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

[BigAlMc]

1 hour ago, limniscate said:

@BigAlMc

 

I'm not sure there's a difference between a modded sMS-200 and a sMS-200 Ultra.  One thing Lee did say to us at RMAF is that, in theory, each sCLK-EX board should be using the same frequency for all of the taps.  I think he also said that my setup with two sCLK-EX boards (one in the sMS-200 Ultra and one in the dX-USB Ultra) should be better than @austinpop's setup, but we haven't noticed any significant differences.

Thanks Eric, 

 

Appreciate the feedback. 

 

Cheers, 

Alan 

[magnuska]

 Hello!

 

I´m on the fence of also getting a audiophile switch. 

I have today a 3G router/switch with 5 ports. I wonder how important would adding another switch be

when I have a 3G router that is not connected to a wired ethernet. 

 

IF I would need a switch I have been thinking about 3 different scenarios. I already have a LPS-1 to power it.

One obvious move would be sending a TX-usb ultra to sotM and have it modified there, 

I could also choose either the AQVOX switch 8  or the newly updated switch8 SE model with a supposed superior SQ and doubled the price , http://www.myhifishop.de/Devices/AQ-SWITCH-SE-Audiophile-High-End-Network-Switch-audio-LAN-isolat::87.html?language=en

 

The soTm modified switch features a really good clock but the new SE seems to have a good clock and more goodies built in.

 

Thanks

 

Magnus

[austinpop]

2 hours ago, BigAlMc said:

A few questions for @austinpop and other Trifecta users.

 

I have a small window, so will try and answer.

 

1 hour ago, limniscate said:

I'm not sure there's a difference between a modded sMS-200 and a sMS-200 Ultra. 

 

Yes, I think in one of our early experiments, we compared my modded sMS-200 with Eric's sMS-200ultra, and couldn't hear a meaningful difference. I think the key is to keep the clock cables short. Don't use the 1m SMB cables SOtM sends you. I'm using 0.3m Digikey SMB cables. If you can manage it, they also have 0.15m (6") cable lengths.

 

1 hour ago, limniscate said:

One thing Lee did say to us at RMAF is that, in theory, each sCLK-EX board should be using the same frequency for all of the taps. 

 

What Lee was saying is that there are always penalties with any approach. While the sCLK-EX board's 4 taps can be programmed to distinct frequencies, there is some interference (noise) when you do so, compared to generating the same frequency. I took that to mean that really hard core users may choose to use a separate sCLK-EX board per clock to get the ultimate performance. He was talking in terms of the technology. I don't believe we know how measurable this effect is, how much it practically impacts SQ, and whether anyone has employed this extreme approach!

 

1 hour ago, limniscate said:

I think he also said that my setup with two sCLK-EX boards (one in the sMS-200 Ultra and one in the dX-USB Ultra) should be better than @austinpop's setup, but we haven't noticed any significant differences.

 

Exactly, which is why I'm a big proponent of maximizing value for the lowest cost by using a single sCLK-EX board where possible, like my trifecta.

 

2 hours ago, BigAlMc said:

2. Does the Trifecta improve Tidal streaming as much as locally stored files? Or is Tidal still at the mercy of the Router. Unlike @romaz (who is going to be sorely missed here ) I doubt i'm in a position to get the router placed close enough to the other stuff to run a different clock on it. 

 

Yes it does. It raises the SQ of both local and Tidal streaming. In my setup, a gap still remains where local streaming sounds better than the identical content from Tidal. Some day, if I can replace my router clocks, perhaps I can achieve parity. 

 

1 hour ago, BigAlMc said:

The other mental arithmetic I'm trying to balance is that my Directstream DAC is apparently best on I2S which is why I got the SU-1. But if I make the leap to the Trifecta I'd be looking to go reclocked USB into the DAC to see if the Trifecta magic was sufficient to outperform the SU-1/I2S. 

 

This gets complicated. Since SOtM will not modify the SU-1 to accept sCLK-EX, I'm not aware of an end-to-end I2S solution. If you're handy, they will program the sCLK-EX clock points for you, and you would have to figure out how to mod the SU-1 to accept them. Roy did something like this for his ISO-R.

 

The other possibility is to go AES. Does the DirectStream DAC perform better with AES or S/PDIF rather than USB? If so, you could get a dX-USB HD Ultra. You would need to ensure that SOtM modify not just the USB clock, but also the 2 sample rate clocks in the dX. Yes, the dX is an older XMOS platform than the SU-1, but Lee told us in Denver that he doesn't feel that the SQ improved with the newer XMOS, just features like higher sample rate support.

 

In that case, you could do something like this:

1. modded switch > sMS-200ultra > dX-USB HD Ultra > AES > DAC, and also try
2. modded switch > sMS-200ultra > ISO-Regen > dX-USB HD Ultra > AES > DAC

In contrast, my trifecta would be cheaper, but would go into the USB input of the DAC:

1. modded switch > modded sMS-200 > tX-USBultra > USB > DAC, or 
2. modded switch > modded sMS-200 > ISO-Regen > tX-USBultra > USB > DAC

Choices, choices!

[toetapaudio]

4 hours ago, thyname said:

Anyone knows where to purchase a Aqvox Ethernet switch in USA?

 

Hi thyname, only direct from Germany. There are two versions now in case you weren’t aware.

[BigAlMc]

2 hours ago, austinpop said:

In that case, you could do something like this:

1. modded switch > sMS-200ultra > dX-USB HD Ultra > AES > DAC, and also try
2. modded switch > sMS-200ultra > ISO-Regen > dX-USB HD Ultra > AES > DAC

In contrast, my trifecta would be cheaper, but would go into the USB input of the DAC:

1. modded switch > modded sMS-200 > tX-USBultra > USB > DAC, or 
2. modded switch > modded sMS-200 > ISO-Regen > tX-USBultra > USB > DAC

Choices, choices!

Thanks Rajiv, 

 

Yeah there are certainly a few options but my thinking was to try and raise the quality of the USB signal (Trifecta at first and then add a better clock) in the hope that it betters the SQ despite I2S theoretically being the optimum input for my DAC. 

 

I saw that @Bamber went the road of getting an SU-1 modded for sCLK-ex input and kudos to him on it. But I was kinda aiming more to replace my set up rather than added more components to the chain. Hence my initial attraction to an SMS-200ultra & modded switch with reference clock to follow.

 

Bases on feedback so far it sounds like that's probably too simplistic and I might need more. 

 

Oh well - plenty fun researching and trying to suss out a potential plan. 

 

Thanks again for the feedback. 

 

Cheers, 

Alan 

[rickca]

1 hour ago, toetapaudio said:

There are two versions now in case you weren’t aware.

Thanks, I didn't know about the new AQ-SWITCH-SE.  It's a good thing.  The original AQ-SWITCH just wasn't expensive enough. 

[thyname]

4 hours ago, toetapaudio said:

 

Hi thyname, only direct from Germany. There are two versions now in case you weren’t aware.

 

Wow, that’s expensive once applied the exchange rate to USD and shipping from Europe.

 

i read somewhere that @JohnSwenson recommend a more reasonably price Ethernet switch (or two), but cannot find it

[seeteeyou]

http://www.myhifishop.de/Devices/AQ-SWITCH-SE-Audiophile-High-End-Network-Switch-audio-LAN-isolat::87.html?language=en

Quote

- The internal power supply circuits are now carefully optimized with additional Ultra-Fast capacitors.

 

Holy smoke, are we running Yottabit Ethernet with this kinda speed now?

 

Ultrasound Beyond the Speed of Light - Acoustics.org
http://acoustics.org/pressroom/httpdocs/150th/Mobley.html

 

 

Seriously, this particular one doesn't seem to be small enough to fit inside a switch

 

https://cdme.osu.edu/sites/cdme.osu.edu/files/uploads/0006a_capacitorbank.pdf

 

Now I wonder if those Ultra-Fast capacitors were any good if they're powering something else.

[d_elm]

2 hours ago, thyname said:

 

Wow, that’s expensive once applied the exchange rate to USD and shipping from Europe.

 

i read somewhere that @JohnSwenson recommend a more reasonably price Ethernet switch (or two), but cannot find it

Those would be the Netgear FS108 and FS105, both version 3 using 12V rather than v2 at 7.5v.  An FS105 v3 with stock SMPS is noiser than an FMC, MC100CM, powered by LPS-1.  Expect the FS105 would be improved by a good power supply but did not have one to try.

[Forehaven]

Just curious, are you all using these switches to simply get an ethernet feed to small computers/streamers nearer to your audio setups?  Are these switches run from a feed from a router or secondary switch? 

Thanks

[JohnSwenson]

There are two aspects to sound quality from a network connection:

1) clocking, this includes the phase noise of the clock in the switch, but it also includes the phase noise from other network devices connected to the switch,

 

2) leakage current. This includes the leakage of the power supply feeding the switch, but it also includes the leakage from power supplies powering other network devices. This leakage current can go right through Ethernet transformers.

 

The "audiophile" switches, use a much better clock for the local switch, but don't do anything about the phase noise coming in over the wire from other equipment or the network leakage.

 

The mentioned recommendation of mine is for dealing with #2, leakage current. What I found is that it is easy to get rid of the leakage current from the local power supply ( the PS for the switch)) by grounding the negative of the PS output. This will get rid of its leakage current. But you still have to deal with the leakage coming in over the network. IF you use the FS108 or FS105 (see above for the models) AND you ground the negative of the supply on the FS108 oe FS105 you will block the leakage from both the local PS AND all the others coming into the switch.

 

Using the switch without the grounding does not improve anything. Using the grounding with other models switches gets rid of the leakage from the local PS, BUT still lets all the others through.

 

Note this does not get rid of the clocking issues, just the leakage.

 

Personally I would get one of these switches and use it just to drive the computer/streamer whatever is the last point where a network connection is made to the audio system. Ground the negative of its output and you have blocked network leakage from getting into  your system.

 

See the SMPS grounding thread in the UpTone forum for details on how to do this grounding.

 

The fun part is going to be figuring out how to deal with the network clocking issues, that is much more difficult, but I am definitely working on how to deal with that.

 

John S.

[ElviaCaprice]

On ‎10‎/‎14‎/‎2017 at 4:06 AM, Hauser said:

 

Hi ElviaCaprice,

 

I've been following some of your posts and am considering modifying a NUC.  One of your previous posts mentioned discussing with May at Sotm whether it was possible to use a Sata 2 cable to connect to the Sata 3 Mobo; I assume so you could use their Sata 2 filter.  Is this what you have done?

 

Regards,

Martin

Hi Martin,

I received an answer from May about the SATA II filter.  Which filters both the data output/input and power input.  Unlike the SATA III filter which filters only the power input. 

The answer for whether the SATA II filter can work on a SATA III device is sometimes.  It's hit and miss.  May did not distinguish between a SATA II port on a mobo or SATA III.  It's all determined by the device itself.

 

Hope that helps.  I'll probably give it a try, the SATA II filter.  I don't find the SATA III filter of any value since I already use a high quality separate power supply, thus I am interested in filtering the DATA line. 

 

But if you did as Roy suggested and work with a mobo that can operate two tXUSBexp PCIe cards, then the cards filtering would suffice along with sCLK-EX clocking and clean power.  The only issue I see with this is having to use USB 3.0.  So now you couldn't just run a SATA cable to a HDD free standing.  I tried looking around for a data only SATA to USB 3.0, but they all incorporate the power input on the HDD.  Maybe this wouldn't be so critical if the clean power to the tXUSBexp is powering the HDD??

[AmusedToD]

18 hours ago, austinpop said:

Let's all take a step back, please. 

 

I am not advocating for any brands in particular, and as long the discussion is relevant, I'm fine with any components being mentioned. If these are super expensive components being discussed, like the Blu-Dave combo, or some of the dCS DACs, I know I won't be buying them in this lifetime.

 

On the other hand, cool technologies do tend to trickle down. So if a year from now, dCS bring out a cheaper DAC, as they are rumored to be working, or a cheaper M-scaler/DAC combo arrives, then this research becomes interesting.

 

The key word here is research. I believe many on this thread - myself among them - wonder if eventually a single box will obviate the need for our front end medleys.

 

The dCS network bridge is intriguing on that score, as it is priced at a point comparable to what I've eventually spent on my trifecta medley.

 

So at this point, I am just curious to see how it compares on SQ. I like comparisons.

 

The dCS network bridge was primarily designed to bring the older dCS DAC's (Elgar, Delius, Paganini, Debussy, etc.) into the streaming era. Of course it can be used with other manufacturers' DACs, but it will always perform at its best with DACs that feature dual AES connections and have a word clock input so that the DAC can be slaved to the NB. 

 

On the other hand, the SOtM gear was designed primarily to be used with USB DACs.

 

Your Codex does not have dual AES inputs, that's why I suggested testing both your trifecta and the NB with the Rossini DAC by simultaneously hooking both into the USB and dual AES inputs of the DAC. This way we can also establish how good the USB input of the Rossini is.

 

[AmusedToD]

17 hours ago, austinpop said:

 

I'll see if it's possible. My dealer was rather interested in the trifecta, so he should be amenable to spending a couple of hours doing comparisons on his setup. Even if he knows that your explanation of the USB port does make the NB less interesting for me.

 

So we'll see, if I can make a comparison happen, that will at least be useful data for this thread.

 

I am inclined to do the test with the QX-5 DAC, since it's one I'm more familiar with, and has both USB and AES inputs. The dual AES-thing could be a one off with a dCS DAC, just for curiosity.

 

Will keep y'all posted.

 

Yes, the QX-5 is a great candidate too as it has both an USB input and dual AES inputs. This way a very clear A/B test can be done by simply changing the inputs of the DAC. You can even hook up your headphones directly to it. 

[thyname]

6 hours ago, JohnSwenson said:

 

 

Personally I would get one of these switches and use it just to drive the computer/streamer whatever is the last point where a network connection is made to the audio system. Ground the negative of its output and you have blocked network leakage from getting into  your system.

 

 

 

Excellent write up!

 

For the quoted phrase above, are you saying that when connecting the endpoint / streamer (such as microRendu or sMS-200) to that switch, nothing else should go to that switch (i.e. Apple TV, Xbox, etc)?

[thyname]

9 hours ago, d_elm said:

Those would be the Netgear FS108 and FS105, both version 3 using 12V rather than v2 at 7.5v.  An FS105 v3 with stock SMPS is noiser than an FMC, MC100CM, powered by LPS-1.  Expect the FS105 would be improved by a good power supply but did not have one to try.

 

Thanks!

 

This one?

 

https://www.amazon.com/NETGEAR-ProSAFE-FS108NA-8-Port-Ethernet/dp/B00002EQCS/ref=sr_1_1?ie=UTF8&qid=1508161224&sr=8-1&keywords=netgear+switch+fs108

 

Then, either:

 

1 - use a 12v linear power supply, or

2 - do the grounding think on existing power supply it comes with as @JohnSwenson describes

 

Correct?

 

Then since I am using an sms-200 (soon to be upgraded to either ultraRendu or sms-200ultra), the clocking issue it taken care of by these streamers / endpoints?

 

Thanks!

[austinpop]

7 hours ago, JohnSwenson said:

There are two aspects to sound quality from a network connection:

1) clocking, this includes the phase noise of the clock in the switch, but it also includes the phase noise from other network devices connected to the switch,

 

2) leakage current. This includes the leakage of the power supply feeding the switch, but it also includes the leakage from power supplies powering other network devices. This leakage current can go right through Ethernet transformers.

 

The "audiophile" switches, use a much better clock for the local switch, but don't do anything about the phase noise coming in over the wire from other equipment or the network leakage.

 

The mentioned recommendation of mine is for dealing with #2, leakage current. What I found is that it is easy to get rid of the leakage current from the local power supply ( the PS for the switch)) by grounding the negative of the PS output. This will get rid of its leakage current. But you still have to deal with the leakage coming in over the network. IF you use the FS108 or FS105 (see above for the models) AND you ground the negative of the supply on the FS108 oe FS105 you will block the leakage from both the local PS AND all the others coming into the switch.

 

Using the switch without the grounding does not improve anything. Using the grounding with other models switches gets rid of the leakage from the local PS, BUT still lets all the others through.

 

Note this does not get rid of the clocking issues, just the leakage.

 

Personally I would get one of these switches and use it just to drive the computer/streamer whatever is the last point where a network connection is made to the audio system. Ground the negative of its output and you have blocked network leakage from getting into  your system.

 

See the SMPS grounding thread in the UpTone forum for details on how to do this grounding.

 

The fun part is going to be figuring out how to deal with the network clocking issues, that is much more difficult, but I am definitely working on how to deal with that.

 

John S.

 

John,

 

Is grounding necessary if using an LPS-1 to power the switch?

[d_elm]

8 minutes ago, austinpop said:

 

John,

 

Is grounding necessary if using an LPS-1 to power the switch?

I tried that, grounding of SMPS is needed with LPS-1 for high impedance noise.

[ducatirider]

I'm having trouble bridging ethernet ports with the new 64 bit Jriver MC23.  For some reason my network doesn't see the NAS but when type in the NAS IP address in a browser it sees it.  I used to have this problem before but I fixed it but re-starting msconfig in normal mode but now that doesn't work.  It does work if I use the 32 bit version.  So weird.  I have Fidelizer and Audiophile Optimizer running under Windfows 10 on a Surface Pro if that matters.

[austinpop]

1 minute ago, d_elm said:

I tried that, grounding of SMPS is needed with LPS-1 for high impedance noise.

 

Ah - you mean the SMPS energizing supply for the LPS-1 - aka Meanwell?

 

Also, @JohnSwenson - do you know what the FS105/FS108 are doing that is unique to suppress leakage through the transformers in the signal path? 

[d_elm]

1 hour ago, thyname said:

 

Thanks!

 

This one?

 

https://www.amazon.com/NETGEAR-ProSAFE-FS108NA-8-Port-Ethernet/dp/B00002EQCS/ref=sr_1_1?ie=UTF8&qid=1508161224&sr=8-1&keywords=netgear+switch+fs108

 

Then, either:

 

1 - use a 12v linear power supply, or

2 - do the grounding think on existing power supply it comes with as @JohnSwenson describes

 

Correct?

 

Then since I am using an sms-200 (soon to be upgraded to either ultraRendu or sms-200ultra), the clocking issue it taken care of by these streamers / endpoints?

 

Thanks!

That switch is the one, 12V.  I used the provided SMPS and grounded the -ve output but try a better power supply.  I do not know if clocking is about jitter or noise.  This switch with good power needs to be compared to an FMC with good power.  Also, FMCs and switches can be modified for lower noise output.  SOtM has been doing this and there may be others.

[octaviars]

8 minutes ago, d_elm said:

Also, @JohnSwenson - do you know what the FS105/FS108 are doing that is unique to suppress leakage through the transformers in the signal path? 

 

I suppose one would need to check what transformers they use and how they are connected to the groundplane.

 

This is an example of a Pulse transformer that I have seen in 10/100 switches but if this specific connection is the one in a FS105/108 I dont know. 

[Forehaven]

8 hours ago, JohnSwenson said:

There are two aspects to sound quality from a network connection:

1) clocking, this includes the phase noise of the clock in the switch, but it also includes the phase noise from other network devices connected to the switch,

 

2) leakage current. This includes the leakage of the power supply feeding the switch, but it also includes the leakage from power supplies powering other network devices. This leakage current can go right through Ethernet transformers.

 

....John S.

Thank you so much John!! NOW I see why the emphasis of adding another switch and ps.  Very clear and concise that even I understood lol

 

I'll def work on this aspect.

 

The 5 port/7.5vdc ps...will the LPS=1 at 7 work here John?

[octaviars]

17 minutes ago, Forehaven said:

 

The 5 port/7.5vdc ps...will the LPS=1 at 7 work here?

 

If there is a DC/DC converter that most of this stuff have internaly I think it will work with 7V but perhaps John can answer this.  

 

For example my router runs on 19vdc and I tried it on my adjustable PSU and it would work all the way down to 12vdc so 0,5vdc lower on a switch might just work.

[toetapaudio]

1 hour ago, austinpop said:

 

John,

 

Is grounding necessary if using an LPS-1 to power the switch?

 

https://www.computeraudiophile.com/forums/topic/35191-review-dc-power-supplies-for-aqvox-audio-grade-switch/