[adrianm]

I was really happy with my Bartók system in its then current state – Innuos Statement – Vivaldi Clock – Bartók non-Apex – headamp GSX mk2 - Focal Utopia – Ansuz Titanium Rack – Ansuz dtc power.

When I upgraded to Vivaldi the reaction was – and this is supposed it? Plus I was unhappy listening. It took a lot of time tweaking and other upgrades to get to a new satisfactory state. Mostly cables, power conditioning and upstream tweaks, but also a switch to Riviera AIC-10 and DCA Stealth.

Now I can confidently walk into any 2-channel dealer or audio show room and say, thanks but no thanks. Plus enjoy myself greatly.

Long story short, as one moves up the ladder with dCS gear, system building becomes ever more important. And it costs quite a bit of money beside the gear itself.

I have to say, I recently tried the Nucleus and Innuos Zen, expecting the servers not to make a difference. Because why would they? To my surprise, Nucleus sounded clearly better than my gaming pc as a server/client. the Innuos even better on some tracks, but worse on others.
Hot take : The only logical conclusion for me is that Roon is not bit perfect, unless ran on Nucleus. I assume this is the trick up their sleeve, because it sounded identical to my gaming pc, just less "grainy". I think Innuos , as well is also CLEARLY doing EQ. I found it louder than Nucleus to begin with, and accentuating different parts of the FR.
While i'm not disputing it might be more pleasant in some cases, it's exactly the kind of tomfoolery i'm not spending my money on, and am bothered by the intent itself. There's nothing "better", because there's no improving a standard that is functioning properly. There's just mixing and matching distortion. As with a lot of cables , power stuff, etc. This is why I went with dCS in the first place, was sick of improving Chord's gear for more money than the gear itself.
That said. i've kept my Isotek Aquarius and Genesis one regenerator for the clock, as it makes a considerable difference in my current place.

 

[adrianm]

Was the benefit worth it over the Lina or Rossini clocks?

I can tell you that the Rossini clock makes bigger difference to the Rossini than the Lina clock did to the Lina FWIW.

 

[572217]

Yes I love mine as well - it's superb.

On Friday I will listen to the Tambaqui and also a Dave (out of a Grimm MU1) just to see how they sound but it's just out of curiosity really. It's so the fellow that works at the store can hear some of my tubes as it saves him buying blind.

I was interested in the Tambaqui as I find the Bartok a bit flat sometimes with 80s music but that's more the recording than the Bartok itself. It's just showing you what's in the recording. Feed it good material and it's sublime. I just want to see how the Tambaqui presents the music as a stand alone DAC and how this compares to the Bartok.

I really hope the Grimm isn't as good as it has been described as I don't have £12k lying around. I suppose I could sell a kidney?

Regarding the price - I agree it's ridiculous. The 2.0 Firmware was such an improvement (containing code designed for Rossini and Vivaldi) that's probably worth another £1k and general price increases maybe another £1k but to go from about £10k to £17k is absolutely ridiculous.

You could argue that the 2.0 upgrade should have been an expected improvement given the nature of FGPAs as they can be re-programmed. Just a free Firmware update - not also resulting in a price increase.

I’m interested to read your findings.

Heard a Grimm MU1 + Chord TT2 with MScaler in Amsterdam last year and knew in maybe five minutes the combo is not for me. What was doing what, hard to tell. I only know the Hugo 1 I bought in 2018 went back to the dealer within three days. Chord appears to not be my cup of tea. On the technical side...

Depending on what you use, the Innuos gear is really nice. If you use Roon a lot, the Nucleus might be better, as Innuos opts for smaller processors. What I get out of the Innuos Statement via UPnP is noticeably better than streaming, and I have tweaked the upstream stuff responsible for streaming quite a bit.

 

[572217]

So you used a 24.500€ Vivaldi clock with the non Apex Bartok?

Was the benefit worth it over the Lina or Rossini clocks?

It’s a less common configuration, but yes. With great results. I never auditioned Rossini, so I can not say whether Rossini and clock would have been much better. One dCS community member heard a comparison with Bartók + Vivaldi clock vs. Rossini and said the Rossini only pulled ahead with the Rossini clock, not alone.

I purchased both DAC and clock at the same time and have not tried Rossini or Lina clock (which didn’t exist at the time). I came from Naim ND555 and was looking for an upgrade somewhat above that price point. External clocking makes a big difference in music presentation on dCS gear. Hence the choice of Vivaldi clock. I wanted to enter at the lower end of the DACs, because I could not know how much I will like the experience over time. How different the clocks sound? Unfortunately no idea. I can only tell you that plugging in the Black Cat Trøn clock cables for Vivaldi dropped my jaw, the power cord makes a difference and resonance control. Another dCS owner subsequently bought the Mini Trøn for his Rossini and his jaw dropped as well. So I would assume, that clock precision and surrounding components inside the box are quite noticeable too.

 

[ThanatosVI]

It’s a less common configuration, but yes. With great results. I never auditioned Rossini, so I can not say whether Rossini and clock would have been much better. One dCS community member heard a comparison with Bartók + Vivaldi clock vs. Rossini and said the Rossini only pulled ahead with the Rossini clock, not alone.

I purchased both DAC and clock at the same time and have not tried Rossini or Lina clock (which didn’t exist at the time). I came from Naim ND555 and was looking for an upgrade somewhat above that price point. External clocking makes a big difference in music presentation on dCS gear. Hence the choice of Vivaldi clock. I wanted to enter at the lower end of the DACs, because I could not know how much I will like the experience over time. How different the clocks sound? Unfortunately no idea. I can only tell you that plugging in the Black Cat Trøn clock cables for Vivaldi dropped my jaw, the power cord makes a difference and resonance control. Another dCS owner subsequently bought the Mini Trøn for his Rossini and his jaw dropped as well. So I would assume, that clock precision and surrounding components inside the box are quite noticeable too.

It's just such an unusual combination.
With current pricing the Apex Bartok + Vivaldi clock costs more than the Vivaldi Apex without clock.

 

[572217]

I have to say, I recently tried the Nucleus and Innuos Zen, expecting the servers not to make a difference. Because why would they? To my surprise, Nucleus sounded clearly better than my gaming pc as a server/client. the Innuos even better on some tracks, but worse on others.
Hot take : The only logical conclusion for me is that Roon is not bit perfect, unless ran on Nucleus. I assume this is the trick up their sleeve, because it sounded identical to my gaming pc, just less "grainy". I think Innuos , as well is also CLEARLY doing EQ. I found it louder than Nucleus to begin with, and accentuating different parts of the FR.
While i'm not disputing it might be more pleasant in some cases, it's exactly the kind of tomfoolery i'm not spending my money on, and am bothered by the intent itself. There's nothing "better", because there's no improving a standard that is functioning properly. There's just mixing and matching distortion. As with a lot of cables , power stuff, etc. This is why I went with dCS in the first place, was sick of improving Chord's gear for more money than the gear itself.
That said. i've kept my Isotek Aquarius and Genesis one regenerator for the clock, as it makes a considerable difference in my current place.

Digital audio is far more complex than vinyl to set up. Unfortunately.

One of the dCS community people works at a large switch manufacturing company. He checked, and the Roon stream is bit perfect unpacked. I still chose to not use it, despite the convenience as I prefer Mosaic’s sound for streaming and the Innuos for local music. It’s as if a layer of tech goes away. I am in a minority with this assessment, but I know what I hear and like and others are right with their seemingly incompatible perception as well. We just perceive differently.

My hunch is the unpacking process of RAAT inside the streamer is audible for some and not for others. Lucky if one hears no difference using RAAT, much more convenience with Roon.

I have no proof but would doubt Innuos alters the bitstream without telling the user. Would be an unnecessary business risk to take.

Some people use optical to solve for galvanic issues of the network. I did not, because I don’t want a nasty little piece of electronic (optical converter) directly attached to my resolving DAC. Audio over the network is still a bit of a mystery. Computer science tells me the layers of abstraction in the network protocol plus buffering in the streamer should eliminate all issues. Experience tells me, that power supplies and power conditioning make a tremendous difference for network and digital gear.

The sound differences you hear with Innuos are probably down to two things. The way they eliminate interference inside the box through component selction and a good power supply. Pretty confident, those are two main causes. Adding a great power conditioner to server + switch in the basement was quite the eye opener.

 

[572217]

It's just such an unusual combination.
With current pricing the Apex Bartok + Vivaldi clock costs more than the Vivaldi Apex without clock.

I bought my stuff before the price increases happened and saw it as a hedged bet – intuiting that Bartók alone would have a more difficult time leapfrogging the ND555. And in case I really liked the dCS experience, I knew I could keep the clock in any scenario.

Vivaldi is a system and unlocks its potential with Upsampler, Clock and DAC together. I wouldn’t recommend buying just the DAC. The fourth box, the disc player is only necessary if you want to keep discs.

 

[adrianm]

We just perceive differently.

Mostly based on our biases. For example:

Some people use optical to solve for galvanic issues of the network. I did not, because I don’t want a nasty little piece of electronic (optical converter) directly attached to my resolving DAC. Audio over the network is still a bit of a mystery. Computer science tells me the layers of abstraction in the network protocol plus buffering in the streamer should eliminate all issues. Experience tells me, that power supplies and power conditioning make a tremendous difference for network and digital gear.

There's a very simple explanation here, and it is based on reality, not audiophile marketing. Electrical noise and jitter makes absolutely no difference to SQ, until it reaches the analogue portion. How come Netflix works perfectly without tons of snake oil? Not to mention far more complex things than audio. Ironically it's exactly the thing you are discarding that makes all of the difference. The Innuos sounds better than Nucleus and other solutions when used as a server, because you run Ethernet via its filters and it prevents an albeit small amount of electrical noise to reach the dac, which does have an impact on SQ.
I did this experiment while I had the Innuos here. Ethernet from Innuos to Rossini sounded better than Ethernet from my router, even with the Innuos used as a server in both cases. That's where the improvement is. However, add an optical isolator powered by a very low noise LPSU (all for less than 300$ btw), and it's a dramatic improvement over the Innuos, and completely eliminates any advantage it has over Nucleus (weirdly, not over the gaming pc used as a roon endpoint, which is why i still suspect Roon does something funny with the stream, or there is an inherent disadvantage to hosting Roon on a windows pc compared to their optimized OS), leaving the Innuos with a difference in sound which is clearly EQ. I suspect this is one of the reasons why it's not actually Roon certified, even after all this time .

We are in agreement that power supplies and conditioning makes a big difference, which is why i'm still keeping my Isotek Genesis one regenerator for the clock. I also did another experiment to see if the improvements come from their advertised "ultra stable power", or just the isolation. To my surprise, the very clear improvements it brings when connected to the clock are also audible when the clock is disabled. Meaning what's actually improving things is noise not being able to get into the Rossini via the clock, not the regenerator actively improving the clock's performance.

 

[572217]

Mostly based on our biases. For example:

There's a very simple explanation here, and it is based on reality, not audiophile marketing. Electrical noise and jitter makes absolutely no difference to SQ, until it reaches the analogue portion. How come Netflix works perfectly without tons of snake oil? Not to mention far more complex things than audio. Ironically it's exactly the thing you are discarding that makes all of the difference. The Innuos sounds better than Nucleus and other solutions when used as a server, because you run Ethernet via its filters and it prevents an albeit small amount of electrical noise to reach the dac, which does have an impact on SQ.
I did this experiment while I had the Innuos here. Ethernet from Innuos to Rossini sounded better than Ethernet from my router, even with the Innuos used as a server in both cases. That's where the improvement is. However, add an optical isolator powered by a very low noise LPSU (all for less than 300$ btw), and it's a dramatic improvement over the Innuos, and completely eliminates any advantage it has over Nucleus (weirdly, not over the gaming pc used as a roon endpoint, which is why i still suspect Roon does something funny with the stream, or there is an inherent disadvantage to hosting Roon on a windows pc compared to their optimized OS), leaving the Innuos with a difference in sound which is clearly EQ. I suspect this is one of the reasons why it's not actually Roon certified, even after all this time .

We are in agreement that power supplies and conditioning makes a big difference, which is why i'm still keeping my Isotek Genesis one regenerator for the clock. I also did another experiment to see if the improvements come from their advertised "ultra stable power", or just the isolation. To my surprise, the very clear improvements it brings when connected to the clock are also audible when the clock is disabled. Meaning what's actually improving things is noise not being able to get into the Rossini via the clock, not the regenerator actively improving the clock's performance.

I’m a computer scientist by training and understand most things around data in principle.

I chose to stay copper based for the stated reason. I find it unlikely an LPS would take care of all the noise a tiny circuit like an optical converter creates. Resonance control is clearly audible and that’s just taking care of minute vibrations in circuits. I have no doubt you get that leap going optical. What I doubt is whether that’s the end of the line. I will try optical when dCS incorporates an interface into their Upsampler. Right now, I stopped missing vinyl and I owned really lovely players.

Compartmentalised arguments about complex systems are not so useful. Especially by us relative amateurs in lots of topics. Collective sense making and empiricism are useful.

 

[buzzlulu]

Another dCS owner subsequently bought the Mini Trøn for his Rossini and his jaw dropped as well.

hmm…that might be me
I even had two headphone friends over - one who also listens to 2 channel via a Rossini- and both were impressed by the mini Tron cables as well.

I use a Graceline level 2 xlr in my Stax/T2 system and just bought a second hand set of Black Cat 3202 xlr's for another system. Chris Sommovigo was a talented person. A shame losing someone like that

PS - do we know each other from the dCS forum - possibly via PM? Are u a friend of Ben?

 

[572217]

hmm…that might be me
I even had two headphone friends over - one who also listens to 2 channel via a Rossini- and both were impressed by the mini Tron cables as well.

I use a Graceline level 2 xlr in my Stax/T2 system and just bought a second hand set of Black Cat 3202 xlr's for another system. Chris Sommovigo was a talented person. A shame losing someone like that

PS - do we know each other from the dCS forum - possibly via PM? Are u a friend of Ben?

There’s no escaping the dCS fraternity.

I only learned of Chris’s death in April. He was an extraordinarily talented man and a pleasure to deal with. I wish him well for his onward journey.

Mostly based on our biases.

I looked into sense perceptions a bit more deeply in the past couple years. What comes in raw is too complex to be processed by our brains in its entirety and all the time. It would be too ‘expensive’ and paralyzing to process all of it. Over the years our brains form pathways and interpretation patterns to reduce complexity. That process and its result is not uniform across people.

Some of it is dispositions and some of it training – nature and nurture if you will.

A world class sound engineer listens differently from me. And the mastering engineer yet in another way. The conductor in another way than the cellist than the composer. A top notch street photographer sees a very different city than the tourist does. Two people watching a usual busy scene on Times Square in New York will give you markedly different accounts of what they see.

If you are a seasoned meditator and gain some agency over what the mind does, perception changes again. The cognitive overlays can take a back seat.

 

[adrianm]

I’m a computer scientist by training and understand most things around data in principle.

Compartmentalised arguments about complex systems are not so useful. Especially by us relative amateurs in lots of topics. Collective sense making and empiricism are useful.

I'm a software engineer/technology consultant, with a focus on data atm. Ironically, I've spent 30 of the last 48 hours doing root cause analysis on some data issues. Enterprise software, not audio, so doesn't make me an expert by any means. Still, when comparing complexity of systems, audio companies are going to have a hard time convincing me they've reinvented the wheel Especially regarding streamers, networking, and software in general.

 

[572217]

I'm a software engineer/technology consultant, with a focus on data atm. Ironically, I've spent 30 of the last 48 hours doing root cause analysis on some data issues. Enterprise software, not audio, so doesn't make me an expert by any means. Still, when comparing complexity of systems, audio companies are going to have a hard time convincing me they've reinvented the wheel Especially regarding streamers, networking, and software in general.

That’s an interesting exchange. I used to work for one of the bigger systems integrators helping design software.

Of course, hifi components are tiny systems compared to – say – Google’s collective internet services, which clocked in at 2 billion lines of code in 2015. The large hadron collider had a paltry 50 million lines of code at the time. The genome of a mouse in comparison has 120 million DNA base pairs. The most complex machines built so far, are the ones manufacturing cutting edge chips. And looking at the complexity of structure and operations of for example Amazon Web Services is mind boggling.

What I was getting at is: putting together a hifi system is a complex issue. The number of influencing and interacting factors is immense. For me it is not feasible or desirable to get to a sufficient understanding of the science and engineering involved. That leaves me with informed experimentation and fruitful exchanges with peers.

I really like this quote:

“A little learning is a dangerous thing;
drink deep, or taste not the Pierian spring:
there shallow draughts intoxicate the brain,
and drinking largely sobers us again.”
—Alexander Pope, An Essay on Criticism, 1709

 

[Andrewteee]

I posted this in another forum and thought it could be informative here on Head-Fi too. I have not tried the Ferrum stack with the Aurender yet because that would require moving around components, but I suspect there could be an improvement in SQ there too. Because my hunch is the Nucleus+ is holding things back in comparison to the Aurender.

I know a lot of Head-Fi folks are interested in the Bartok HPA and I've read many comments from some who believe the Bartok HPA holds it back, but my experience implies deeper potential depending on the source. I can't understand why that would make a big difference with the HPA, but I can't untangle source to Bartok to headphone amp to headphones as outlined below. The Bartok HPA is really delivering the goods now...

I had an interesting experience last night (and confirmed today).

On a whim, I decided to listen to the headphone output of the Bartok, which I have not done in months, and compare it to my Ferrum stack (new Wandla DAC, Oor amp and Hypsos power supplies). I was comparing Roon to Roon (ethernet on the Bartok, USB on the Wandla). Sounded fine, but reinforced why I prefer the Ferrum Oor for headphone listening. The Bartok DAC may be better, but the HPA was holding it back. And the Ferrum system sounded really, really good together.

Then I switched over to the Aurender AES/EBU input on the Bartok playing through the Aurender app and OMG!

There was an immediate and dramatic improvement in the Bartok headphone capabilities. It jumped ahead of the Ferrum stack by a margin. It sounded wonderful on the Meze Elite and really kicked the DCA Stealth into gear! Gave me that thrill of rediscovering music I love.

This was the first time I had listened to the Bartok HPA with the Aurender (N20). I knew there was an improvement playing through my two channel speaker system, but the improvement was even more clear and significant through the HPA (probably more direct).

This is making the Bartok a whole new DAC experience and now I’m considering the APEX upgrade, after hearing the potential, and keeping in mind the simplicity of a more streamlined system. I still have to think about it, given the hefty cost as a % of the original cost of the Bartok, but I’m closer to a decision, as an ongoing investment in the Bartok. Throw in a clock too, why don’t you.

Oh my, source matters! I’ve been Roon based for so long, never really giving thought to anything different. I had purchased the Aurender for GAS reasons and to serve as an AES/EBU source for the Wandla DAC. But I started by trying it in the two channel system – the improvement was clear, so there it has remained. I was using the Roon Nucleus+ with the Wandla.

Clearly, a lot of credit goes to the Aurender, but it also demonstrates that the Bartok can scale with the source to further improve SQ. And it implies the Bartok HPA has more potential that some people think, depending on the source.

I was not expecting this, but there it is. Dang! What other potential lies undiscovered?!?

 

[572217]

I was not expecting this, but there it is. Dang! What other potential lies undiscovered?!?

I found the same, the source matters a lot. Power is crucial too. Adding a JMF PCD-102 filter to the system was a jaw dropper. It easily eclipsed the Ansuz d-tc power distributor I had before that. Might be another upgrade avenue to explore as well.

dCS clocking paired with the N20 might be an issue. When I had an N20 for auditioning at home it did not automatically switch sample rates. Which made the combination too cumbersome for my use case.