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

[austinpop]

9 hours ago, romaz said:

Also, it remains to be seen at what point does adding more clocks no longer make a significant difference and will replacing all the bad clocks in a motherboard require as many reparative downstream devices.

 

Yes, this is a huge are of interest now, and we eagerly await your findings.

 

I had arbitrarily assumed that - going upstream from the DAC - each successive clock would have a lower effect, to the point of diminishing returns. Instead, my findings with the complementary nature of the effect of the tX-USBultra and the modded switch have me really wondering what the effect of better clocks further upstream will be.

 

Can't wait to find out! 

 

9 hours ago, romaz said:

 

For those looking for high value, as Rajiv has also found, a cheap switch thrown in can make a huge difference.  Perhaps, the MVP of the group given how its low cost is the switch.  I'm thinking that an sMS-200 Ultra with 2 switches could be a better value than an sMS-200 Ultra, tX-USB HD and switch.

 

Intriguing thought. I'd be surprised if SOtM aren't already working on a switch in their product portfolio.

 

9 hours ago, romaz said:

 

Just a couple of comments...

 

First, I'm not sure anyone yet knows just how good the clock implementation in the ISO Regen is compared against SOtM's new clock and so no assumptions should be made.  

...

 

I guess we'll see soon enough but I love having all these great options.

 

I could not agree more!

[austinpop]

1 hour ago, ChrisG said:

It's pretty straight forward - PCM+ for PCM, DSD+ for DSD...oh, and make sure you have the HF on

 

OK, OK, now all you DAVE-heads™ - it's a technical term - have me really intrigued. 

 

I'm now motivated enough to schlep my Cavalli LAu over to Eric's and test out the amazing headphone amp on the DAVE. I'm excited to find out how it sounds, although sadly I'll have to revert to the stock single-ended cables on my HD800, as - best I can tell - DAVE-o has no balanced headphone out.

 

Points about DSD+, PCM+, and crossfeed duly noted.  

[mozes]

9 hours ago, romaz said:

The question remains whether straight USB can sound just as good.

in a pure USB path, I am also intrigued by the outcome of having two tX-USBultra in series although not a very good value!

[Superdad]

2 hours ago, rickca said:

Now that is really interesting insight!  Maybe Uptone can design something even more effective than a switch with an upgraded clock or passive network isolators like Emo. 

 

How about it @Superdad and @JohnSwenson? 

 

Yes Rick, up-thread I already outlined the recipe for seriously great "audiophile" Ethernet switch with high signal integrity and isolation.  Oh, here it is:

 No time soon though.  Way too much ahead of that.  More broad appeal products.  I am afraid that network "tweakers" are a rather small group compared to other areas.  Not to say we won't do it.  Just not sure when.

[tboooe]

36 minutes ago, Superdad said:

No time soon though.  Way too much ahead of that.  More broad appeal products.  I am afraid that network "tweakers" are a rather small group compared to other areas.  Not to say we won't do it.  Just not sure when.

Alex, hope you are enjoying your vacay!

 

Hey, can I just open a tab with you guys?  That way you can just send me whatever new products you make and we dont have to deal with silly things like invoices, placing orders, etc? 

[Solstice380]

1 minute ago, tboooe said:

Alex, hope you are enjoying your vacay!

 

Hey, I can I just open a tab with you guys?  That way you can just send me whatever new products you make and we dont have to deal with silly things like invoices, placing orders, etc? 

UpTone Bar and Grill (and audio products,too) ?

[rickca]

41 minutes ago, Superdad said:

Yes Rick, up-thread I already outlined the recipe for seriously great "audiophile" Ethernet switch with high signal integrity and isolation.

Thanks for the link, Alex.  I'd forgotten that you had already provided quite a lot of detail.

[Cooler]

Anyone tried to compare sotm/mRendu with pinkfaun i2s pci-e?

Instead of connect

PC -> ethernet -> sotm/mRendu -> usb DAC to

PC -> pindfaun i2s -> i2s DAC. 

 

Ive heard usually usb worse then i2s in modern DACs and sotm/mRendu cant fix that, they only can send best possible signal to DAC via usb.

[gsquared]

12 hours ago, romaz said:

 

I'm not sure if you've figured it out yet or not.  It seems to be easier to create a bridge on a Mac than on some PCs.  In your case, trying turning off your WiFi and establishing a connection to your router via your wired Thunderbolt ethernet connection.  Once this has been established, then try bridging this connection with your other LAN connection that is connected to your mR.

Hi Roy, great tip. I hadn't thought of turning off my WiFi connection, but that's probably why I could not see both the ethernet and thunderbolt at the same time.

 

I'll give that a try this evening to see if that does the trick.

 

Thanks!

[greenleo]

After reading so many recent great reports, I have the feeling that the clock is more important than the device, whether it is a switch, a 200Ultra or an txUltra.  

 

It would be very interesting to compare

Switch-200Ultra-txUltra vs 200Ultra-txUltra-txUltra.  If they are of similar SQ, then this hypothesis may be verified.  Then it would've very interesting to compare the SQ of 

 

4x txUltra vs  2x Switch with 200Ultra vs switch with 200Ultra with txUltra vs 200Ultra with 2x txUltra

4 clocks(USB end points) vs 4 clocks (NAA configured in different ways)

 

I suspect the 4 clocks(USB end points) wins in SQ.

 

It would take a group of members to gether the gears together for such a shootout and the point of diminishing return may be found unless more than 4 clocks are needed!

[Johnseye]

On 5/8/2017 at 8:03 PM, austinpop said:

We both agreed the effect of the SOtM stack was more dramatic than the the DAC difference between the Codex and the DAVE, which is notable in itself!

 

This is a huge take away.  As we make all these changes to our chain, especially as they relate to power and clocks, the question I ask myself is whether the money would be better spent elsewhere.  That you think the SOtM clock changes are more significant than a $10k difference in the cost of a DAC encourages me to go ahead with the same modifications.  Thanks Rajiv!

[austinpop]

6 hours ago, greenleo said:

After reading so many recent great reports, I have the feeling that the clock is more important than the device, whether it is a switch, a 200Ultra or an txUltra.  

 

It would be very interesting to compare

Switch-200Ultra-txUltra vs 200Ultra-txUltra-txUltra.  If they are of similar SQ, then this hypothesis may be verified.  Then it would've very interesting to compare the SQ of 

 

4x txUltra vs  2x Switch with 200Ultra vs switch with 200Ultra with txUltra vs 200Ultra with 2x txUltra

4 clocks(USB end points) vs 4 clocks (NAA configured in different ways)

 

I suspect the 4 clocks(USB end points) wins in SQ.

 

It would take a group of members to gether the gears together for such a shootout and the point of diminishing return may be found unless more than 4 clocks are needed!

 

That's a great thought, but you're right, the logistics are daunting. I'd be up for participating with my stack.

[ElviaCaprice]

How about using an older Regen in line with a replaced EX clock?  Might be a great way to put an older cheaper component to effective use.  Could share an LPS-1 with two Regens.

 

Personally I would rather use a Regen over the TX-USB Ultra, much easier input, output for usb adapters.

[austinpop]

5 hours ago, Johnseye said:

 

This is a huge take away.  As we make all these changes to our chain, especially as they relate to power and clocks, the question I ask myself is whether the money would be better spent elsewhere.  That you think the SOtM clock changes are more significant than a $10k difference in the cost of a DAC encourages me to go ahead with the same modifications.  Thanks Rajiv!

 

To be honest, this surprised me too. I expected a lot more from such an extravagant DAC upgrade.

 

Of course, please apply the usual caveats that YMMV - I can only report what I heard in my system.

[Confused]

@austinpop - My apologies if I have missed the details somewhere (this is a long thread now!), but what exactly is the mod to the Zyxel GS108b V3 switch that you mention?  Are there any details of this posted elsewhere?

 

I have some interest in your good results.  Recently I have been trying the Aqvox switch, and to be honest I have been a little disappointed with it.  Maybe it offered some tiny improvements, but I wasn't 100% convinced that it was not adding some tiny negatives also.  So disappointing.

 

I notice that JCat are now selling a modified switch with TCXO clock, is this a similar concept to what you have been trying?

 

http://jplay.eu/jcat/

[skyline]

6 hours ago, austinpop said:

 

To be honest, this surprised me too. I expected a lot more from such an extravagant DAC upgrade.

 

Of course, please apply the usual caveats that YMMV - I can only report what I heard in my system.

 

It's all  about the systematic approach and not 1 holy grail component.

 

Took me a while in the hobby to figure this out.

 

I use a Waversa hub ethernet filter with my Lumin, I'm willing to bet it yields more of a performance increase than going up the Lumin line.

 

I noticed you used Pangea power cords. You will get a big leap in performance by going with a reference cord like the Audience AU24se.

[rickca]

Is the SOtM sCLK-EX based on an OCXO?  I've read the material on the SOtM website and it just says it's a custom designed ultra low phase noise oscillation circuit.

[ElviaCaprice]

4 minutes ago, rickca said:

Is the SOtM sCLK-EX based on an OCXO?  I've read the material on the SOtM website and it just says it's a custom designed ultra low phase noise oscillation circuit.

No it's not.  If it we're you would see the ovens, which are pretty big. 

And to answer your next question.  We don't know why the sclk EX clocking board is so effective.  But it seems it obviously can be heard.

[rickca]

1 minute ago, ElviaCaprice said:

No it's not.  If it we're you would see the ovens, which are pretty big. 

Thanks.  I can barely recognize the oven in my kitchen, never mind on a circuit board.

[ElviaCaprice]

PPA uses one on his version 3&4 USB cards.

[amir57bs]

On 1/10/2017 at 11:05 AM, romaz said:

This is a very good question. I have a lot of experience with various single box purpose-built music servers with USB streams including a TotalDac-d1 server, Aurender N100 with Ken Lau PSU, Aurender N10, Aurender W20, Lumin U1, CAD CAT, Auralic Aries with femto clock and LPS, various custom-built CAPS servers and more. Here is the thread I started on Head-Fi:

 

REVIEW: Comparison of 5 High End Digital Music Servers - Aurender N10, CAD CAT server, TotalDac d1-Server, Auralic Aries, Audiophile Vortex Box

 

None of those single box units ever sounded anywhere close to what I am hearing now. That is how special both the microRendu and sMS-200 are, imo. This direct connection has much to do with it also but so does my Paul Hynes SR7 which I just got recently. I am finding these things to be the key pieces so far.

 

do you mean using microRendu is far better than Single Box (with Sotm PCIe to USB CARD) ?

[beautiful music]

I plugged my Dave to sMS-200 via USB but it didn't send any information to Dave despite i checked the Dave in the SOtM web configuration and it was there.

While i switched to my Mojo the sMS-200 send a data to it and i can run my music.

 

Any thoughts and advice.

[mrmb]

On 3/29/2017 at 11:10 PM, romaz said:

http://www.msbtechnology.com/faq/how-to-wire-your-house-for-good-power/

 

To be oblivious to what good power cables can provide with respect to line resistance is to be oblivious to a whole other world of dynamics hidden within your electronic gear trapped by poor cables.

romaz:  Thanks for this link and article.  We're preparing to build a new house and I'm looking for wiring suggestions, recommendations etc.  The ones you posted from MSB are very interesting and helpful....Thanks!

[austinpop]

7 hours ago, Confused said:

@austinpop - My apologies if I have missed the details somewhere (this is a long thread now!), but what exactly is the mod to the Zyxel GS108b V3 switch that you mention?  Are there any details of this posted elsewhere?

 

No there are no details anywhere. Unlike Roy and others here, I do not have experiences with, or opinions of, specific brands and models of electronics parts! I should also say that I am shamelessly riding @romaz's coattails on this "cascaded good clocks" approach.

 

So I have no explanations, just observations.

 

Specific to the Zyxel switch, I followed Ray's advice, and asked SOtM to replace the regulators with linear ones. I gather SOtM swapped in 2 linear regulators and replaced 1 or more capacitors. I don't have specific part numbers - sorry.

 

7 hours ago, Confused said:

I have some interest in your good results.  Recently I have been trying the Aqvox switch, and to be honest I have been a little disappointed with it.  Maybe it offered some tiny improvements, but I wasn't 100% convinced that it was not adding some tiny negatives also.  So disappointing.

 

I highly recommend you go back up thread and look for the posts where Roy talked about his premise for trying this mod. The emerging theory - and it is only a theory - is that there is something about the following combination immediately upstream of your DAC that yields these SQ gains:

1. Find a clock of outstanding quality. I believe he talks about how he was alerted to the sCLK-EX quality by another vendor
2. Create a cascade of reclockings immediately upstream of the DAC with this clock
3. Figure out how many reclockings upstream before this ceases to matter.

What he (and now I) observed is that so far, 4 reclcokings (from the 4 taps of the sCLK-EX resident in the tX-USBultra) have each had a noticeable and cumulative effect. These reclockings are:

• one in the switch, as Ethernet frames flow from one port to the other
• one on the Ethernet port of the sMS
• one on the USB port of the sMS
• one on the USB ports of the tX

Relating this back to your question about the Aqvox, I think the issue is not whether a switch with a high-quality clock can have an impact, it is more about what is the cascade of clocks from end-to-end all the way to your DAC.

 

The evidence seems strong to me that the real key is getting a sequence of excellent clocks in a chain immediately upstream of the DAC. Now we are all awaiting Roy's results with his mobo mods, which goes back to point 3 above. What happens when he goes past n=4, where n is the number of "best" reclockings?

 

Stay tuned!

 

 

[micesol]

On May 8, 2017 at 11:58 PM, romaz said:

Wow, I took a few days off to tend to other things and came back to see that lots has happened!  Lots of great reporting.  Thank you, Rajiv and everyone else!  I am enjoying reading everyone's experiences.

 

Yes, it seems consensus is building that clocks do matter and can make a huge difference regardless of the DAC.  The impact seems much greater within the "direct connection" pathway because when I break up my bridged LAN configuration and revert back to the traditional method, the impact of my reclocking switch is significantly less and so the impact of these devices, I believe, is conditional.  Bridging still matters and matters even more.  The question remains whether straight USB can sound just as good.  Also, it remains to be seen at what point does adding more clocks no longer make a significant difference and will replacing all the bad clocks in a motherboard require as many reparative downstream devices.

 

For those looking for high value, as Rajiv has also found, a cheap switch thrown in can make a huge difference.  Perhaps, the MVP of the group given how its low cost is the switch.  I'm thinking that an sMS-200 Ultra with 2 switches could be a better value than an sMS-200 Ultra, tX-USB HD and switch.

 

Just a couple of comments...

 

First, I'm not sure anyone yet knows just how good the clock implementation in the ISO Regen is compared against SOtM's new clock and so no assumptions should be made.  Personally, I am expecting the ISO Regen to be fantastic and am open to the possibility that it could be better or at least it could do some things better than the tX-USB Ultra and that possibly the two devices could even be complementary in a pure USB setup, especially if your DAC's USB input has no GI at all.  If it turns out the Cyrstek clock is not at the same level as the SOtM clock, I could see the combo of music server > ISO Regen > tX-USB Ultra > DAC being something worth considering, especially given the ISO Regen's excellent price point.  I guess we'll see soon enough but I love having all these great options.

 

Second, regarding the DAC, I still find this to make the greater difference, at least in my setup although a DAC can only be as good as the source that feeds it and will also be limited by the amp that follows it.  It has been said that the Chord DAVE is the most transparent headphone DAC in the world when headphones are directly connected to its headphone output and I believe this to be true.  With speaker setups, much of DAVE's bandwidth, speed, noise floor, dynamic range and time resolution is lost or buried by the speaker amp that follows it because no speaker amp in existence can match its characteristics and so DAVE will sometimes fail to distinguish itself as readily in speaker setups although there are those of us who use DAVE to directly drive speakers (without the use of a speaker amp) and this is something I've been doing now for nearly a year.  

 

Because of DAVE's ultra low output impedance of 0.055 ohms and because it can output 2 watts at 8 ohms via its RCA outputs, DAVE can directly drive high efficiency speakers (>95dB sensitivity) without having to use a separate speaker amplifier to astonishing levels of resolution and transparency, so good that I pretty much listen only to speakers these days.  Later this year, Rob Watts will be releasing the first of his digital amps (where DAVE will connect to these amps via digital and not analog connectors) and these amps will have the same transparency as what I am listening to now in case there are those who are intrigued by what this level of resolution and transparency sounds like and don't use high efficiency speakers.  The first amp will be 20 watts per channel (or 70 watt monoblocks) and will scale to as high as >200 watts in future releases.  The point of aspiring to such transparency in your downstream components is that the quality of the upstream source begins to really matter all the more and it now appears that there is so much still to be gained by paying attention to your source.

 

Roy, I am curious as to how to hook up the Chord Dave through its RCA outputs to a pair of speakers. My Audio Note AN-E silver signature speakers are 94.5 efficient, so they may be candidates for direct hook up to the Chord Dave. I think this may have been described in this thread but can't remember which post. I know Louis offers a special way to hook up to his monitors, but how will I do this with Audio Note speakers? Thanks!