DCS Vivaldi

[microstrip]

Finally received the Vivaldi DAC on Wednesday and set it up. Only interesting snafu in setup was that although everything was wired correctly and logically on the software side of things, trying to play MQA files caused insanity between the Upsampler and DAC. But I was able to get things to work correctly after powering off/on the Upsampler. My suspicion is that it's the "driver" in this arrangement and needed to "sync" up with the DAC. I surmise this as I had no issue when swapping the dCS Network Bridge with the new Upsampler and using with my existing Berkeley DAC.

Other odd bit is that shutting down the Upsampler and powering it on again, I lost all of settings I'd made to it. Would not have thought that would happen.

Anyway, I'm ensconced at the moment "breaking in" the new DAC (figure a few hundred hours will do. And trying to discern the various filter settings as they affect overall sonic results based on how I've set up the Upsampler for various bitrates. Currently enabled DSD passthrough and DSX for all PCM. The various filter settings are really subtle.

Expect to have the Clock by sometime in early March, given the dCS backlog on this device.

Perhaps you already have done it, but while you do not have an external clock it is possible to use the DAC clock output to feed the upsampler clock input, synchronizing both units.

 

[stevebythebay]

I’d already purchased 4 of the Transparent XL BNC-BNC 75 ohm cables before receiving the new DAC. So yes, I’ve got one connected from the DAC’s word clock out into the Upsampler wordclock 1 input. Also using dual AES between the Upsampler and DAC. Luckily I was able to get Synergistics Research to make another Galileo LE for me (it’s no longer in production). So now all I am awaiting is the clock, which I hope to have in house by the end of next month.

One last bit: I’m looking forward to purchasing the EtherREGEN from UpTone Audio, once it gets released, to see if it helps keep any residual noise out of the system.

 

[microstrip]

I’d already purchased 4 of the Transparent XL BNC-BNC 75 ohm cables before receiving the new DAC. So yes, I’ve got one connected from the DAC’s word clock out into the Upsampler wordclock 1 input. Also using dual AES between the Upsampler and DAC. Luckily I was able to get Synergistics Research to make another Galileo LE for me (it’s no longer in production). So now all I am awaiting is the clock, which I hope to have in house by the end of next month.

One last bit: I’m looking forward to purchasing the EtherREGEN from UpTone Audio, once it gets released, to see if it helps keep any residual noise out of the system.

I have also been dreaming about an Ethernet regenerator since long - but IMHO it should do a lot more than keeping residual noise from the system. I can't understand why someone does not pick a dual RJ45 such as the JCAT NET Card FEMTO, a good motherboard and a decent linear supply with adequate software and sells it with a good profit ... Unfortunately the industry focused on USB for a decade and only now seems to be thinking seriously about Ethernet.

Minimizing the noise from the whole server machine seems a waste of resources.

Great clock cables, BTW.

 

[stevebythebay]

Well you got to start somewhere I suppose. Just doing all you can to isolate the network from everything certainly helps. But the interfaces, cables, etc. all have the potential to carry noise of one kind or another.

By the way nice amps. I used a VTL 300 for about 25 years, which I got from David Manley back in ‘87. Modded by Luke & Bea over the years. Then I finally replaced with Spectral gear. Interesting how the best characteristics of both tube and solid state have converged over the years.

 

[LL21]

...By the way nice amps. I used a VTL 300 for about 25 years, which I got from David Manley back in ‘87. Modded by Luke & Bea over the years. Then I finally replaced with Spectral gear. Interesting how the best characteristics of both tube and solid state have converged over the years.

Agree! I am a big tube fan...did have tube amps at one point and was stunned and enjoyed them...found myself investigating super-high powered tubes for quite some time. Ended up with Class A SS...that was nearly 10 years ago when I got my first Gryphon. And I have now moved up twice within the family and not looking back...though Aries Cerat has certainly impressed me with their very loyal following of extremely discerning audiophiles.

 

[Elberoth]

I have also been dreaming about an Ethernet regenerator since long - but IMHO it should do a lot more than keeping residual noise from the system. I can't understand why someone does not pick a dual RJ45 such as the JCAT NET Card FEMTO, a good motherboard and a decent linear supply with adequate software and sells it with a good profit ... Unfortunately the industry focused on USB for a decade and only now seems to be thinking seriously about Ethernet.

Minimizing the noise from the whole server machine seems a waste of resources.

Great clock cables, BTW.

There already is the SOtM sNH-10G Ethernet Switch, which is said to do just that. I got one myself, but didn't have the time to test it just yet.

The UpTone Audio's EtherRegen that is similar in concept and is in development.

 

[Elberoth]

I have just posted in another thread my impressions about the $3990 Mutec Ref-10 external 10MHz clcok.

https://www.whatsbestforum.com/thre...hootout-op20-vs-op17.26610/page-4#post-556935

My friend has borrowed the Ref-10 clock to try on his Vivaldi stack, where it replaced the CyberShaft's older OCXO Premium (which has a phase noise of -110dBc/Hz) and his initial reaction is very positive. The difference between the Mutec Ref-10 and CyberShaft OCXO Premium is much greater than the CyberShaft OCXO Premium and no 10MHz external clock.

I wonder, how much does the adding the Mutec clock is closing the gap to the MSB Select II DAC (which some people report is there).

 

[stevebythebay]

Finally received the Vivaldi Clock yesterday. Surprised at just how big an improvement it makes. I suspect it may get somewhat better as the Transparent XL cables break in. In any case, the system sounds more holographic with greater depth and specificity of the soundstage. Seems the instruments seem far more realistic and cohesive and oddly I can raise the volume without feeling blown back in my chair. The lower bass seems more textured, and musical elements that I’d previously let escape my attention are now clearly expressed. And this is especially true at the very end of passages. All in all, a great addition to my sound system.

 

[Legolas]

Finally received the Vivaldi Clock yesterday. Surprised at just how big an improvement it makes. I suspect it may get somewhat better as the Transparent XL cables break in. In any case, the system sounds more holographic with greater depth and specificity of the soundstage. Seems the instruments seem far more realistic and cohesive and oddly I can raise the volume without feeling blown back in my chair. The lower bass seems more textured, and musical elements that I’d previously let escape my attention are now clearly expressed. And this is especially true at the very end of passages. All in all, a great addition to my sound system.

Steve
Can I ask, what is the Roon Nucleus? Is it the new Roon+ Nucleus music server? I am after thoughts on how it performs, as seems promising to me...

 

[stevebythebay]

Steve
Can I ask, what is the Roon Nucleus? Is it the new Roon+ Nucleus music server? I am after thoughts on how it performs, as seems promising to me...

The Roon Nucleus is a hardware product designed by the same Roon Labs which makes the software that runs on Intel and Apple hardware. It was conceived as a purpose built hardware and software optimized to run the Roon Core. There are two versions. One for more limited (size) libraries and where a user is less likely to want to use DSP and other Roon functions. The other is a more powerful version. See more details at roonlabs.com.

 

[CKKeung]

Finally received the Vivaldi Clock yesterday. Surprised at just how big an improvement it makes. I suspect it may get somewhat better as the Transparent XL cables break in. In any case, the system sounds more holographic with greater depth and specificity of the soundstage. Seems the instruments seem far more realistic and cohesive and oddly I can raise the volume without feeling blown back in my chair. The lower bass seems more textured, and musical elements that I’d previously let escape my attention are now clearly expressed. And this is especially true at the very end of passages. All in all, a great addition to my sound system.

Congratulations!

But please go for buying a Mutec Ref10 or Cybershaft 10M clock to supply the Vivaldi Clock.
You will be very surprised!

 

[stevebythebay]

Rather get an atomic clock...

 

[Elberoth]

Rather get an atomic clock...

The so called 'atomic clocks' (with Rubidium oscillators) are no match performance wise for this Mutec or CyberShaft models.

 

[CKKeung]

Rather get an atomic clock...

You mean a rubidium clock such as Antelope Audio?

Sorry, they are not as good as the ocxo-based Mutec or Cybershaft if used for audio.

 

[CKKeung]

The so called 'atomic clocks' (with Rubidium oscillators) are no match performance wise for this Mutec or CyberShaft models.

Yes I did direct comparison between Antelope 10M vs Cybershaft OP13.
The cheaper OP13 was slightly better.

Then I compared Antelope 10MX vs Cybershaft OP20.
The latter KO the Antelope 10MX in seconds!
Huge diff!

 

[stevebythebay]

You mean a rubidium clock such as Antelope Audio?

Sorry, they are not as good as the ocxo-based Mutec or Cybershaft if used for audio.

Guess I’ll go with a GPS receiver.

 

[CKKeung]

A GPS clock, or GPS disciplined oscillator(GPSDO), is a combination of a GPS receiver and a stable oscillator whose output is controlled to agree with the signals broadcast by GPS or other GNSS satellites.

It's essential to have a good oscillator with low phase noise to suit audio application.
Long term clock accuracy is not our purpose.

I have experience on 2 GPSD0 for auidio uses.
The first one is a relatively low price one.
The audio performance is mediocre and far inferior to Antelope and Mutec, not to mention the Cybershaft.
I guess it's because the crystal oscillator inside was a cheap one.


In the contrary, the CH Precision T1 was excellent!
I am eager to compare it to Cybershaft OP20.

Therefore not all GPSDO are the same. It depends on implementation/execution.

 

[stevebythebay]

Thanks for your insight. I’ll get a recommendation from dCS as they provide an input for such a device.

 

[Elberoth]

Thanks for your insight. I’ll get a recommendation from dCS as they provide an input for such a device.

Using GPSDO for audio makes exactly ZERO sense. None, nada.

Their internal XO (or Rubidium oscillator) is continously 'pulled' to track the GPS signal by a tracking loop. The disciplining mechanism works in a similar way to a phase-locked loop (PLL) and is a source of jitter, decreasing GPSDO 's short term stability.

GPSDOs are built for long term accuracy, not short term stability, and there is a range of applications where they work great. Unfortunately, D/A conversion in audio applications, is not one of them.

You need an XO with lowest possible short term stability, i.e. phase noise figures. This is why fixed frequency XO work best and have lowest phase noise figures.

Here you can compare the phase noise plots for several different oscillator modules:

Rubidium oscllator module SRS PRS-10 used in Esoteric Rubidium clocks (green trace)
Rubidium oscllator module FE-5680A used in Jay's Audio CD transport (blue trace)
OCXO used in Mutec Ref-10 clock (yellow trace)

As you can see, the Mutec's OCXO clock is far superior to any rubidium oscilators. Even the excellent (and expensive at $1400 OEM cost) SRS module only gets -105dBc/Hz @ 1Hz. Mutec is rated -116dBc/Hz and CyberShaft OP20 is -120dBc/Hz.

The Antelope 10M clock mentioned here, which is based on the Spectratime LCR-900 module, is worse than ALL oscillators plotted above - its plot would be ~20dB above the blue plot of FE-5680A! It is literally laughably bad.

 

[CKKeung]

Using GPSDO for audio makes exactly ZERO sense. None, nada.

Their internal XO (or Rubidium oscillator) is continously 'pulled' to track the GPS signal by a tracking loop. The disciplining mechanism works in a similar way to a phase-locked loop (PLL) and is a source of jitter, decreasing GPSDO 's short term stability.

GPSDOs are built for long term accuracy, not short term stability, and there is a range of applications where they work great. Unfortunately, D/A conversion in audio applications, is not one of them.

You need an XO with lowest possible short term stability, i.e. phase noise figures. This is why fixed frequency XO work best and have lowest phase noise figures.

Here you can compare the phase noise plots for several different oscillator modules:

Rubidium oscllator module SRS PRS-10 used in Esoteric Rubidium clocks (green trace)
Rubidium oscllator module FE-5680A used in Jay's Audio CD transport (blue trace)
OCXO used in Mutec Ref-10 clock (yellow trace)

The Antelope 10M clock mentioned here, which is based on the Spectratime LCR-900 module, is even worse than those plotted - its plot would be ~20dB above the blue plot of FE-5680A!

Elberoth : you are an expert in clocks for audio use!