Jump to content
IGNORED

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


Message added by The Computer Audiophile

Important and useful information about this thread

Posting guidelines

History and index of useful posts

Most important: please realize this thread is about bleeding edge experimentation and discovery. No one has The Answer™. If you are not into tweaking, just know that you can have a musically satisfying system without doing any of the nutty things we do here.

Recommended Posts

1 hour ago, rickca said:

Yes I saw that comment from May.  So I'm curious about the source of @JJ Braham's information that these switches have sCLK-EX and sCLK-OCX10 clocks in them.  It doesn't make sense.

Agree, BTW, SOTM are also working on their own audiophile switches which might see the light early in 2018

Link to comment
7 hours ago, Johnseye said:

 

I didn't realize that.  Then what's the point?

 

If you think about all the frequencies that typically apply, the system frequencies - mobo (25 MHz), Ethernet (25 MHz), USB (12/24) are needed continuously, and don't lend themselves to the switched approach.

 

With the sample rate frequencies, it does make sense, because at any given time, you're only playing one stream through the device, so you either need one or the other. So making do with one clock point for both is eminently practical.

 

EDIT: Oops - I see that @elan120 and @lmitche already made this point eloquently.

Link to comment
11 minutes ago, austinpop said:

I plan to do another eval at home once the NB firmware update to enable USB output comes out.

 

No talk about when this will happen?

Main system
TAD D1000mk2, TAD M2500mk2, TAD CE-1, Ansuz Mainz 8 C2, Ansuz Darkz D-TC, 
Qobuz Studio -> Roon ROCK on NUC -> Uptone etherREGEN -> dCS Network Bridge -> AES/EBU -> DAC
HD Plex 200W PSU (4 rail for ISP fiber, router, etherREGEN and NUC)
 
Second system
Qobuz Studio -> Devialet Silver Phantom, Devialet Tree
Link to comment
22 minutes ago, austinpop said:

I spent an afternoon at my dealer's listening to my trifecta medley in his system, along with the dCS Network Bridge.

I know you enjoy major orchestral pieces.  How did the dCS NB compare to the trifecta in busy passages when nearly all the instruments are playing?

Pareto Audio AMD 7700 Server --> Berkeley Alpha USB --> Jeff Rowland Aeris --> Jeff Rowland 625 S2 --> Focal Utopia 3 Diablos with 2 x Focal Electra SW 1000 BE subs

 

i7-6700K/Windows 10  --> EVGA Nu Audio Card --> Focal CMS50's 

Link to comment
7 hours ago, JohnSwenson said:

It seems a lot of people are getting confused by my posts on the subject of network leakage, I will try and state things in a more concrete manor.

 

This is relating to a switch which is what your network endpoint into your audio system is connected, this may be streamer (microRendu etc) a laptop or other computer Mac mini, PC etc). We shall call this the audio endpoint (AE)

 

Leakage current can get into the AE through the switch in two ways, from the power supply powering the switch, or from the cable connected to the rest of the network. The leakage coming from the network comes from the SMPS powering THOSE devices.

 

If the switch connected to the AE is powered by an SMPS, grounding the negative of output of the SMPS will shunt the leakage from that supply, but the leakage from the NETWORK will still go through. There is one exception, see the next section.

 

IF the switch is one of FS105 and FS108, grounding the negative of the supply will get rid of BOTH the leakage from the SMPS AND the network leakage. NOTE, this ONLY happens for these two switch types. Grounding the supply to a different switch type does NOT block network leakage.

 

If you are powering a switch from a linear supply, this gets rid of the leakage going through the PS of the switch, but NOT the network leakage. The only way to get rid of the network leakage is to use one of the above switches AND ground the negative of the supply powering the switch, no matter WHAT that supply might be. (linear, SMPS, LPS-1 etc)

 

If you are using an LPS-1 to power the switch, see the above rules for ANY supply. ANY supply includes the LPS-1. Thus IF you have one of the two named switches and you are powering the switch from an LPS-1, you must ground the output of the LPS-1 in order to block the network leakage. This will only work with one of those two switches. Grounding the output of the LPS-1 will NOT block leakage if you are using some other switch. It will not help if the LPS-1 is driving some other type of device. Thus there is no reason to ground the output of an LPS-1 if it is NOT driving one of the above named switches.

 

There is one exception to the last point. IF the LPS-1 is driving an ISO REGEN there can be a situation where the whole audio system is floating with respect to earth ground and a charge can build up which can show up as clicks and pops. ONE earth ground in such a system can alleviate this. ONE way to do this is to ground the negative of the supply powering the ISO REGEN. If this supply is an LPS-1 then you can try grounding the output of the LPS-1 to see if it fixes the clicks and pops.

 

Grounding the INPUT to the LPS-1 can help in other situations by shunting the high impedance leakage.

 

I hope this makes things clear, I'm running out of ways to say this.

 

John S.

 

 

 

 

 

 

 

What if the power supplies are connected to a floating isolation transformer? Where should the grounding go?

 

Thanks

 

Link to comment
1 minute ago, rickca said:

I know you enjoy major orchestral pieces.  How did the dCS NB compare to the trifecta in busy passages when nearly all the instruments are playing?

 

Very favorably. We used two pieces for the A-B comparisons:

  1. Mahler Symphony 5 - Zander (DSD64) - 3rd movement (scherzo)
  2. Sibelius Symphony 5 - Vänskä (DSD64) - 1st movement

In both, the trifecta seemed to render the french horns more dynamically - my "floating off the chair" metaphor. In the Sibelius, one of the key tests is to distinguish the trumpet and flute playing the same theme. I'd say resolution wise, the trifecta and the NB were comparable.

 

The NB just had a slightly more smooth and natural tonality, while the trifecta rendered a bigger, deeper soundstage.

 

Honestly, I don't know if I'd call either "better" than the other.

Link to comment

OK some of the difference might just be attributable to USB vs AES, right?  As you say, you need to revisit this once the dCS NB has USB output.

 

I guess you listened to the Rossini without the Rossini Master Clock?  It's just a wee bit more money.

Pareto Audio AMD 7700 Server --> Berkeley Alpha USB --> Jeff Rowland Aeris --> Jeff Rowland 625 S2 --> Focal Utopia 3 Diablos with 2 x Focal Electra SW 1000 BE subs

 

i7-6700K/Windows 10  --> EVGA Nu Audio Card --> Focal CMS50's 

Link to comment
12 hours ago, ElviaCaprice said:

I didn't ask May which was which.  But I am using 3 points off the sCLK-EX, one set to 12MHz, the other 2 to 25MHz (at different voltage???)  So now I am going to have to ask may what the 12MHz point was used on???

Here is the answer from May concerning the exact details of my NUC modified sCLK-EX.  This is just to clarify and be exact.  When you send a mobo to SOtM, they will change out the clocks accordingly and check for frequency/voltage.  No need to tell them.

12Mhz goes to USB(2.5v), and 25Mhz goes to the main(0.8v) and lan(1v), and they can’t be combined to the one physically cause they require the different power voltages. if they requires the same voltage, it can be combined.

 

(JRiver) Jetway barebones NUC (mod 3 sCLK-EX, Cybershaft OP 14)  (PH SR7) => mini pcie adapter to PCIe 1X => tXUSBexp PCIe card (mod sCLK-EX) (PH SR7) => (USPCB) Chord DAVE => Omega Super 8XRS/REL t5i  (All powered thru Topaz Isolation Transformer)

Link to comment
13 minutes ago, ElviaCaprice said:

Here is the answer from May concerning the exact details of my NUC modified sCLK-EX.  This is just to clarify and be exact.  When you send a mobo to SOtM, they will change out the clocks accordingly and check for frequency/voltage.  No need to tell them.

12Mhz goes to USB(2.5v), and 25Mhz goes to the main(0.8v) and lan(1v), and they can’t be combined to the one physically cause they require the different power voltages. if they requires the same voltage, it can be combined.

 

 

This still begs the question of why the mobo USB? Unless you're using it for something. 

Link to comment
9 minutes ago, Johnseye said:

 

This still begs the question of why the mobo USB? Unless you're using it for something. 

I agree, but as it so happens I do play Hi res videos via USB 3.0 at times, so it does come in handy.  Really I had no other use for those points.  It sounds fantastic.  I suppose if someone was into streaming via Ethernet that one point could have been used for Router?  Switch?

(JRiver) Jetway barebones NUC (mod 3 sCLK-EX, Cybershaft OP 14)  (PH SR7) => mini pcie adapter to PCIe 1X => tXUSBexp PCIe card (mod sCLK-EX) (PH SR7) => (USPCB) Chord DAVE => Omega Super 8XRS/REL t5i  (All powered thru Topaz Isolation Transformer)

Link to comment
3 hours ago, austinpop said:

dCS Network Bridge vs. Trifecta Medley

 

I spent an afternoon at my dealer's listening to my trifecta medley in his system, along with the dCS Network Bridge.

 

System

 

His system was as follows:

  • AC: Everything plugged into Shunyata Triton
  • DAC: Ayre QX-5
  • Amplification: Nagra Classic Preamp and Amp
  • Speakers: Rockport Avior

Digital Sources

  • Apple Airport Media bridge with following connected to the ports
    • Roon Core on Synology NAS
    • dCS NB with AES output to QX-5 
    • my trifecta medley with USB output to QX-5

Trifecta Medley (just to remind)

  • Zyxel switch (sCLK-EX mod) > SOtM dCBL-Cat7 > SOtM sMS-200 (sCLK-EX mod) > USPCB > ISO-Regen > USPCB > SOtM tX-USBultra > Lush USB cable
  • Reference Clock: Cybershaft Premium OCXO OP-14 10MHz clock driving the tX-USBultra

Listening Impressions

 

I'll keep this short. This was not a slam dunk for either source. Both the NB and the trifecta had their strengths and weaknesses.

  • The dCS NB clearly had a more refined and natural tonal balance. It sounded smoother, richer, and would probably be less fatiguing over long listening sessions.
  • The trifecta, on the other hand, sounded a bit more dynamic, had a bigger soundstage, and more air, but did display a hint of coarseness and edginess. 

Again, this is all relative. I must say I was very pleasantly taken by the NB. I have some conjectures that would explain what I was hearing, but I don't want to speculate. This was my first lengthy exposure to the dCS sound, and I found it wonderful. I am not going to make any pronouncements at this time, as there were too many variables different from my system.

 

I plan to do another eval at home once the NB firmware update to enable USB output comes out. During that eval, it would also be interesting to audition it in Eric's setup, comparing his trifecta driving the Yggy via AES.

 

Coda

 

The final part of our session was to kick back and listen to the ($24k) Rossini DAC, fed via Ethernet, and configured as a Roon endpoint. Wow! This is an incredible piece. The improvement from the trifecta/NB + QX-5 to the Rossini was not subtle. It's hard to describe something that just sounds right - natural, organic, relaxed, but with incredible resolution.

 

Just so you guys don't start sending me PMs - this session was just for fun. I am not, and will likely never be - in the market for a $24k DAC.

 

Of course, next time I want to listen to the Vivaldi and the Vivaldi Master Clock!

 

@austinpop, I guess you understand my point now in wanting to upgrade to a dCS DAC. Yes, they are bloody expensive, but they play in a different league altogether. Thank you for this short review! 

 

Intersting findings regarding the Network Bridge vs trifecta comparison. I believe your Trifecta is even more expensive than the NB as it includes a 10Mhz Reference Master clock. Pretty large number of boxes in the chain plus cables to connect them plus PSU’s to power them. Do you have the exact price of your entire chain before the DAC (less the server)?

 

I see you used a single AES cable from the NB to the QX-5. Did you try the dual AES connection? When we tested both options at my friend’s place (between a Paganini CD transport and Rossini DAC), the dual AES sounded more refined and relaxed, with a bigger and more holographic soundstage. I guess it has to do with further jitter reduction because, as per dCS, the dual AES connection splits the sample rate of the digital file played between the two AES outputs, so each AES cable transfers only one half of the total sample rate, reducing jitter in the process.

 

The QX-5 certainly can’t play at the Rossini level (and most certainly not even at the older Puccini/Paganini level) as we are talking about an off the shelf ESS Sabre chip in the QX-5 vs the Ring DAC with 40 different in house developed chips and a powerful FPGA to control them all. Plus the difference in quality of the power supplies and probably even the audio clocks (though this one is just a guess). And have in mind many people have chosen the QX-5 over the Chord Dave (I read this on other forums).

 

To level the playing field with the jitter reduction and clocks, ideally you should try the dual AES connection from the Network Bridge and also a 75ohm BNC spdif cable from the NB’s word clock output to the word clock input on the DAC (unfortunately the QX-5 does not have such an input). Supposedly the NB uses the same audio clock as the one found in the Rossini, so I am still buffled as to how they managed to keep the price of the NB so low (it’s streaming module is also more powerful than the one found in the Rossini DAC).

 

So looks like these “boutique” brands have juice in them after all :):):) 

 

 

Link to comment
44 minutes ago, AmusedToD said:

 

@austinpop, I guess you understand my point now in wanting to upgrade to a dCS DAC. Yes, they are bloody expensive, but they play in a different league altogether. Thank you for this short review! 

 

What??

 

I think you must have me confused with someone else. I didn't say anything to disparage your interest. In fact I should hope that this shows I am not wedded to a brand or approach. I keep an open mind.

 

44 minutes ago, AmusedToD said:

Intersting findings regarding the Network Bridge vs trifecta comparison. I believe your Trifecta is even more expensive than the NB as it includes a 10Mhz Reference Master clock. Pretty large number of boxes in the chain plus cables to connect them plus PSU’s to power them. Do you have the exact price of your entire chain before the DAC (less the server)?

 

As I've posted before, about $5k.

 

44 minutes ago, AmusedToD said:

I see you used a single AES cable from the NB to the QX-5. Did you try the dual AES connection? When we tested both options at my friend’s place (between a Paganini CD transport and Rossini DAC), the dual AES sounded more refined and relaxed, with a bigger and more holographic soundstage. I guess it has to do with further jitter reduction because, as per dCS, the dual AES connection splits the sample rate of the digital file played between the two AES outputs, so each AES cable transfers only one half of the total sample rate, reducing jitter in the process.

 

I thought dual AES only works with dCS DACs, not with the QX-5. 

 

44 minutes ago, AmusedToD said:

The QX-5 certainly can’t play at the Rossini level (and most certainly not even at the older Puccini/Paganini level) as we are talking about an off the shelf ESS Sabre chip in the QX-5 vs the Ring DAC with 40 different in house developed chips and a powerful FPGA to control them all. Plus the difference in quality of the power supplies and probably even the audio clocks (though this one is just a guess). And have in mind many people have chosen the QX-5 over the Chord Dave (I read this on other forums).

 

I don't speculate on things like this, nor do I disparage anyone's design choices. To define the QX-5 as an "an off the shelf ESS Sabre chip" is to do it a great disservice, and it sounds patronizing. Why would you expect it to compete with something 3x it's price?

 

44 minutes ago, AmusedToD said:

So looks like these “boutique” brands have juice in them after all :):):) 

 

Who is this addressed to? Did I ever say they didn't?

Link to comment
4 hours ago, austinpop said:

Trifecta Medley (just to remind)

  • Zyxel switch (sCLK-EX mod) > SOtM dCBL-Cat7 > SOtM sMS-200 (sCLK-EX mod) > USPCB > ISO-Regen > USPCB > SOtM tX-USBultra > Lush USB cable
  • Reference Clock: Cybershaft Premium OCXO OP-14 10MHz clock driving the tX-USBultra

Hi Rajiv, 

Apologies if this has been covered and I missed it (which given the detailed articulate posts you contribute I'm sure it has) but why is the Cybershaft only driving the tx-USBultra? 

 

Perhaps I got the wrong end of the stick but I thought the sCLK-ex model involved the ability to add an external clock to the drive the sCLK-ex, which in turn drives the items its connected to (such as SMS-200ultra, modded switch & tx-USBultra in your case). 

 

Did I misunderstand this as I thought synchronisation of several components to a great clock was the essence of the secret sauce in the spaghetti setup? 

 

Cheers, 

Alan 

 

PS great work comparing the DCS to the Trifecta! 

Synergistic Research Powercell UEF SE > Sonore OpticalModule (LPS-1.2 & DXP-1A5DSC) > EtherRegen (SR4T & DXP-1A5DSC) > (Sablon 2020 LAN) Innuos PhoenixNet > Muon Streaming System > Grimm Audio MU1 server > (Sablon AES) Mola Mola Tambaqui DAC > PS Audio M1200 monoblocks > Salk Sound Supercharged Songtowers

Link to comment
3 minutes ago, BigAlMc said:

Hi Rajiv, 

Apologies if this has been covered and I missed it (which given the detailed articulate posts you contribute I'm sure it has) but why is the Cybershaft only driving the tx-USBultra? 

 

Perhaps I got the wrong end of the stick but I thought the sCLK-ex model involved the ability to add an external clock to the drive the sCLK-ex, which in turn drives the items its connected to (such as SMS-200ultra, modded switch & tx-USBultra in your case). 

 

Did I misunderstand this as I thought synchronisation of several components to a great clock was the essence of the secret sauce in the spaghetti setup? 

 

Cheers, 

Alan 

 

PS great work comparing the DCS to the Trifecta! 

 

I suppose because the 10Mhz input is installed  on his txUSBUltra which houses the sCLK-ex clock board. But @austinpop should be able to explain. 

Link to comment
12 minutes ago, BigAlMc said:

Hi Rajiv, 

Apologies if this has been covered and I missed it (which given the detailed articulate posts you contribute I'm sure it has) but why is the Cybershaft only driving the tx-USBultra? 

 

Perhaps I got the wrong end of the stick but I thought the sCLK-ex model involved the ability to add an external clock to the drive the sCLK-ex, which in turn drives the items its connected to (such as SMS-200ultra, modded switch & tx-USBultra in your case). 

 

Did I misunderstand this as I thought synchronisation of several components to a great clock was the essence of the secret sauce in the spaghetti setup? 

 

Cheers, 

Alan 

 

PS great work comparing the DCS to the Trifecta! 

 

The Cybershaft has one output. It drives the reference clock input of the sCLK-EX board. In my setup, that lives in the tX-USBultra.

 

The 4 clock outputs of the sCLK-EX board inherit the ultra low phase-noise of the reference clock, so in effect, the reference clock's quality is propagated across the whole chain.

Link to comment
53 minutes ago, austinpop said:

 

What??

 

I think you must have me confused with someone else. I didn't say anything to disparage your interest. In fact I should hope that this shows I am not wedded to a brand or approach. I keep an open mind.

 

 

As I've posted before, about $5k.

 

 

I thought dual AES only works with dCS DACs, not with the QX-5. 

 

 

I don't speculate on things like this, nor do I disparage anyone's design choices. To define the QX-5 as an "an off the shelf ESS Sabre chip" is to do it a great disservice, and it sounds patronizing. Why would you expect it to compete with something 3x it's price?

 

 

Who is this addressed to? Did I ever say they didn't?

 

Thanks again for the time and effort you undertook in comparing the NB with your Trifecta. Those were my general observations, as question were raised (not by yourself) whether it’s proper to compare “boutique” equipment to the more “down to earth” equipment which is primarily discussed here.

 

Regarding the QX-5, it’s most probably a great sounding device (I never heard it personally with a speaker system, but did fiddle with it in Munich with a pair of headphones they had at the stand), but it does use an off the shelf chip and that’s a fact, albeit with some pretty fancy filters designed by Charlie Hansen.

 

It was on my short list too as I respect Ayre very much for what they do. If you remember, I asked you to compare the QX-5 to your Trifecta, which you did and I thank you for that once again. I accepted your findings and thus erased the QX-5 from my list and refocused on dCS only. 

 

My intention was to say that Ayre (probably) doesn’t sound as good as the Puccini/Paganini level of dCS DACs (let alone Rossini), and the price difference is not that huge, although the Puccini/Paganini are now discontinued. The QX-5 is around £8,000 in the UK, whereas the Puccini/Paganini were around £11,000 when they were available until late 2015. And the most logical reason for this is the DAC circuit design, at least that’s how I see it and that’s what dCS have been bragging about for more than 20 years. That’s what Chord brags about too when their DACs are mentioned.

Link to comment

@austinpop If I understood correctly your trifecta was connected to the QX-5 using USB of course and the NB using AES/EBU. What if the USB input of the QX-5 is not as good as the AES? Perhaps the results would be much more similar when could use for example a MC3+ USB between your trifecta and the DAC. Just a thought. I understand if you don’t have access to a MC3+ USB of course... If I would live closer to you I would lend you my modified Mutec. 

Link to comment
18 hours ago, JohnSwenson said:

IF the switch is one of FS105 and FS108, grounding the negative of the supply will get rid of BOTH the leakage from the SMPS AND the network leakage. NOTE, this ONLY happens for these two switch types. Grounding the supply to a different switch type does NOT block network leakage.

I remember reading earlier in this thread about a version 3 of these switches. How important is the version of these switches? Will all versions work the same? It appears the obvious difference is in the dc voltage. I see a 7.5 vdc and a 12 vdc version. Newegg doesn't specify the specific version number on the product page and to make it worse, they're showing pictures of both the 7.5 V and 12 V models in the same product page.

Link to comment
7 minutes ago, lasker98 said:

I remember reading earlier in this thread about a version 3 of these switches. How important is the version of these switches? Will all versions work the same? It appears the obvious difference is in the dc voltage. I see a 7.5 vdc and a 12 vdc version. Newegg doesn't specify the specific version number on the product page and to make it worse, they're showing pictures of both the 7.5 V and 12 V models in the same product page.

 

My question as well.  Amazon sells an FS105NA which looks like it's the v2 switch.  I like that it's 7.5v instead of 12v as it makes it much easier to power with an LPS-1 if necessary.  @JohnSwenson mentioned the 12v having a better jack but not sure on the details.  This v2 version also has a grounding plug.  See picture.  Not sure how that plays into it all and whether using it provides any added benefit.

 

51FuPnWByJL._SL1400_.jpg

Link to comment
31 minutes ago, afrancois said:

@austinpop If I understood correctly your trifecta was connected to the QX-5 using USB of course and the NB using AES/EBU. What if the USB input of the QX-5 is not as good as the AES? Perhaps the results would be much more similar when could use for example a MC3+ USB between your trifecta and the DAC. Just a thought. I understand if you don’t have access to a MC3+ USB of course... If I would live closer to you I would lend you my modified Mutec. 

 

This is why I was careful not to draw any lasting conclusions yet. I want to try it in my system first. 

 

Yes, I understand that there could be DAC performance differences between USB and AES inputs and who knows - the NB USB output, once enabled, may be suboptimal too. According to my dealer, dCS are confident it will be just as good as the other outputs.

 

i also envision a trifecta comparison on @limniscate's system too, which IS AES based, but his trifecta is not master clocked. And he has 2 sCLK-EX in his chain. The Cybershaft at our disposal only has one clock output. And, his clock inputs are SMB, not BNC, so we don't have the right cable. 

 

Like I said, I can't control everything! :)

 

Ultimately I do these comparisons for my own explorations and my own system. My current DAC prefers USB. So the only question I can go deep on is how well the NB does USB.

 

stay tuned!

 

 

Link to comment
1 hour ago, afrancois said:

could use for example a MC3+ USB between your trifecta and the DAC. Just a thought. I understand if you don’t have access to a MC3+ USB of course... If I would live closer to you I would lend you my modified Mutec. 

 

Questions about the MC-3+ USB for you:

  • The PSU is not linear?
  • When driven by a Ref 10 reference clock
    • This improves the 2 sample data clocks, I've read. But -
    • does it also improve the USB interface clock? If not, the USB clock remains a weak link in the chain.

Perhaps these are questions best posed to @julian.david.

 

OTOH - if anyone wants to lend us a Ref 10, that would be quite acceptable. :D

Link to comment

@austinpop, assuming the attention to clocking in all network components is about lower electrical noise (you have not concluded this) have you tried fibre right before the renderer ?  There is the problem of noise generated by the downstream FMC but SOtM can modify an MC100CM or MC200CM with a linear regulator, improved capacitors, and replace the clock to use an external  sCLK-EX.  This could be interesting until the day we have renderers with fibre input, if ever.

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...