DAC and to my A/V consultant to see if they can shed some light on this.
Thanks for reading
Jason
|
@david_ten, What bridge are you using? Built into your streamer or a separate purchase? Thanks! |
@grannyring Bill, I went with Small Green Computer's internal bridge (i.e. within their server). They had offered it on my now older build. It is offered again on their most current i9 build but not on their i5 unit.
I prefer going through the internal (simple) bridge within the server versus using the switching capability of my Netgear Orbi Satellite. The differences aren't major but definitely noticeable and preferred.
I attribute the differences I hear as more likely due to the isolation of the server and the fact that it is powered off a dedicated LPS with a much, much better power cord.
The LPS to the Orbi Satellite is the same as the one to the server, however, it has dual DC outputs... both being utilized. As mentioned above, it is powered with a basic power cord. Due to the shape of the Orbi, I am unable to isolate it with the footers, etc. that I currently have.
I'm evaluating server choice. If I stay with Small Green Computer, I'll have them customize the build (including a higher quality internal bridge, if possible).
|
@jason_k2017 even the very highest quality digital audio stream is a breeze. This isn’t (solely) about having good or great latency for your streaming purposes. Why would I install something which will change the audio that the studio took so much care to create, and my DAC has taken so much care to faithfully reproduce? Consider a switch as being an ’equal’ / component-level... as you do your DAC. In my limited personal experience, the better the feed to the DAC the better the DAC’s performance. And, using your word, the more "faithful" the system's performance. |
@david_ten
Thanks for the reply
But this doesnt realy answer my question
I currently have a high end DAC reproducing audio exactly as it was digitised. Why would I install a switch that the manufacturers say will change that?
|
@david_ten
and forgive me, but if something is faithfully reproduced, it can not be 'more faithfully' reproduced
|
Jason: please forgive me, for I cannot answer your questions. My use of your word faithful was in quotes.... : ) |
David
I am just trying to understand how a DACs performance can be improved beyond an already faithful reproduction. |
Things can always be more faithfully reproduced. You won't see the endgame for most faithfully reproduced in our lifetimes.
All the best,
Nonoise |
jason_k2017 11-3-2019
Why would I install something which will change the audio that the studio took so much care to create, and my DAC has taken so much care to faithfully reproduce? It just doesn’t make any sense to me. Have I misunderstood what the switch manufacturers, and some of the posters on here, have stated a switch will do to my digital audio? I though a switch was a device that simply connected point a to point b. Understandably you may not have read all of the posts in this lengthy thread. Please see the various posts in the thread by me and by Atdavid, which have offered technically-based explanations consistent with the many experience-based anecdotal reports that have been provided in recent years by highly experienced and very highly respected long-time members (such as David_Ten, Grannyring, and the two members I referred to in my initial post in this thread, among others), to the effect that network switches can significantly affect sonics. Begin with the first of my posts in this thread dated 10-29-2019. As you’ll see, the reasons have nothing whatsoever to do with "changing the audio," and have nothing whatsoever to do with improving the accuracy with which bits are conveyed to the DAC (assuming the Ethernet link is functioning properly). They have everything to do with interactions involving ostensibly unrelated signals and circuitry, including interactions involving circuitry that is downstream of the Ethernet interface in the DAC or other receiving component. Interactions that are dependent on the spectral composition of the signal waveforms on the Ethernet link, which in turn can be presumed to vary significantly as a function of the characteristics of the particular switch and its power supply Putting it all very simply, real-world circuits and systems do not necessarily behave in accordance with idealized conceptions of how they should behave. I’ll add that while various "naysayers" who have posted in the thread have either completely ignored those explanations or have dismissed them as being "silly" and/or ignorant I feel safe in presuming that those members do not have extensive background performing detailed design of high speed electronic circuits comprising a mix of digital, analog, and D/A converter circuitry. If indeed they have any circuit design experience at all. Regards, -- Al |
@nonoise are you suggesting that I can improve on hearing a piece of music that sounds exactly as it sounded when it was encoded? And I can acheive that with a switch? Or are you suggesting that my DAC does not decode properly? @almarg
interactions involving circuitry that is downstream of the Ethernet interface in the DAC or other receiving component.
Interactions that are dependent on the spectral composition of the
signal waveforms on the Ethernet link, which in turn can be presumed to
vary significantly as a function of the characteristics of the
particular switch and its power supply I would have thought that in such an expensive piece of equipment the audio analog output would have been isolated from the ethernet digital input I really don't understand this stuff. But I guess my answer is switches should be avoided if possible |
@almarg I have said this before but you choose to ignore common sense and continue to spew techo-babble.
Your arguments assume that the ethernet signal as received by your audiophile switch is perfect and that your audiophile switch will transmit the signal to your DAC perfectly, counteracting (your words)...
"
interactions involving ostensibly unrelated signals and circuitry, including interactions involving circuitry that is downstream of the Ethernet interface in the DAC or other receiving component. Interactions that are dependent on the spectral composition of the signal waveforms on the Ethernet link, which in turn can be presumed to vary significantly as a function of the characteristics of the particular switch and its power supply".
The fact is that this signal has passed through hundreds of routers, repeaters, data centers, and switches prior to arriving at your router. Are we to understand that all those networking devices have had no effect on the signal, thus allowing that signal to arrive at your digital doorstep in pristine condition? Are we to further understand that the only place deterioration of the signal can occur is within the final switch and hence that switch needs to be a magical audiophile switch.
Your whole argument sounds like the same pseudo-scientific verbiage used to describe other incredibly overpriced nonsense products that plague hi-end audio. |
@jnorris20025, Almarg is the last person here who you'd want to accuse of spewing techno-babble.
@jason_k2017, are you suggesting that what you're hearing is exactly as it was encoded? That your DAC is perfect?
All the best,
Nonoise
|
@nonoise
@jason_k2017, are you suggesting that what you're hearing is exactly as it was encoded? That your DAC is perfect?
The one thing I do know is that if you decode an encoded stream in a quality DAC
with the the same codec
with which it was encoded then it will be as near to perfect as anything else I can throw at my sound system. My turntable (which is no longer used) is not perfect, my CD player is not perfect, my DB receiver is not perfect. Nor are yours. None are You should read up a little on codecs, digitzation and the capablilities and limitations of the human ear. There is a point at which nobody can detect an changes to audio. If you can't grasp that, think of your eyes instead and why film only needs only to be at 24 frames per second and why you have to be reasonably young to see any difference between 1080 and 4K TV |
@jason_k2017,
I can grasp lots of things, thank you, I do so by keeping an open mind.
Even at my ripe old age of 65 my eyes can still see the difference between 1080 and 4K TV on a decent set up (that was one bad analogy, by the way).
All the best,
Nonoise |
+1, @nonoise.
I see that @jnorris2005 still trying to convince us with the same old analogy of signal passing through hundreds of routers, repeaters, data centers...blah blah.
It’s the same signal / data sent to each subscriber yet our experiences differs, why....because how we choose to decode that incoming signal in the comfort of our homes. What these naysayers failed to understand is this, the same identical signal or data stream sent to his or my home results in different aural experience based on our choice of electronics. In jnorris whimsical world a $5K streamer/DAC shouldn’t sound any different than $40 Google Chromecast streaming device because we are dealing with a same digital signal, basically 1’s and 0’s.
In a digital chain everything matters.
Let’s take the 4K analogy, why does X brand of TV able to display the identical 4K stream better than Y brand of TV. It’s all in implementation and how we choose to decode that incoming audio or video signal. |
You appear to be attempting to communicate a level of technical astuteness with your post, but you appear to have missed the point completely. Almost no one in this thread is questioning the digital communication of bits perfectly ... minus the standard data errors that do occur, albeit almost never with the confines of the last data transfer. With few exceptions in this thread, and we can discount them, has anyone suggested that the signal or data degrades. That you are using that argument suggests you have not read this thread or do not understand the contents of this thread. What has been communicated, specifically by Almarg, is that the Ethernet connection, while transformer isolated, is still an entry point for EMI, both magnetically and capacitively.
jnorris200575 posts11-04-2019 12:01am
The fact is that this signal has
passed through hundreds of routers, repeaters, data centers, and
switches prior to arriving at your router. Are we to understand that
all those networking devices have had no effect on the signal, thus
allowing that signal to arrive at your digital doorstep in pristine
condition? Are we to further understand that the only place
deterioration of the signal can occur is within the final switch and
hence that switch needs to be a magical audiophile switch.
Your
whole argument sounds like the same pseudo-scientific verbiage used to
describe other incredibly overpriced nonsense products that plague
hi-end audio.
|
Actually 24 frames per second was a compromise, not perfect, but good enough in the early days when film was expensive. As "film" dies, and costs decrease, we will move towards higher frame rates to improve perception of motion. You know that "film" was flashed 3 times per second, which again was not ideal but an acceptable compromise. Being able to see differences between 1080P and 4K is more a factor of corrected vision and screen distance, not age, at least to a somewhat advanced age unless there is specific macular degeneration. Knowing everything about CODECs, digitizations and the limits of human hearing will not tell you anything about a specific DAC implementation and its susceptibility to EMI.
jason_k20176 posts11-04-2019 12:56am@nonoise
@jason_k2017, are you suggesting that what you're hearing is exactly as it was encoded? That your DAC is perfect?
The
one thing I do know is that if you decode an encoded stream in a
quality DAC
with the the same codec
with which it was encoded then it will be as near to perfect as anything
else I can throw at my sound system. My turntable (which is no longer
used) is not perfect, my CD player is not perfect, my DB receiver is not
perfect. Nor are yours. None are
You should read up a little on
codecs, digitzation and the capablilities and limitations of the human
ear. There is a point at which nobody can detect an changes to audio. If
you can't grasp that, think of your eyes instead and why film only
needs only to be at 24 frames per second and why you have to be
reasonably young to see any difference between 1080 and 4K TV
|
@atdavid
With few exceptions in this thread, and we can discount them, has anyone suggested that the signal or data degrades. This has thrown me Yes, many of you have. It is the whole point of this thread and my original question Several people have states that a 'special' switch will prevent it from doing so. Or are you now saying that a special switch is totally unnecessary as the signal or data does not degrade ? Please make up your mind |
No, only a one or two people who do not understand digital data transmission have suggested degradation of the digital signal. That vast majority of the posts have spoke to electrical noise being the most likely culprit.
I have also posited, because it is real and it happens, that data frame arrival, because it is periodic (and pseudo random) can also cause pumping of the power supply lines in the end equipment as the power supply load increases and decreases. That will be in the frequency range of audio and whether audible will be implementation dependent. |
No, only a one or two people who do not understand digital data
transmission have suggested degradation of the digital signal. That vast
majority of the posts have spoke to electrical noise being the most
likely culprit.
So if electrical noise does not degrade the digital signal, why do I need this special switch? |
I don't understand why people think it's technically impossible for a switch to introduce noise in the component chain. Switches are powered by electricity and contain fans, chips, printed circuit boards and power supplies. This isn't an argument about bits are bit and 0's and 1's or buffering or any of the well understood ethernet data transfer protocols - its about noise. Why is this so difficult? |
@jason_k2017
This isn't about degrading a digital signal, it's about electrical noise riding on the line. What do you think shields ethernet cable? It's wrapped in foil or braided wires - yes, conductors of electricity. So if electrical noise does not degrade the digital signal, why do I need this special switch? |
Still nobody is answering me
Why, if the digital signal is not being degraded and is therefore reaching its destination intact (which is all that matters), do I need a 'special' switch
|
Why, if the digital signal is not being degraded and is therefore reaching its destination intact (which is all that matters), do I need a 'special' switch?
@jason
I don't think a special switch is needed unless the hardware is adding noise to the chain via RF/EM interference. One way to address this noise, if being introduced via the switch hardware, "could" be using a different switch that doesn't introduce this noise on the line. I personally do not subscribe to any theory about digital signal getting degraded but I know as a fact that electrical gear can be sources of noise in an audio system. Do you deny this???? I strongly disagree with your assertion that a digital signal reaching its destination intact is all that matters. Introducing new electrical noise along the path also matters! Why are you choosing to ignore this? It's because you're cherry picking your argument. Very few disagree with your assertion on the bits arriving intact! Yes - we get that.
|
