A novel way to massively improve the SQ of computer audio streaming

[Nenon]

2 hours ago, Tommd64 said:

Compared the prices and see whats going on. I ll say without hearing please buy a Farad3, it's also possible to order it with highgrade componends..or better dc cable. Coz he 700p for a Sean Jacobs is a lot of money for a realitively simple PS. Compared to a Farad or SotM.

 

And how did you come to that conclusion? You can tell how things sound by looking at a picture? 

 

The LPS1.2 for example has some outstanding engineering for example. But it does not sound as good the 1.5A Sean Jacobs LPS nowhere in my system. I've had some exposure to the SOTM and was not impressed. The Farad looks interesting but I've never had one to try. 

[austinpop]

4 hours ago, drjimwillie said:

 What I’m asking about is a little bit of out-of-the-box thinking.   

 What if you do what I have in the sketch? 

 

Why on earth would you connect both ports of the server to the switch?

 

1 hour ago, Nenon said:

What are you trying to accomplish with this configuration? It does not make much sense to me.

 

+1.

[drjimwillie]

I thought about this because of two reasons.

 1.  Let me please preface this with the fact that we read a lot of things here connected to audio. I thought that I had read somewhere regarding ethernet connections that more  Ethernet connections made for a better sound. But I could not find the reference. From my memory the person had two ethernet ports on their server and they connected them both to their switch and because of the increased data flow the sound was better. I was assuming that possibly one ethernet cable would be used for the incoming signal and the other ethernet cable would be used for the outgoing ethernet signal.

 Then either by pontification or test the theory was if you had two cards with 4 ethernet ports that 4 cables made even better sound.  I ask the question because I wondered if anyone here had actually tried it? 

2.  The other reason I am asking this question is because I am doing my homework to be prepared for the EtherRegen.   If there is complete isolation from one side of the switch to the other then there may not be a detriment to hooking up multiple cables. 

[austinpop]

7 minutes ago, drjimwillie said:

I thought about this because of two reasons.

 1.  Let me please preface this with the fact that we read a lot of things here connected to audio. I thought that I had read somewhere regarding ethernet connections that more  Ethernet connections made for a better sound. But I could not find the reference. From my memory the person had two ethernet ports on their server and they connected them both to their switch and because of the increased data flow the sound was better. I was assuming that possibly one ethernet cable would be used for the incoming signal and the other ethernet cable would be used for the outgoing ethernet signal.

 Then either by pontification or test the theory was if you had two cards with 4 ethernet ports that 4 cables made even better sound.  I ask the question because I wondered if anyone here had actually tried it? 

2.  The other reason I am asking this question is because I am doing my homework to be prepared for the EtherRegen.   If there is complete isolation from one side of the switch to the other then there may not be a detriment to hooking up multiple cables. 

 

Can you provide a link to these reports? And do you recall if the 2 (or 4) ports were bridged or not? Were they using "link aggregation?" What was the rationale?

 

I'm not averse to trying wacky things, obviously, but this particular topology makes no sense to me. 

 

11 minutes ago, drjimwillie said:

because of the increased data flow the sound was better.

 

Sorry, but that's just not how Ethernet works. 

[Nenon]

34 minutes ago, drjimwillie said:

I had read somewhere regarding ethernet connections that more  Ethernet connections made for a better sound.

 

This post explains that pretty well:

On 6/4/2019 at 7:07 PM, JohnSwenson said:

The understanding of "isolation" in digital audio has been my passion for at least 10 years. There is a LOT of misunderstanding on the subject floating around in audio circles.  Here is a quick summary of my current understanding and how the current products fit in with this.

 

There seems to be TWO independent mechanisms involved: leakage current and clock phase noise. Various amounts of these two exist in any system. Different "isolation" technologies out there address one or the other, but very rarely both at the same time.  Some technologies that attenuate one actually increase the other. Thus the massively confusing information out there.

 

Leakage current is a property of power supplies. It is the leakage of AC mains frequency (50/60 Hz) into the DC output. It is usually common mode (ie exists on BOTH the + and - wires at the same time, this makes it a bit difficult to see. There seems to be two different types, one that comes from linear supplies and is fairly easy to block, and an additional type that comes from SMPS and is MUCH harder to block. An SMPS contains BOTH types. They are BOTH line frequency.

 

Unfortunately in our modern times where essentially all computer equipment is powered by SMPS we have to deal with this situation of both leakage types coming down cables from our computer equipment. There are many devices on the market (I have designed some of them) for both USB and Ethernet, most can deal with the type from linear supplies but only a few can deal with the type from SMPS.

 

Optical connections (when the power supplies are completely isolated from each other) CAN completely block all forms of leakage, it is extremely effective. Optical takes care of leakage, but does not deal with the second mechanism.

 

Clock phase noise

 

Phase noise is a frequency measurement of "jitter", yes that term that is so completely mis-understood in audio circles that I'm not going to use it. Phase noise is a way to look at the frequency spectrum of jitter, the reason to use it is that there seems to be fairly decent correlation to sound quality. Note this has nothing to do with "pico seconds" or "femto seconds". Forget those terms, they do not directly have meaning in audio, what matters is the phase noise. Ynfortunately phase noise is shown on a graph, not a single number, so it is much harder to directly compare units. This subject is HUGE and I'm not going to go into any more detail here.

 

Different oscillators (the infamous "clocks" that get talked about) can have radically different phase noise. The level of phase noise that is very good for digital audio is very difficult to achieve and costs money. The corollary is that the cheap clocks used in most computer equipment (including network equipment) produce phase noise that is very bad for digital audio.

 

The important thing to understand is that ALL digital signals carry the "fingerprint" of the clock used to produce them. When a signal coming from a box with cheap clocks comes into a box (via Ethernet or USB etc) with a much better clock, the higher level of phase noise carried on the data signal can contaminate the phase noise of the "good" clock in the second box. Exactly how this happens is complicated, I've written about this in detail if you want to look it up and see what is going on.

 

The contamination is not complete, every time the signal gets "reclocked" by a much better clock the resulting signal carries an attenuated version of the first clock layered on top of the fingerprint of the second clock. The word "reclocked" just means the signal is regenerated by a circuit fed a different clock. It may be a better or a worse clock, reclocking doesn't always make things better!

 

As an example if you start with an Ethernet signal coming out of a cheap switch, the clock fingerprint is going to be pretty bad. If this goes into a circuit with a VERY good clock, the signal coming out contains a reduced fingerprint from the first clock layered on top of the good clock. If you feed THIS signal into another circuit with a very good clock, the fingerprint from the original clock gets reduced even further. But if you feed this signal into a box with a bad clock, you are back to a signal with a bad fingerprint.

 

The summary is that stringing together devices with GOOD clocking can dramatically attenuate the results of an upstream bad clock.

 

The latest devices form Sonore take on BOTH of these mechanisms that effect sound: optical for blocking leakage and multiple reclocking with very good clocks. The optical part should be obvious. A side benefit of the optical circuit is that is completely regenerates the signal with a VERY low phase noise clock, this is a one step reclocking. It attenuates effects from upstream circuits but does not completely get rid of them. This is where the opticalModule comes into play, if you put an opticalModule in the path to the opticalRendu you are adding another reclocking with VERY good clocking. The result is a very large attenuation of upstream effects. It's not completely zero, but it is close.

 

The fact that the opticalRendu is a one stage reclocking (which leaves some effects from upstage circuits) is why changing switches etc can still make a difference. Adding an OpticalModule between the switch and opticalRendu reduces that down to vanishingly small differences.

 

So an optical module by itself adds both leakage elimination and significant clock effects attenuation. TWO optical modules in series give you the two level reclocking .

 

An opticalRendu still has some significant advantages over say an ultraRendu fed by a single opticalModule, the circuitry inside the opticalRendu has been improved significantly over the ultraRendu. (it uses new parts that did not exist when the ultraRendu was designed). In addition the opticalRendu has the reclocking taking place a couple millimeters away from the processor which cuts out the effects of a couple connectors, transformers and cable. The result is the opticalRendu has some significant advantages.

 

An opticalModule feeding an ultraRendu does significantly improve it, but not as much as an opticalRendu. So you can start with an opticalModule, then when you can afford it add an opticalRendu, also fed by the opticalModule and get a BIG improvement.

 

I hope this gives a little clarity to the situation.

 

John S.

 

 

 

 

 

39 minutes ago, austinpop said:

Sorry, but that's just not how Ethernet works. 

Exactly.

 

56 minutes ago, drjimwillie said:

2.  The other reason I am asking this question is because I am doing my homework to be prepared for the EtherRegen.   If there is complete isolation from one side of the switch to the other then there may not be a detriment to hooking up multiple cables. 

 

Plug your Endpoint to the B side of the EtherRegen, and everything else to the A-side. I would probably isolate the router with fiber too. 

[numlog]

Previously, both the onboard ''SoC' SATA and USB of Ryzen CPUs showed promising performance compared to the chipset and even upgraded PCIe cards mounted in primary PCIe slots but were still too harsh/metallic.

The onboard SATA now appears to have benefited greatly from upgrading to linear CPU power supply, much smoother with further improvements to resolution, it is the best SATA has sounded.

The 2.5'' SATA drive connects via M.2 slot with SATA to M.2 adapter, this means my M.2 Optane OS drive is moved to a PCIe slot with an adapter.

This happens to a more optimal configuration since the Optane drive can only realistically be fed clean, external power when mounted to  one of these adapters, the 2.5'' SATA drive has no problem being fed external power when connected to the M.2 slot via an adapter.

Im using I2S PCie card for the DAC interface now so cant test how modded USB card compares to Ryzen's onboard USB, I would guess it must also benefit from improved CPU power. 

 

Also, regarding CPU clock speed - underclocking the Ryzen 1700 resulted in the  typical underclocked ''sound'', more liquid/smooth but flatter and duller, a more boring sound.

AFAIK adjusting clock speed in BIOS with any CPU will automatically lower the core voltage.

With the underclock the core voltage drops from ~1V to ~0.8V, adjusting the core voltage back to  its 'default'' 1V  counter acted the some of the negative affects we have associated with underclocking, the underclocked sound is much better this way.

Still need more time evaluate whether I think its better with or without the underclock.

I dont use any oversampling or conversion for playback.

 

 

 

 

[drjimwillie]

 I read about trying to make my system sound better for a couple of hours every day, unfortunately this is more time than I get to listen to my system. 😕

 Do that every day for a few years and some of the references get lost.  And I’m sure some of the metaphors get mixed. 

 I remember reading about link aggregation but I’m not sure if that’s the basis for a better sound? 

 

 I thought it was kind of a wacky idea, which is why I asked the question here. 

 I was wondering if you could use or how you could use a switch in conjunction with the bridge  

 I would try my idea, but I’m having a couple of technical difficulties in my system is not conducive to critical listening at the moment.  When I try it I will let you know. I am surprised that no one else has tried it up to this point. 

 

 I am not sure how the link to the JS article proves my point. I cannot find a correlation after reading it.   Would you please explain what you’re thinking? 

 

 I have learned that I should try the switch after the server rather than before it. Although the switch before it did increase the SQ.  I think the increase in SQ is because the ethernet cable from the router to the server is 30 feet long and it benefited from the signal regeneration.

 

Thank you 

[Nenon]

The JS comment I quoted was the best explanation I have seen about the effect of (multiple) clocks. I thought you might be referring to the same post.

 

It's not that complicated. You have electrical noise you need to worry about and you have reclocking. You want the electrical noise to be as isolated from your endpoint as possible. You also want to have the best clocks as close as possible to your Endpoint. That's why @austinpop recommended the following design for you:

---- Inbound from router <-> (port 1) Server (port 2) <-> Switch <-> Endpoint. ---

The bridge configured on the server would isolate your noisy router from your endpoint. And the (audiophile) switch would reclock the signal (with a good clock) just before the Endpoint. It's a much better solution than what you are thinking. 

 

Isolation can be achieved in multiple ways - you can use fiber optic connection, you can configure network bridge on your server, you can use a filter like the gigafoil v4, there are medical ethernet filters like the EMO EN-70HD or Baaske, etc. Some work better than others. Some kill dynamics, others work miracles. There is no consensus what's best, because it is system/environment dependent. My choice at the moment would be a pair of Sonore opticalModules with good LPS and single-mode long range 1Gbps transceivers. But once the EtherRegen is released it would be my choice for isolation. Obviously you need to maintain clean power after the isolation, so good LPS's play an important role there.

 

Reclocking is the second important factor, and that is even more complicated. The quality of the clock is impacted by the quality of the power, cables, vibrations, etc. You need to take really good care of all those factors. 

 

But those are the two major factors you need to worry about. When I looked at your design with the server bridge, first it did not make sense, because the Ethernet traffic would not flow like you thought. Also, your router and endpoint were sharing the same switch, which means there is no electrical noise isolation.

 

When the EtherRegen is released, if what they explain is true, that would be an easy solution to all your problems. And it would simplify things a lot. You put the Router and server on one end and you put the endpoint on the other end. No need for network bridging or anything complex.  I am a bit sceptical the EtherRegen would work so well and do a complete isolation plus excellent recklocking, but we would see. I really hope it does. 

 

Until we know it does for sure, the OCD in me is thinking about something like this, but that of course is a huge overkill: 

On 7/1/2019 at 12:46 PM, Nenon said:

 

If the claims Uptone Audio is making about their new switch are true, this can become a flat network with one switch only and no bridging...

[johndoe21ro]

On 7/25/2019 at 6:26 AM, austinpop said:

A Mini Review of the Farad Super3 Power Supply

 

Here is my review of the Farad Super3 Power Supply. I know this has taken longer than anticipated, but I wanted to ensure that the PSU and all associated cables were fully burned in. In addition, as you'll see, Mattijs Vries (designer of the Farad) and I explored some more cable alternatives, and this took additional shipping and burn in time.

 

Why is the Farad Super3 PSU interesting? First, several listening reports on this and other threads have been very positive. Second, it is a supercapacitor LPS design, claimed to be designed for low noise and low output impedance. Third, it supports current up to a max of 3A, which is significantly higher than the 1.1A max of the Uptone LPS-1.2 and the 2A max of the Paul Hynes SR-4. Finally, at a base price of €439, it competes very strongly with these units.

 

This review is an attempt to do a systematic and thorough listening comparison of the Farad with these existing best-of-breed PSUs commonly in use for digital audio gear. As we on this thread know, the quality of a power supply in a digital component can be as important as it is in an analog component. Yet, the set of really high-quality power supplies is rather small. On the pinnacle of this range is the lofty Paul Hynes SR-7, which in its DR and XL incarnation has reigned supreme in all my listening tests. But this PSU is almost impossible to acquire, now that Paul has stopped taking new orders for his custom PSU business. Next in the spectrum are the Uptone JS-2, LPS-1.2 and the Paul Hynes SR-4, which are more accessible. These are excellent supplies in their own right, while not quite at the level of the SR-7. How does the Farad Super3 compare to these? While I no longer have a JS-2, I do have LPS-1.2s and SR-4s on hand to compare, as well as access to an SR7 DRXL unit.

 

I requested an evaluation unit from Mattijs, who very kindly and promptly shipped me one. By mutual agreement, this unit was configured for 12V. Farad supplies are designed to be configured for a specific voltage, as Mattijs feels the provision of voltage selectors degrades sound quality. In addition to the default DC cable with the unit, Mattijs also sent me his Level 1 (shielded quad core 18 AWG tin-plated copper), and Level 2 (shielded quad core non-inductive mil-spec 16 AWG tin-plated copper) upgrade cables. I picked 12V, since it allowed me to test the Super3 in several different spots in my chain.

 

Before attempting any critical listening, I burned in the Super3 and the Level 2 cable for 500+ hours. I know it took a long time for my SR-4 to reach peak SQ, so I afforded the same courtesy to the Farad. 

 

Listening impressions of the Farad Super3

 

Before getting into comparisons, I tried out the Farad in my system to get a sense of its sonic characteristics. I tested it in 2 separate locations in my chain:

• powering the sNH-10G switch
• powering the tX-USBultra

I also tried this both in my primary headphone system, and in @limniscate's speaker-based system.

 

From the get-go, I liked what I was hearing. The first impression with this PSU is one of tonal calmness. It has a smooth, relaxed character, with a black background, speaking to its low noise floor. The other characteristics that are immediately apparent are its tonal richness, the depth and slam of the bass, and great dynamics. Even without any comparisons, I was very please with the sound quality I was hearing.

 

A note about cables

 

As I was preparing for the comparisons, I realized that all of the other PSUs were using silver cables of one type or the other. Over time, I have gravitated to silver DC cables. Unlike analog silver cables, that can tend to add brightness and sparkle, I find silver DC cables to enhance clarity, neutrality, and depth. Here are the PSUs and associated cables that I was preparing to compare:

• LPS-1.2: Audio Sensibility Signature, and a custom cable that a kind ASer from Japan sent me a few months ago, built with the Oyaide FTVS-910 cable. Both cables are terminated with Oyaide 5.5x2.5 barrel connectors. I've modded the AS cable with JSSG360 shielding.
• SR-4: Paul Hynes fine silver DCFS3XLR 3A cable, modded with JSSG360. Terminators are XLR-4 to Oyaide 5.5x2.1
• SR-7 Paul Hynes fine silver DCFS6XL 6A cable. Terminators are Jaeger to Oyaide 5.5x2.1

I didn't want cable quality and materials to affect the comparison, so I discussed this with Mattijs. At Pink Faun, they had shied away from silver, but he was open to the idea. In fact, he has been experimenting with a "Level 3" silver cable, which is still a work in progress. While he sent me a sample, I won't be talking about that until he's finished tweaking, and is happy with the result. In addition, he supplied me with adapters (GX16 to 5.5x2.5 and GX16 to XLR-4) that allowed me to use all my existing silver cables with the Farad - all but the SR-7 cable.

 

My cable findings? I preferred all 3 silver cables over the Farad Level 2 cable on the Farad PSU. Within the silver cables, the differences were very small. If I had to choose, I would rank them: 1) Audio Sensibility, 2) Oyaide FTVS-910, 3) Hynes DCFS3XLR.

 

I do understand that these adapters are suboptimal, and almost certainly increase the output impedance of the PSU. The fact that the silver cables performed better, despite this disadvantage, suggests that a properly terminated (GX16) silver cable for the Farad will sound even better. 

 

For the remaining comparisons, I used the same DC cable with the Farad as the PSU under comparison.

 

Comparison with the LPS-1.2

 

The LPS-1.2 is the go-to LPS for digital gear, and is a tough-to-beat combination of excellent SQ and reasonable price. I've had as many as 3 of these simultaneously in my system, and currently use 2 of them. For this comparison, I alternated between the LPS-1.2 and Farad, both with the AS cable, first powering the sNH-10G switch and then the tX-USBultra.

 

These two PSUs certainly have different sonic attributes! The LPS-1.2 sounds open and spacious, but also brighter, and slightly more in-your-face. The Farad, in contrast, is smother, darker, with deeper bass. Its resolution and rendition of fine detail is just as good, while being less fatiguing over longer listening sessions. I also felt it was a bit more dynamic than the LPS-1.2.

 

I'd say the Farad is the better PSU sonically, although it does cost approx. 25% more than the LPS-1.2.

 

Comparison with the Paul Hynes SR-4

 

For this comparison, I used the DC3FSXLR cable with both PSUs, with the Farad using a GX16 to XLR-4 adapter. 

 

This was a close one! I had to go back and forth many times to form a lasting opinion on this comparison. Tonally, both the SR-4 and the Farad share the dark tonality that I find so pleasing and fatigue-free. They were neck and neck on dynamics and bass. The one area where the SR-4 pulled  ahead - and only by a nose - was fine detail and resolution. The texture of instruments, the detail in fine cymbal and brush strokes was just a little clearer and better defined with the SR-4. 

 

However, the difference was very small, and let's keep in mind that the Farad was handicapped in the output impedance department due to the use of a GX16 to XLR4 adapter. It would not surprise me at all if the Farad closed the gap with a properly terminated cable.

 

I'm going to declare this a tie between the Farad and SR-4.

 

Comparison with the Paul Hynes SR-7 DRXL

 

Did I really expect the Farad to match or outperform the SR-7 DRXL? Not really, and it didn't. With the SR-7, there was just more of everything. Dynamics, bass, clarity, resolution, and texture - they were all there - and all emerging from an inky black background. You pay dearly ($1200+ per rail, plus the patience of Job to get a unit) for the SR-7 DRXL, and you do get what you pay for!

 

Other Use Cases

 

I haven't yet remarked on the 3A capacity of the Farad, which is in fact a great advantage it holds over the LPS-1.2 and SR-4. One concrete use case where this came in handy was in powering the Chord Hugo TT 2. When I reviewed the TT2 a few months ago, I had found that its SQ improved modestly over the stock SMPS when powered by an SR-7 DR rail at 15V. Since I received my own purchased TT2, and since I don't yet own an SR-7 DRXL, I've been using a PowerAdd Pilot Pro 12V to power the TT2. This battery improves SQ over the stock SMPS supply. However, that is the extent of my PSU experimentation. While the TT2 nominally consumes under 2A in steady state (unless powering speakers), it does have an inrush peak of 3A (per Rob Watts) for several seconds while the supercaps "charge." For this reason, I've been loath to try my SR-4 with it, although Paul Hynes has promised to run some tests with the SR4-19, set to 15V, driving a 3A load, to see if it can handle this transient.

 

However, the Farad, with its 3A capacity was easily able to handle this startup transient, and powered the TT2 beautifully. SQ was improved over the battery, and of course the stock SMPS. The main difference was the Farad sounded more relaxed and tonally richer.

 

Note that this experiment was performed with my 12V review unit. It remains to be seen whether a 15V unit would improve SQ even further.

 

Conclusions

 

The Farad Super3 is an exciting entrant to the world of high-quality LPSUs for digital audio gear. Its sound quality is essentially on par with the Paul Hynes SR-4 when used with a good silver DC cable like the Audio Sensibility Signature, while availability is - at least, for now! - much better than the 3-4 months current lead time for the SR-4. In certain applications, if your current needs exceed the SR-4's 2A, the Farad would even be the preferred option.

 

While the Farad Super3 does not match the sound quality of the vaunted SR-7, its easy availability and generous current capacity make it a very compelling choice for everything ranging from switches, DDC's, and clocks to NUCs and even certain DACs.

 

Highly recommended.

 

Next Steps

 

Mattijs will be shipping me a 15V unit soon to validate whether the Farad @ 15V is a better fit for the TT2. I plan to compare this with an SR7 @ 15V. The TT2 improves only modestly even with the SR-7, so the question here is whether the TT2 powered by the Farad is "close enough" to the SQ of the TT2 powered by the SR-7 DRXL.

 

I also expect to receive, in due course, Mattijs level 3 cable once he's perfected it. I'll be comparing this to my benchmark Audio Sensibility and Oyaide DC cables, and reporting back.

 

Stay tuned for an update in a few weeks.

Great review and a long expected one, Austinpop!
In a couple of words could you describe the differences between how each one (sNH-10G switch and then the tX-USBultra) reacted with the different PS? Was one of the SOtM's more susceptible than the other? Also, I understand that you have a tX-USBultra with a 12V input (as I do). Did you try to internally switch it to 9V? If so, does it sound better with the 9V or the 12V? Thanks!

[Blackmorec]

13 hours ago, Tommd64 said:

Compared the prices and see whats going on. I ll say without hearing please buy a Farad3, it's also possible to order it with highgrade componends..or better dc cable. Coz he 700p for a Sean Jacobs is a lot of money for a realitively simple PS. Compared to a Farad or SotM.

If you wanted to directly compare an SoTM SPS-500 to one of Sean Jacob’s designed units, the closest in sound quality terms would be the DC-2, which I found to be superior to the SoTM in terms of naturalness, PRAT, soundstage cohesiveness and therefore listener involvement.  The DC-3 is a significant step up from the DC-2 and in the right system, well worth the extra ££. 

How do they sound? Anyone familiar with the Zenith MkII SE  or Statement would immediately recognise the sound of a DC-2 and DC-3.  How to describe that sound? Anyone who frequents small/medium live music hangouts will be familiar with the way you can easily follow each instrument, but how the magic comes from the way the instruments and voices combine to produce a highly engaging, bounce you out of your seat rhythmic whole. Above all else, the SJ power supplies bring the individual clarity and depth,  the rhythmic drive and the musical ‘sense’ that hooks and engages the listener without any need to concentrate.  

[baconbrain]

4 hours ago, Nenon said:

Until we know it does for sure, the OCD in me is thinking about something like this, but that of course is a huge overkill: 

4 hours ago, Nenon said:

The JS comment I quoted was the best explanation I have seen about the effect of (multiple) clocks. I thought you might be referring to the same post.

 

It's not that complicated. You have electrical noise you need to worry about and you have reclocking. You want the electrical noise to be as isolated from your endpoint as possible. You also want to have the best clocks as close as possible to your Endpoint. That's why @austinpop recommended the following design for you:

---- Inbound from router <-> (port 1) Server (port 2) <-> Switch <-> Endpoint. ---

The bridge configured on the server would isolate your noisy router from your endpoint. And the (audiophile) switch would reclock the signal (with a good clock) just before the Endpoint. It's a much better solution than what you are thinking. 

 

Isolation can be achieved in multiple ways - you can use fiber optic connection, you can configure network bridge on your server, you can use a filter like the gigafoil v4, there are medical ethernet filters like the EMO EN-70HD or Baaske, etc. Some work better than others. Some kill dynamics, others work miracles. There is no consensus what's best, because it is system/environment dependent. My choice at the moment would be a pair of Sonore opticalModules with good LPS and single-mode long range 1Gbps transceivers. But once the EtherRegen is released it would be my choice for isolation. Obviously you need to maintain clean power after the isolation, so good LPS's play an important role there.

 

Reclocking is the second important factor, and that is even more complicated. The quality of the clock is impacted by the quality of the power, cables, vibrations, etc. You need to take really good care of all those factors. 

 

But those are the two major factors you need to worry about. When I looked at your design with the server bridge, first it did not make sense, because the Ethernet traffic would not flow like you thought. Also, your router and endpoint were sharing the same switch, which means there is no electrical noise isolation.

 

When the EtherRegen is released, if what they explain is true, that would be an easy solution to all your problems. And it would simplify things a lot. You put the Router and server on one end and you put the endpoint on the other end. No need for network bridging or anything complex.  I am a bit sceptical the EtherRegen would work so well and do a complete isolation plus excellent recklocking, but we would see. I really hope it does. 

 

Until we know it does for sure, the OCD in me is thinking about something like this, but that of course is a huge overkill: 

 

 

I don't want to sound like a SoTM fanboy but wouldn't a single sNH-10G fulfill all of these requirements  (isolation, clocking, etc.) while simplifying the setup? (one switch, no bridging, ...)

[Summit]

17 hours ago, drjimwillie said:

Is anyone using this  configuration? 

 How does it sound? 

 What did you compare it to?

 What switch are you using?

 

Strange things can sometimes have an effect 😜. No, I don’t think you will benefit anything with this configuration, but if you use 2 switches, maybe.

 

Router – switch 1- server- switch 2 – endpoint.

[George Hincapie]

Can I ask generally, for those of you who are feeding your server output via ethernet direct to your streamer, and have upgraded the power supply for your music servers from SMPS to LPSU, what impact did that have? What was the scale of perceived improvement if there was any? Conversely, did any of you go through this process and fail to note any improvement?

 

Trying to decide whether to spend £1K on a LPSU for my server, and am trying myself in knots. I don't want to commit unless there's going to be a level of improvement to justify it.

[davide256]

 

5 hours ago, Blackmorec said:

If you wanted to directly compare an SoTM SPS-500 to one of Sean Jacob’s designed units, the closest in sound quality terms would be the DC-2, which I found to be superior to the SoTM in terms of naturalness, PRAT, soundstage cohesiveness and therefore listener involvement.  The DC-3 is a significant step up from the DC-2 and in the right system, well worth the extra ££. 

How do they sound? Anyone familiar with the Zenith MkII SE  or Statement would immediately recognise the sound of a DC-2 and DC-3.  How to describe that sound? Anyone who frequents small/medium live music hangouts will be familiar with the way you can easily follow each instrument, but how the magic comes from the way the instruments and voices combine to produce a highly engaging, bounce you out of your seat rhythmic whole. Above all else, the SJ power supplies bring the individual clarity and depth,  the rhythmic drive and the musical ‘sense’ that hooks and engages the listener without any need to concentrate.  

?? This is a long thread but you should probably go back and refresh your memory that the OP originally had Zenith server but abandoned it

for an AL/NUC solution. While a good power supply is essential for resolution, the drive and musical sense you describe come from the hardware/software

platform capabilities

[Nenon]

1 hour ago, George Hincapie said:

Can I ask generally, for those of you who are feeding your server output via ethernet direct to your streamer, and have upgraded the power supply for your music servers from SMPS to LPSU, what impact did that have? What was the scale of perceived improvement if there was any? Conversely, did any of you go through this process and fail to note any improvement?

  

Trying to decide whether to spend £1K on a LPSU for my server, and am trying myself in knots. I don't want to commit unless there's going to be a level of improvement to justify it.

 

Upgrading the stock SMPS to a good LPS is the single best thing you can do for your computer source. From all the thousands of combinations discussed in this forum, that's the one thing everyone agrees about. It is the most important thing to do for your computer after all! Whether spending £1K is worth, that's something only you can decide. To me, spending 20% on computer and 80% on LPS gives much better results than spending 80% on computer and 20% on a LPS.

 

The impact of a good LPS has been discussed many many times here. You just need to do some reading about it. 

[George Hincapie]

18 minutes ago, Nenon said:

 

Upgrading the stock SMPS to a good LPS is the single best thing you can do for your computer source. From all the thousands of combinations discussed in this forum, that's the one thing everyone agrees about. It is the most important thing to do for your computer after all! Whether spending £1K is worth, that's something only you can decide. To me, spending 20% on computer and 80% on LPS gives much better results than spending 80% on computer and 20% on a LPS.

 

The impact of a good LPS has been discussed many many times here. You just need to do some reading about it. 

 

Does that hold true regardless of how the data leaves the server to the streamer?

 

Thanks for the response.

[dctom]

16 hours ago, George Hincapie said:

 

Can you describe how it is better? What did it replace - SMPS? How much of an improvement? How is your server connected to your streamer? Ethernet or USB?

 

 

It is difficult to isolate one component's contribution when it is part of an overall drive for improvement when the whole system's  sound has improved. However out of curiosity I put a switch mode  ps in instead of the SJ DC3. The music collapsed some what, scale, depth, a bit less dynamic. There was a definite improvement in the overall SQ, so as far as I am concerned a worthwhile upgrade.

[dctom]

20 hours ago, Tommd64 said:

Compared the prices and see whats going on. I ll say without hearing please buy a Farad3, it's also possible to order it with highgrade componends..or better dc cable. Coz he 700p for a Sean Jacobs is a lot of money for a realitively simple PS. Compared to a Farad or SotM.

 

The SJ DC3 does produce 5 amp output, so is more versatile if I change hardware.

[Nenon]

36 minutes ago, George Hincapie said:

Does that hold true regardless of how the data leaves the server to the streamer?

 

There is a short and a long answer to that. I'll go with the short answer as we need to get back on topic in this thread: YES. 

[Blackmorec]

3 hours ago, davide256 said:

 

?? This is a long thread but you should probably go back and refresh your memory that the OP originally had Zenith server but abandoned it

for an AL/NUC solution. While a good power supply is essential for resolution, the drive and musical sense you describe come from the hardware/software

platform capabilities

Nothing to do with memory. Experience 😉. I own (ed) a Zenith SE, Statement, DC-2, 3 x DC-3s and had an SoTM SPS-500 on a 6 weeks loan. The OP may have preferred an AL/NUC solution above a Zenith SE like I prefer the Statement but like both the Zenith SE and Statement the AL/NUC still requires a great power supply to sound its best.  How did I reach the conclusion I did?  Both the SE and Statement provide exceptional musical sense and rhythmic drive.....but guess what you get when you use a DC-2 or DC-3 to power the system’s network components?  Yes,  an even greater sense of the depth,  drive, excitement and rhythmic interplay between musicians that makes the SE and Statement so appealing. 

[austinpop]

14 hours ago, drjimwillie said:

 I read about trying to make my system sound better for a couple of hours every day, unfortunately this is more time than I get to listen to my system. 😕

 Do that every day for a few years and some of the references get lost.  And I’m sure some of the metaphors get mixed. 

 I remember reading about link aggregation but I’m not sure if that’s the basis for a better sound? 

 

 I thought it was kind of a wacky idea, which is why I asked the question here. 

 I was wondering if you could use or how you could use a switch in conjunction with the bridge  

 I would try my idea, but I’m having a couple of technical difficulties in my system is not conducive to critical listening at the moment.  When I try it I will let you know. I am surprised that no one else has tried it up to this point. 

 

 I am not sure how the link to the JS article proves my point. I cannot find a correlation after reading it.   Would you please explain what you’re thinking? 

 

 I have learned that I should try the switch after the server rather than before it. Although the switch before it did increase the SQ.  I think the increase in SQ is because the ethernet cable from the router to the server is 30 feet long and it benefited from the signal regeneration.

 

Thank you 

 

Really briefly, because this is going off-track here, I will say 2 things. First, it doesn't appear that anyone here has tried what you're proposing, so feel free to try it, and report back whether it a) works at all, and b) does it improve SQ? 

 

The second point is to suggest you exercise caution and do some reading on this. When you connect multiple network interfaces on the same machine to the "network," there are multiple ways to do it, and some do not have a happy ending.

1. Each interface gets its own IP address. This is called a multihomed system. If the interfaces are connected to the same subnet, behavior can be unpredictable, and is not considered correct practice. Just do a google search on "connect multiple ethernet ports to same subnet." That said, you may be able to get it to work on your particular OS.
2. The ports are bridged, and so the machine only gets a single IP address, associated with the bridge. Here again, attaching all the ports to the same subnet will have unpredictable behavior. I would argue that since bridging puts all the adapters in promiscuous mode, this might cause increased interrupt load on the system. 
3. The ports are aggregated, using a feature called link aggregation. Here again, the system gets a single IP address. This may be the most promising option for what you're proposing.

#2 and #3 require some setup and configuration, so it's not as simple as just connecting multiple ports physically. As far as I know, none of the common endpoints (Rendu, SOtM) or OSes (AL, Euphony, etc) have an option to enable link aggregation. You'd have to drop into the command line as root, and manually do it.

 

Have fun!

[austinpop]

12 hours ago, johndoe21ro said:

Great review and a long expected one, Austinpop!
In a couple of words could you describe the differences between how each one (sNH-10G switch and then the tX-USBultra) reacted with the different PS?

 

I already wrote that in the review. Not sure why you're asking me to repeat myself? Or are you asking if I heard different things when comparing PSUs on the switch vs. the tX-USBultra? If so, the answer is no. The same findings held for both.

 

12 hours ago, johndoe21ro said:

Was one of the SOtM's more susceptible than the other?

 

In my setup - and I hesitate to generalize - the tX-USBultra benefits the most from the superior PSU. That's not to say the switch did not also benefit, but the way to think about it is if you had only one good PSU, where would you put it? My answer is - experiment! What's right for my system isn't necessarily right for yours.

 

12 hours ago, johndoe21ro said:

Also, I understand that you have a tX-USBultra with a 12V input (as I do). Did you try to internally switch it to 9V? If so, does it sound better with the 9V or the 12V? Thanks!

 

Please search back on this thread. I think I've described this in the distant past. Bottom line - I think 12V sounds best. Let it be.

[austinpop]

4 hours ago, George Hincapie said:

Can I ask generally, for those of you who are feeding your server output via ethernet direct to your streamer, and have upgraded the power supply for your music servers from SMPS to LPSU, what impact did that have? What was the scale of perceived improvement if there was any? Conversely, did any of you go through this process and fail to note any improvement?

 

While I have experienced the benefit of replacing the stock SMPS with an LPS on other components, I have never done that particular comparison on a music server. I think if you do some creative searches on this thread, using terms like Corsair and HDPlex, you should find @romaz's and others posts about replacing an SMPS PSU with an LPSU.

 

4 hours ago, George Hincapie said:

Trying to decide whether to spend £1K on a LPSU for my server, and am trying myself in knots. I don't want to commit unless there's going to be a level of improvement to justify it.

No one can  give you any assurances. You do realize this, don't you? All you can do is try it, and hear for yourself.

 

2 hours ago, Nenon said:

 

Upgrading the stock SMPS to a good LPS is the single best thing you can do for your computer source. From all the thousands of combinations discussed in this forum, that's the one thing everyone agrees about. It is the most important thing to do for your computer after all! Whether spending £1K is worth, that's something only you can decide. To me, spending 20% on computer and 80% on LPS gives much better results than spending 80% on computer and 20% on a LPS.

 

The impact of a good LPS has been discussed many many times here. You just need to do some reading about it. 

 

+1

[austinpop]

4 hours ago, davide256 said:

the OP originally had Zenith server but abandoned it for an AL/NUC solution.

 

Please don't drag me into this,   or put words in my mouth. Yes, I moved on from the SE, but that was because I wanted to experiment with Roy's findings about servers and NUCs, not because I thought the SE was bad. And it is indeed true that one of the major reason's the SE (and Statement) sounds so damn good is because of the Sean Jacobs PSUs within.

[FredM]

11 minutes ago, George Hincapie said:

 

Eh? My question is on topic?! What's the long answer.

 

Still not satisfied anyone has properly answered what I've asked and that's concerning. Because what that says is you can't tell. Or am I being unfair?

 

Perhaps I need to establish whether I can buy one sale or return. If it does nothing then I can avoid an expensive mistake.

 

Really surprised I haven't been inundated with positive responses.

 

Imho this forum just isn’t a store where you can ‘demand’ after sale service. Please be happy with the answers people give you in their spare time. About 2 hours ago Nenon gave you a great answer, if want more, please use search