The EtherREGEN thread for various network, cable, power experiences and experiments

[dbastin]

Hi, 

Sorry if this has already been addressed earlier, I don't have time to read 14 pages and 1400 posts of this thread.  So feel free to redirect me to where it has been addressed.

 

I have been experimenting with EtherRegen as an isolator only (not using the switch to connect a number of devices) and other similar isolators.  Like some have experienced daisy chaining ERs and other switches, I have experienced considerable sound quality when daisy chaining ER and these other isolators. 

 

My query is, why is this the case?  Is it reducing phase noise, jitter?

 

Would an external clock give an equivalent result?

 

Thanks

[dbastin]

Hi again,

 

Firstly, let's remember the system context is quite important.  I note @PYPuses Mola Mola, and I know of another Mola Mola owner whose experience is that changes to ethernet makes little difference.  He has ER > Antipodes CX > ethernet > Makua (endpoint).  This is probably rare.

 

In my case, it has evolved to become elaborate (unfortunately also untidy and costly) as follows:

 

ISP (fibre) connection box > Cat 5e > Ubiquiti EdgeRouter X > Afterdark LAN > Antipodes EX > Wireworld Platinum > ER > Wireworld Platinum > Gigafoilv4 > JCAT Signature Gold > Gigafoilv4 > Synergistic Research Atmosphere X Ref > Devialet Pro 220 (endpoint).

 

All high quality LPSs, AC power cords.  There's more to it, but that is the backbone.

 

I has discarded a few ethernet cables too. Needless to say the SR Atmos X Ref and JCAT provide a considerable improvement on their own.  I was very impressed with them even before ER was installed.  Little did I know ... what I thought was impressive them was the tip of the iceberg, and in hindsight is a fraction of the SQ I now have gained just from ethernet improvements.

 

I got Gigafoilv4 at used cost simply to compare it to ER as isolators only.  They're both excellent, and I presently prefer Gigafoil by a small margin, but an external clock, better LPS and grounding may give ER an edge if/when implemented.

 

As the Antipodes EX has an inbuilt switch with very low noise, I effectively have 4 devices isolating/reclocking daisy chained between the server and renderer.  Once inside the Devialet, timing is governed by a single reference clock and Roon RAAT 'pull' protocol.

 

I added each ER/Gigafoil upstream of EX and router, which provided improvements, but greatest improvement is the current arrangement.  From memory, each addition was incrementally less improvement.

 

The cost of cables and LPSs is racking up!

 

Hence I was hoping the Uptone guys might be able to explain why.

 

I had read the review mentioned above some time ago, and noted the use of external clock.  At the time I felt that expense was extreme ... but now with all my incremental expense I am having 2nd thoughts.

 

Its multiple isolators (incl cables, LPSs etc)  -v- external clock

 

Cheers

[dbastin]

On 2/2/2021 at 4:23 PM, dbastin said:

I have experienced considerable sound quality when daisy chaining ER and these other isolators. 

 

My query is, why is this the case?  Is it reducing phase noise, jitter?

 

Would an external clock give an equivalent result?

@JohnSwenson, I would greatly appreciate your insights into this, as per my earlier post.

 

I think this issue applies to daisy chaining any ethernet switches, or media convertors, including daisy chainingvsay a few ERs.  I have wondered at what point the incremental improvement of adding one more device is too little to notice - ie. Is it the 5th, 6th or 8th one that is not worthwhile??

 

I dont mean to disrepect by referring to Gigafoil in this thread, I would be daisy chaining ERs if I could purchase them at low cost for my (crazy) experiment - ERs have a very good resale price, which is saying something in itself.

[dbastin]

17 hours ago, Rsbrsvp said:

Next question.  My ether regen in this setup would receive an optical cable on side A which originates at my router.  But on side A I also have my server (a NUC) conneted to the etherregen with a rj45.  My question is will this connection ruin the benifit of optical connetion?

 

17 hours ago, Rsbrsvp said:

However,  my Intel NUC server is also connected to side A of the etnerregen by rj45.  Will this connection ruin the benifit of the optical connection ?

 

Try the NUC plugged into the ER Side A, and then into the router/modem/switch and listen to the difference.  In my experience ER is that works its very best when nothing else is plugged into it except 1 x data in Side A and data out Side B. I've had fibre from switch to ER too, and moved a very low noise server from Side A to the switch and gained an improvement.

[dbastin]

3 hours ago, Rsbrsvp said:

Don't understand

I mean, try 2 options

1. NUC connected to Side A

2. NUC Connected to router

 

Choose whichever sounds best to you.

[dbastin]

It would be great to get an equally elegant explanation from @JohnSwensonof why daisy chaining 2 ERs, or 3 ERs, improves sound quality.

[dbastin]

3 hours ago, JohnSwenson said:

I don't know at this point

Thanks, that is fascinating.  You have probably heard of many people daisy chaining ERs and other similar audio grade switches.  I have done similar.  And it does bring great benefits.  Mysterious!?

 

By daisy chain  I mean 2 or more switches in series.

 

Also, to clarify, this enquiry is not a criticism of your great little EtherRegen, or any other manufactures of other audio grade switches that also benefit from daisy chaining.

 

So we remain puzzled by this mystery, and keep enjoying experimenting, exploring and discovering more mysteries.

 

Thanks for your time.

 

If I discover any insights from other manufacturers, I will share if you are happy for me to do so here.

 

Incidentally, it has crossed my mind that the next gen of switches could have more 'layers' of isolation (eg. more than 1 MOAT) in one box interconnected on the PCB rather than by ethernet cables, which avoid the issues cables can cause, duplicate power supplies and the like.

 

Without mentioning names, there is already a couple of multi-switches out there, incl a dual switch in one box interconnected by fibre, and a quad switch in one box but still with 4 interconnecting ethernet cables.

 

And a switch featuring a 3ppb 25MHz OCXO, massive individual network isolation transformers, 2 internal linear power supplies, and 1 in - 3 out RJ45s was recently released, for 5 x cost of EtherRegen.

 

The journey goes on.

[dbastin]

@ambre  are you saying the Cisco RJ45 to SFP adapter you have plugged into Side A SFP is better than using fibre (ie. SFP module in FMC > fibre > SFP module in Side A)?

[dbastin]

7 hours ago, ambre said:

But indeed the normal RJ45 input sounds better in terms of timing, midrange, bass and much better stereo image

Hmm, that turned out to be a worthwhile query.

 

The trial and error means we need to continuously review our assumptions. In this case, it took a fresh 3rd persons perspective to prompt it.

 

21 hours ago, JohnSwenson said:

I cannot possibly image how this could possibly sound better than using an A side port which IS clocked by a very good clock and uses very good power networks.

So I think the lesson here is, the Side A SFP is (or should be) inferior to Side RJ45s.  In other words, the SFP is last resort if RJ45 cable is not possible.

 

The assumption is that the benefit of fibre isolation is diminished or largely eliminated by the SFP and whatever module we use.

 

Objectively ... Maybe we should test this rather than imagine.  It could be another unexplainable mystery.

 

I recall reading that the people using Taiko servers prefer the fibre input.

[dbastin]

5 hours ago, R1200CL said:

So I think the lesson here is, the Side A SFP is (or should be) inferior to Side A RJ45s.  In other words, the SFP is last resort if RJ45 cable is not possible.

 

5 hours ago, R1200CL said:

This is a totally wrong conclusion.

Yes, it is, my mistake, sorry, I misread John's text.  He actually said

 

On 3/8/2021 at 9:54 AM, JohnSwenson said:

The RJ-45 SFP modules are a completely different ballpark all together. With them ... I cannot possibly image how this could possibly sound better than using an A side port which IS clocked by a very good clock and uses very good power networks

Hence his clarification

 

an SFP WITH AN RJ-45 JACK is almost certainly inferior to the built in RJ-45 jacks. An OPTICAL SFP module is a completely different beast , and is what what most people are using. It is EXTREMELY IMPORTANT to distinguish between the optical SFP and the RJ-45 SFP.

 

[dbastin]

1 hour ago, andru26 said:

he reason I asked about the 100BASE SFP was that there are reports that the lower 100Mbps speed these have gives a better sound quality. Do you have any information regarding the SFP speed < - > audio quality relation?

I suspect this is more relevant to wired connections (rather than fibre) as I gather it is that 100Mbps enable better reduction of noise generated by a switch (for instance, a CPU/processor etc running at 1000Mbps will make more noise??), which in turn could be passed on via wired connection and in my experience somehow also impacts the sending switch in a way that is audible, even with fibre downstream!  

 

In simple terms I have this ...

 

ISP > wire > ER > fibre > Mikrotik CRS SFP+ router > wire > server > wire > wifi > endpoint

 

I can here when I change this wire ... go figure?!

[dbastin]

6 minutes ago, Billy_SP said:

For your reference, I got better results placing Router> Fiber > ER> Server, ER>endpoint. 

Thanks, I had something like that previously, amongst trying other combinations.

If you haven't tried a SFP+ router, it is worth considering.  It seems to me even the previous Ubiquiti EdgeRouter X SFP was quite a weak link.  In my case, I feel the wifi and CRS SFP+ router are critical building blocks for the SQ I am getting.

 

After reviewing cabling (UTP v STP etc), I will be trying:

 

ISP > wire > ER > fibre > Mikrotik CRS SFP+ router > fibre > ER > wire > server > wire > wifi > endpoint

 

ISP > wire > ER > fibre > Mikrotik CRS SFP+ router > fibre > ER Side A > wire > server

and ER Side B > wire > wifi > endpoint

 

My Server is Antipodes EX which has built in switch with dual PHY.  The 2nd config above will basically compare the EX switch with the ER.

[dbastin]

7 hours ago, Encore said:

Have you tried cabling the last connection to the endpoint? I have only bad experiences with wifi to the endpoint, even though in theory it should have the benefit of no galvanic connection.

In short - yes - it is not optimal in my system.   in summary, wifi:

• can be extremely good (in my case far, far superior to ER)
• performance depend on the device/streamer
• can be improved by tweaking upstream network (eg. my WAP is powered by a 5v USB powerbank).

Here's an outline of that journey.

 

I got my hands on an Auralic Aries G1, after hearing about its performance particularly with wifi.  In the end I concluded its filters and whatever else is does was deviating from accuracy in favour of producing a desirable 'impressive' sound.

 

I initially connected to G1 via $5 Cat 5e UTP, probably as per Ubiquiti EdgeRouter X SFP > server > G1.  The aim was the most basic connection.  I was impressed, the G! did produce a very desirable sound, at the time I felt better than direct to my Devialet Pro as streamer/DAC.  I gradually inserted more between the server and G1, and had 1 x ER and 2 x Gigafoil's with LPSs, and power cords and ethernet cables of $1k ea. Each additional ER/Gigafoil improved the sound, each incrementally last than the last, I felt another ER would have given another very small improvement, but ran out of LPSs.

 

I tried AES, USB and toslink from G1 to the DAC - Devialet 220 Pro.  I felt toslink was best, although the USB was flakey so I didn't really give that fair chance.  Note the toslink (Wireworld Supernova) was say $150 and the USB and AES (Audience, Shunyata) were $900, but I concluded the toslink has best galvanic isolation.  Note, the only connections to the Devialet were the power cord, speaker cables and ground wire.  Also, I had G1 upsampling to 96kHz and the Devialet natively upsamples to 192kHz.  I even tried roon core upsampling to 192kHz, G1 downsampling to 96kHz which sounded very good.

 

I think it was than I upgraded the power cord to the G1 from Synergistic Research Atmosphere Level 3 Digital to Galileo SX - a silly cost ratio but just experimenting.

 

With that benchmark, I disconnected the wired connection and turned on wifi instead.  The wifi was router > Cat 5e UTP > WAP.  it was quite obviously better again! 

 

Then I improved the wired connection to the WAP, to my surprise each upgrade (extra cable and ER/Gigafoil) provided further incremental improvements, although a lot less than via wired connection the G1.

 

With that benchmark, I removed the G1 and switched to wifi to Devialet. The difference was minor - moderate; I feel the Devialet is more accurate whereas the G1 adds more pleasing and desirable things to the sound - that is its signature, brand sound and market advantage.  In this way, G1 is more suited to less resolving systems, it brings forth detail a system may not, but in a very resolving system that, in my view, is too much, inaccurate and unnatural, even though it is intoxicatingly addictive.

[dbastin]

1 minute ago, Encore said:

Thanks for describing your journey. Just goes to show why it's so difficult to generalize ... 😊

Also to beware of marketting hype disguised in technical explanations.  Auralic has talked up its wifi, with good reason.  Devialet on the other hand, not a peep promoting its wifi, and so its potential goes somewhat unnoticed - other products may suffer the same unrealised performance.

[dbastin]

On 4/7/2022 at 11:19 PM, audiobomber said:

"ER Side B > wire > wifi > endpoint"

 

I don't understand, can you explain this please? How do you go from wire to wi-fi? How do you connect via wi-fi downstream of the ER? 

Sorry, I overlooked or missed your question.  I also realised I should have put

ER Side B > wire > WAP > endpoint.  Here is a longer description.

 

ER Side B > ethernet cable (copper) > Mikrotik mAP lite > .. wifi .. > Devialet Pro

 

Mikrotik mAP lite was chosen because it has 5v inpt (so hopefully less noisy regulators built in) and so is powered by a 5v USB powerbank and the only traffic connected to it is to/from the Devialet.  I also plug the other hole with a short USB C to USB A cable and insert AQVOX USB Detoxer QL2 - USB-A Terminator power cleaner jitter-ex.  I have WAP on another 'leg' of my network to access roon for control, it is here ... 

 

router > fibre > Ubiquiti EdgeSwitch > Cat 5e > WAP (TP link)

 

On 4/7/2022 at 11:19 PM, audiobomber said:

My exaSound PlayPoint streamer can receive wi-fi, but I found that moving the wi-fi reception upstream improved the sound. I added an extender with upgraded power supply, which receives wi-fi, but does not broadcast. The extender is connected via ethernet to the A side of my ER. The B side goes through a cable and LAN Isolator/Filter, to the PlayPoint. 

 

Let me see, are you saying you have ... wifi ... > Extender > copper> ER > copper > Playpoint.

 

I p[resume you can't run fibre to your audio room either.  Perhaps the Playpoint doesn't minimise noise from its wifi.  Or your extender may be noisy and thus the ER is providing benefit.    And you have a double isolation being wifi then MOAT (umm triple including the Playpoint ethernet isolation transformers).  But one thing is for sure, the ER clock and power regulation is superior to whatever is in the extender - and that is a very good idea!. 

 

Perhaps I should try a 2nd mAP as per

 

ER > copper > mAP1 > ... wifi ... > mAP2 > copper > Devialet,

ER > copper > mAP1 > ... wifi ... > mAP2 > copper > ER > copper > Devialet

 

Then I cold disable the Devialet wifi (which I hope would shut it down and any noise it generates).

 

Also, an idea I am going to do ... the mAP is so small and light it could be connected with e say 1 inch Cat6a UTP and 'hang' from a RJ45 port like a dongle. 

[dbastin]

On 4/8/2022 at 4:06 AM, Encore said:

 

OK, so with @GryphonGuy's help I got a couple of 1310 nm SFP modules and some single mode fiber optic cable.

 

In first instance, I have inserted it as shown in this diagram. From experimenting with Kapton tape on the earth connectors of Ethernet cables, I had established that the biggest improvement came from putting it on the 15 m long Ethernet cable that runs from my EoP outlet (probably some nasty SMPS in there).

 

So this cable goes in on the B side of the first eR. The Ethernet cable from the Roon core goes into the A side, so not quite ideal – it would have been better if I could connect both on the opposite site of the SFP port. But worth the try.

 

From the A side fiber optic goes to A side of the second eR. From the B side Ethernet cable goes to the streamer PC.

 

This gave a clear improvement. Transparency improved, making the soundstage feel bigger, I tend to turn the volume up more. Maybe the mids now feel a little recessed, something that is also a trait of my Emission Labs 300B XLS tubes. So maybe the sound from the DAC has become more balanced, exposing this character of the power tubes more than before.

 

In a funny way the sound has become almost too clean. Difficult to explain but especially at low volume the sound is a little less engaging. However, it is still early days, and I think the pros outweigh the cons. I don’t see myself be going back!

 

Some more experimenting is necessary, though. I want to try to replace the 15 meter Ethernet cable with fiber optical. However, that will move the SMPS of the first eR to a non-dedicated outlet. Difficult to predict what that will give. It gave an improvement to move the SMPS for the Macbook Pro away from the dedicated circuit breaker onto a regular one.

 

Looks like a good effort.  For what it's worth, something to consider - with the 15m run, either detach the shield from the ground at the downstream end or use UTP.  This might be better than inserting a FMC, its power supply, and fibre

 

Transparency improved, making the soundstage feel bigger, I tend to turn the volume up more. Maybe the mids now feel a little recessed, 

Difficult to explain but especially at low volume the sound is a little less engaging. 

 

I have had this a few times.  I suspect it is because there is a band of noise and distortion that has been removed making it more comfortable to play at higher volume., and higher volume helps to dig deeper into the sound and soundstage.

 

In a funny way the sound has become almost too clean.

 

Maybe what you were accustomed to was too dirty, and you had come to enjoy the noise.  Keep stripping away noise sources and you will probably hear more of the clean, including previously unheard details that bring natural-ness to the sound and its venue ... that is unless the overall tone needs re-balancing.

[dbastin]

22 hours ago, zoltan said:

I have a feeling that in this set-up, you make the ER useless for two reasons. ER cleans the ethernet signal from electric and also reclocks the signal. Electric noise will not travel though wifi, so it doesn't matter. Secondly, the Mikrotik will reclock the signal again, as any network device in the chain and unless it has a better clock than the ER, it doesn't matter. I'm not even sure if reduced jitter makes any sense when we are talking about wifi. You should do your last suggested method, except that you won't make use of the first ER as I said above: 
ER > copper > mAP1 > ... wifi ... > mAP2 > copper > ER > copper > Devialet
BTW, I also have a Devialet (Expert D800) and I think the wifi is switch off when you program it that way.  

 

Good point.  I hand in mind that this would have provided better isolation from noise for both my server/core and mAP; in which case my complete network chain to the audio endpoint would be:

 

ISP > Cat 6a UTP > ER > fibre (10m) > Mikrotik CRS 305 (as router) > fibre > 2nd ER side B > copper > mAP1 > ... wifi ... > Devailet

2nd ER side B > copper > Antipodes EX (server)

 

Note, copper is whatever ethernet cable I am exploring.  An ER upstream of wifi and fibre still makes a considerable difference, compared to a generic FMC, presumably cleaning up the mess that comes out of the ISP box.

 

So perhaps the 2nd ER would serve best and another bookend.

 

ISP > Cat 6a UTP > ER > fibre (10m) > Mikrotik CRS 305 (as router) > copper > mAP1 > ... wifi ... > mAP2 > copper> 2nd ER > copper > Devailet

This will certainly challenge the Devialet which already has an extremely low noise floor.

 

18 hours ago, Encore said:

I have had this a few times.  I suspect it is because there is a band of noise and distortion that has been removed making it more comfortable to play at higher volume., and higher volume helps to dig deeper into the sound and soundstage.

Exactly. But I think it's a combination with what I've always experienced when I have improved the digital chain: The frequency extremes become better. I.e. more weighty bass and crisper highs. In some cases I think the ear may pick this up as a more recessed midrange. This then prompts you to turn up the volume. And I think I just had to get used to it--I really dig it now! Listening to the Dune soundtrack, I was totally blown away yesterday 😆

 

I too experience the better definition of the frequency extremes, which seems to extend the range but probably just makes it intelligible.  I agree, it is a combination ... the midrange is relatively less loud 'noise' and instead quieter, more refined purity. 

 

My actual system is Devialet 440 Pro driving TAD ME1 about 85Hz upwards, and below that a pair of subwoofers each channel.  Who would've thought the small ME1s would benefit from 220w each?!  Over my journey, the bass initially got more/bigger and louder and then as the network got improved the bass became smaller/quieter but far more defined.  I gather this is due to less 'blur' which is demonstrated by the impulse curve.

 

When distortion, noise, ringing etc is reduced, bass is less bloated because there is essentially less in the decay.  This shows the difference in impulse between corrected and uncorrected signal, in this case using PEQ, but I gather it is the same concept.

 

 

13 hours ago, barrows said:

If I understand John Swenson's explanation properly, accumulated clock phase noise products will be combined with the signal such that this noise does travel over both WiFi and optical Ethernet.  The Etheregen blocks this noise from upstream sources from A-B side.

Noted, thanks.

 

13 hours ago, audiobomber said:

I have heard from several sources that wi-fi should be kept out of the streamer. I believe someone from Sonore, (maybe @barrows?), once said here that there would never be onboard wi-fi in a Sonore renderer.

 

My initial setup to test wi-fi in the Playpoint vs. upstream used a TP Link RE650 and did not include the ER. Still, this provided significant improvement. I now use a TP-Link 580D extender instead of the RE650, as the 580D accepts a linear power supply. The ER most definitely has a better clock.

 

13 hours ago, audiobomber said:

As I understand it, the ER is intended to be the last active device before the music renderer. Does that not work for you?

I will explore these possibilities as noted above.  For instance, ERs as book ends.

 

12 hours ago, barrows said:

I would say: "never say never"...  Wifi is a two way communication, so a WiFi "receiver" still has to be a transceiver.  As a transceiver, a WiFi access point is a source of additional noise-building such into a high end Renderer where we are going to great lengths to reduce all possible high frequency noise seems counterproductive at best.  As we are trying to eliminate as many noise sources as possible from proximity to the audio system, I do not believe it is a good idea to have a WiFi access point of any kind in close proximity to the audio system.  This is just what I would term "good practice".  Whether, or not, a WiFi access point in close proximity to the audio system actually will degrade audio performance is going to be highly system dependent, and would need to be tested on a case by case basis for verification.  Personally, I choose to not risk it, and I just try and keep all potential additional noise sources away from the audio system (this means all extraneous electronic components, any cheapo SMPS, noisy lighting circuits, etc).

I previously felt the same.  And my past experience with wifi was not at all encouraging.  Aurlaic Aries G1, and then Devialet, got me taking risks to revisit all those assumptions.  I don't keep changes and devices that do not make an improvement.

 

There are so many rabbit holes in this hobby.

[dbastin]

14 hours ago, emailists said:

@dbastinA bit off topic but have you tried running your ME1’s full range with subs coming in below that?

 

That’s how I run my TAD CR1’s- with dual subs coming in under 40hz.  
 

I think I posted about this previously, but as far as bass benefitting from network connection, after getting the Onti Cat8 cables to replace 6a, into and out of the ER, the entire lower end of the spectrum filled out, and vocalists had more body.  I had to lower the levels of my subs.  Incredible that network cables had such  a profound effect, and benefited other areas of the spectrum as well. 

 

Thanks.  Here is some food for thought.

 

85Hz crossover to subs is based on acoustics and speaker placement: my Devialet has high and low pass filters and delays in the digital domain.

 

Subs almost against the front wall couple with the wall acoustics and mostly seems cleanest bass - no reflection of bass from front wall or1/4 wavelength cancellation.

 

One pair of subs is about 6 inches off the floor and handle 20 - 55Hz high excursion which my JL e112 do very well.  The e112 also provide crossover to the upper pair.

 

The upper pair are above those about 2 feet off the floor which is cleaner upper bass than on the floor, so handle 55 - 85 Hz and I use Martin Logan Balanced Force 210s for this task.

 

The ME1s can then be placed out into the room, away from the influence of the wall,  and again far enough to minimise1/4 wavelength cancellation and optimise for listening position.

 

The ME1s midrange and mid bass is not spoilt by also producing low frequencies.  It sounds cleaner and clearer.  More powerful but far less reverberant and excessive overloading (and blowing windows out).

 

I say all this because it is quite critical to judging and fully appreciating  improvements to ethernet which yield considerable improvements in low end .. and its a lot more then just more and deeper, it is meaningful, articulate, dynamic, controlled, fast, etc.

 

I have had more and deeper, but as I continued to improve the ethernet system, it got more lean, precise, and lifelike.

 

So, the profound effect of your cables may actually still be affecting the music rather than being transparent.  I have Shunyata Sigma ethernet and it is a prime example of leaner but far more detail.  The impulse chart I posted earlier is to show how a cable or some other thing in a network could be resulting in the 'slower' responding curve - it is what some call blur.  This seems most apparent in low frequencies, provided your speaker placement and room acoustics aren't preventing this from being apparent.

[dbastin]

9 hours ago, Encore said:

I have come to the conclusion that there is no such thing as absolute isolation. No matter how good your isolation device, it can always be improved by more of them and better PSUs, clocks, et cetera.

In my experience there is:

• isolation and also
• the impact of 'noise' (or something else mysterious ) on each device regardless of isolation

For instance, using wifi or fibre, the receiving device is isolated for the sending device, and the receiving device can be designed for very low noise (eg. power supply, ground, etc, etc), but a change in power supply, cord or fuse (or some other tweak) at the sending device can cause a difference to be heard, and that can actually be an improvement to sound quality rather than or as well as just different.

 

As much as the science says the data is checked etc for perfection or it is rejected and resent until it is, and then buffered, my ears are not convinced.

 

Therefore, the answer to this ...

 

16 hours ago, Mat said:

Is there any benefit to use Melco S100 in front of Etherregen ? 

It depends on many things that may not be fully understood yet, and also on the overall context (network and system) but typically the answer will be YES.

[dbastin]

10 hours ago, kostas6a3 said:

Has anyone used two EtherREGEN in series and if so with what results?

I could not find anything about the above question in the 107 pages

I am very surprised to hear you cold not find any posts about ERs in series .. surely people have dome this, I am sure I have read it somewhere.  I have, in simple terms two is better than one, however as mentioned you can combine the sonic character of different switches (and cables) and there are lots of combinations, including fibre and various SFP modules.

 

I suggest make sure all your current network is optimal before spending on an ER. or something else.

 

Cabling (and their shields) is as critical and can have a similar scale of improvement as an ER.

 

Consider this alternative first.  Replace your router with a SFP+ capable router.  Mikrotik are low cost, I use Mikrotik CRS 305 but there are others.  The CRS models can be configured as routers (a bit tricky but with instructions, concentration and patience still quite doable).   For their small cost, with good power supply etc, reap rewards that would otherwise cost much much more.  More info over here ...

 

 

For example, I have just one ER and the Mikrotik CRS downstream with great cabling and the results are shockingly good.  I have a spare ER and 2 Gigafoils at my disposal and not in any particular hurry to reinsert them, although bear in mind my last connection to endpoint is via wifi which in my case trumps all these toys put together.

[dbastin]

1 hour ago, audiobomber said:

My exaSound PlayPoint has wi-fi capability, but ethernet connection sounds much better, even without the EtherREGEN in the circuit.

 

A Furutech LAN-8 CNF cable on the B-Side brought great results for me.

Just like any other interface, not all wifi is equal.  I have Devialet Pro and prefer it to using wifi to Auralic Aries G1 as streamer connected to the Devialet.

 

What did the Furutech replace for you to consider it giving great results?  Do you know what Furutech does about grounding the S/FTP shield?

[dbastin]

2 hours ago, R1200CL said:

If people considder that route, I highly recommend to install iPfire on hardware similar to this. 
 

The you can create separate physical networks if you believe is such. And you can also have a very nice VPN solution for Roon. 

To clarify in case it was too subtle, the SFP '+' is 10Gbe and apparently that spec is superior.  I can certainly say in my case the CRS 305 is significantly better for SQ than the Ubiquiti EdgeRouter X SFP (1Gbe) it replaced.

 

I'm not sure if there is hardware like you suggested that is SFP+.

[dbastin]

17 hours ago, kostas6a3 said:

So far the two EtherREGENs in the series have not performed as expected.

The sound became softer, lost in sharpness and the bass looser. Everything is powered by the power supplies that come with them. The power supply of HDPLEX 300W is expected, which will power the two EtherREGEN, the Melco N100 and the NUCK for the final hearings.

Some observatiins to ponder ...

 

Sounds to me like the things you are describing as missing could well be the adverse impacts of noise from the network.

 

Noise can impose a hardness and unpleasant edge, and that may be preferred in the context of some systems, or people become accustomed to it as being normal.  I am curious about the loose bass though.

 

In my experience, bass changes a lot in different ways as the network is improved and generally becomes more well defined, accurate impulse.  Bass parts in music are far more in sync with the rhythm.  In musician terms, the bass playing sounds 'tighter' with the band.

 

But there is also a more extended/wider frequency range, and stronger bass response, as if the bass has been compressed (whereas with the affects of noise the bass is more spread out over longer duration).

 

It is possible the 2 ERs are now highlighting problems elsewhere in your network, room or system.

 

Consider that the new sound could now be exciting your room differently and creating a new perception??

 

Also, take care with how ethernet shields are connected.

 

And, be careful using one power supply for 2 ERs, and server.  It may be possible for noise to transmit from one device to another via the power supply, which could negate the MOAT isolation.  It probably depends on the design of the power supply.

 

Finally, the N100 built in switch might not be providing you maximum benefit if the N100 is your endpoint.  Instead of using NUC as server, you might get better results using N100 as server and having a DAC with an integrated  player.

 

Router > ER1 > N100 > ER2 > endpoint DAC

[dbastin]

52 minutes ago, kostas6a3 said:

Maybe I did not put it correctly. The sound was lost in clarity.

The two EtherREGENs currently use their own pulse power supply and do not have a common power supply.

The Melco N100 is a music server and end point and not a network switch so I do not see how to EtherREGEN could be connected to the DAC.

 

Hmm, I suspect something is not quite correct, perhaps a connection.

 

N100 has at least a network bridge and possibly a full switch.  It seems to have 2 PHY chips.  The 'player' port can be used to connect to a DAC with an integrated player/streamer.(eg. LUMIN D2, Aurender A10, lots more).

 

Off loading player functions from N100 will reduce load on the cpu, in short most likely gain improvement in SQ.  Having a player integrated with the DAC reduces complications if converting formats, cables etc.  Again, probably benefits SQ.

 

[dbastin]

I am afraid it is trial and error.   Perhaps best to use a decent UTP to replace every one of your cables one at a time to find out which connection is causing the problem.  Don't worry about the SQ until you work out which cable connection is causing the problem, and then work out the best sounding cable for that connection that does not cause the problem.

 

In one situation in my network, I have a relatively low cost commercial grade Cat 6a UTP performing better than a $800 audiophile Cat 8 cable that connects shields at both ends with metal connectors.  But the Cat 6a is trumped by audiophile cables elsewhere downstream.

 

This is because:

• there does not appear to be a standard way to connect shields for ethernet cables used for audio and
• there may not be a standard way audio gear deals with the ground of their ethernet ports, and metal connectors. 

In my liaison with the makers of the audiophile ethernet cables I have, they use what ever method achieves the best sound from their cables, and they use different tech and methods in the cable construction.