DCS Vivaldi

I have a similar setup to Kingsrule (Roon -> Upsampler -> DAC) and am having no issues with 2.02 either.

What I would do is revert back to 2.01 (hopefully you have the CD), test that all works and update to 2.02 again (e.g. perhaps the upgrade got corrupted along the way).

MadFloyd said:

I have a similar setup to Kingsrule (Roon -> Upsampler -> DAC) and am having no issues with 2.02 either.

What I would do is revert back to 2.01 (hopefully you have the CD), test that all works and update to 2.02 again (e.g. perhaps the upgrade got corrupted along the way).

Click to expand...

Funny you should mention the possibility of a corrupted upgrade.

When the upgrades to the Transport and DAC first came out, Martin at dCS sent me two CD's in the post.

The CD upgrading the Transport to 2.02 worked without trouble. However, the CD fro upgrading the DAC to 2.02 would start, but about half an hour in would stop without reason.

I later learned that a batch of corrupted discs had been released and my CD for the DAC upgrade was later replaced with a good copy.

Kingsrule said:

Just updated dac to 2.02 and having issues:

Roon, no processing

Using upsampler

set to DXD

When shuffling in Roon, if a 24/96 or 24/192 track comes up, Roon shows its playing but there is no audio. When a 44 track comes up, audio is back

If I set upsampler to DSD everything plays. Tried different resync times but behavior remains the same

Anyone experience this? Any solution besides reverting back to 2.01?

Click to expand...

Wanted to jump in here before you wasted any more time.

Do you have the upsampler slaved to the DAC's clock using Universal Master Mode (in other words no Vivaldi clock)? If so then this is a regression in 2.02 that is impacting a few users. dCS is aware and it's slated for fix in the next release. No timeline on that yet. Best solution is to roll back to 2.01.

As for the Stream 800 vs 820 cards. There's no meaningful difference between the two aside from what has already been mentioned (WiFi, slightly faster CPU). The 820 was chosen for the bridge based on its WiFi capabilities. The CPU difference doesn't have any sonic impact due to the implementation. There are no plans to implement the 820 card in the Vivaldi or Rossini.

As for the Network Bridge vs ??. Well, I've been testing permutations of that since last summer. The bridge feeding the Vivaldi DAC via dual AES with good cables is quite a bit better than the Rossini (with or without the clock on the Rossini or Vivaldi). Hell, USB into the Vivaldi DAC is quite a bit better than the Rossini over the network. That's not to say that the Rossini isn't awesome in its own right (it really is), but the Vivaldi is simply better. It SHOULD be better considering the price differential! One of my Rossini customers didn't believe me so he tried for himself and now has a Vivaldi and a Network Bridge.

When buying the Vivaldi full stack digital Transparent XL cable loom in two phases what is the recommended order - should we go with the clock set or the AES/EBU set first?

AMP said:

As for the Stream 800 vs 820 cards. There's no meaningful difference between the two aside from what has already been mentioned (WiFi, slightly faster CPU). The 820 was chosen for the bridge based on its WiFi capabilities. The CPU difference doesn't have any sonic impact due to the implementation. There are no plans to implement the 820 card in the Vivaldi or Rossini.

Click to expand...

The word I have on the Stream 800 is that dCS worked with the manufacturer on the circuit design for the 800, so in a sense, card is optimized for the Vivaldi Up and Rossini. The 820 is a stock part.

As for this review, I question the credibility of the reviewer. The review was not a direct A:B comparison but the reviewer was comparing what he heard from the Bridge to what he remembers hearing from the Rossini. I was also told that he did not have a Rossini for a long term test but heard it at an audio club event where it was paired with some sketchy vintage speakers and gear. Most people say that sound memory is measured in seconds or minutes at most. Trying to compare to something you heard months ago is just not legit.

ksalno said:

The word I have on the Stream 800 is that dCS worked with the manufacturer on the circuit design for the 800, so in a sense, card is optimized for the Vivaldi Up and Rossini. The 820 is a stock part.

As for this review, I question the credibility of the reviewer. The review was not a direct A:B comparison but the reviewer was comparing what he heard from the Bridge to what he remembers hearing from the Rossini. I was also told that he did not have a Rossini for a long term test but heard it at an audio club event where it was paired with some sketchy vintage speakers and gear. Most people say that sound memory is measured in seconds or minutes at most. Trying to compare to something you heard months ago is just not legit.

Click to expand...

My dealer feels the same. Something was off and it's not the network card being the thing that improved...He still plans to listen to confirm using the same cables, power cords etc etc. One theory is different cables or isolation on the reviewers network bridge vs Rossini...that wasn't mentioned in the review...

AMP said:

As for the Network Bridge vs ??. Well, I've been testing permutations of that since last summer. The bridge feeding the Vivaldi DAC via dual AES with good cables is quite a bit better than the Rossini (with or without the clock on the Rossini or Vivaldi). Hell, USB into the Vivaldi DAC is quite a bit better than the Rossini over the network. That's not to say that the Rossini isn't awesome in its own right (it really is), but the Vivaldi is simply better. It SHOULD be better considering the price differential! One of my Rossini customers didn't believe me so he tried for himself and now has a Vivaldi and a Network Bridge.

Click to expand...

Would the Vivaldi DAC + Network Bridge still be better when compared to the Rossini DAC + Rossini clock with upsampling to DXD/DSD engaged?

ksalno said:

As for this review, I question the credibility of the reviewer. The review was not a direct A:B comparison but the reviewer was comparing what he heard from the Bridge to what he remembers hearing from the Rossini. I was also told that he did not have a Rossini for a long term test but heard it at an audio club event where it was paired with some sketchy vintage speakers and gear. Most people say that sound memory is measured in seconds or minutes at most. Trying to compare to something you heard months ago is just not legit.

Click to expand...

That's what I thought too - that he had only heard it at some audio club meeting. However, from the Polish version, he has had it for a long term test (but probably not as the same time as the Network Bridge?):

http://highfidelity.pl/@main-2645&lang=

https://translate.google.com/transl...://highfidelity.pl/@main-2645&lang=&sandbox=1

AMP....Do these DAC updates do anything besides operational things? Any sonic changes between 2.01 and 2.02?

I just reverted back to 2.01...I think things aren't right. Got message "Applying Software Update..Do Not Switch Off"....50 min+ still there, I rebooted DAC and 2.01 shows in menu. Happily, everything appears to be working

Since dCS products are so software driven, I do not understand why they chose to handle their updates the way they do. Put them on their website to be downloaded, inform users of potential issues and put more resources into software development. Contacting a dealer or distributor is an archaic way to get updates and information. Seems like they have one guy in a backroom writing their stuff.

Kingsrule said:

AMP....Do these DAC updates do anything besides operational things? Any sonic changes between 2.01 and 2.02?

Click to expand...

The only Vivaldi update that had a sonic benefit was the move from 1.x to 2.x. That included the updated mapping algorithm and filters. The subsequent updates have been functionality changes. The big change in 2.02 was an update to the way that the DAC resyncs on a sample rate change so as to speed up that process.

I just reverted back to 2.01...I think things aren't right. Got message "Applying Software Update..Do Not Switch Off"....50 min+ still there, I rebooted DAC and 2.01 shows in menu. Happily, everything appears to be working

Click to expand...

You should have gotten a message on the screen saying, "Please Reboot" or something to that effect. There is a bit a of a delay between the end of the file / disc playback and that message. I've occasionally seen glitches in the updates and in those cases verify that all of the versions reported are correct. If anything is still listed at 2.02 then you should run the update again.

Since dCS products are so software driven, I do not understand why they chose to handle their updates the way they do. Put them on their website to be downloaded, inform users of potential issues and put more resources into software development. Contacting a dealer or distributor is an archaic way to get updates and information. Seems like they have one guy in a backroom writing their stuff.

Click to expand...

dCS has a fairly large engineering team (especially by high-end audio standards). That team is split into hardware, software, and mechanical design and that group is working on a large number of projects at the moment. The universal master mode glitch popped up after the release as something slipped through testing. This is highly unusual for the dCS FPGA code as their testing methods are extensive.

The update process on the Vivaldi is a side-effect of the hardware design. The high-speed bus that connects all of the different FPGAs and their associated flash chips isn't connected to the audio pathway at all. The only way to get the update code onto that bus is to wrap it in a PCM stream and "play" it. In that process the onboard micro controller (which acts as a supervisor) intercepts the "audio" stream, strips off the wrapper and sends it out to the various FPGA flash locations. This is all by design as in order to make updates faster would require a specific interface just for updates and that would be another faster computer running in the box just to manage updates. That's not great for audio performance. Updates are faster on the Rossini because the architecture is different and the Stream card is connected to the high-speed bus.

It varies country by country, but updates are usually managed by the dealer. I know I visit my clients to install updates. The time required to install the update provides a good opportunity to have a couple of beers, catch up, and (sometimes) spin some LPs.

Kingsrule said:

AMP....Do these DAC updates do anything besides operational things?

I just reverted back to 2.01...I think things aren't right. Got message "Applying Software Update..Do Not Switch Off"....50 min+ still there, I rebooted DAC and 2.01 shows in menu. .

Click to expand...

I had that same issue on 2.01. dCS advised me to power off and all was fine. I let is sit there for like 2 hours....before just to be sure

jfrech said:

I had that same issue on 2.01. dCS advised me to power off and all was fine. I let is sit there for like 2 hours....before just to be sure

Click to expand...

Probably it is why my distributor/dealer advised me to hold the upgrade when I asked for the CD.
Still running 2.01.

microstrip said:

Probably it is why my distributor/dealer advised me to hold the upgrade when I asked for the CD.
Still running 2.01.

Click to expand...

John Quick of dCS gave me the 2.02 disc at the LA Audio Show. I put it in my Oppo UDP-203 using coax digital out to the Vivaldi, and everything worked flawlessly.

Anyone compare a Vivaldi stack vs MSB Select II ?

jfrech said:

I had that same issue on 2.01. dCS advised me to power off and all was fine. I let is sit there for like 2 hours....before just to be sure

Click to expand...

I let mine sit for 6 hours until I got brave enough to power it off and on again.

Wow, I've been going nuts with clocking bugs and just now read AMP's posts. Good to know dCS is aware of it. Here's hoping a solution comes soon.

2.01 still dogs it changing sample rates when I shuffle in Roon...not as much as my copy of 2.02 but really??

I know that has been discussed previously at least by some forum members here, even I used to provide my feedback on that. Interestingly perhaps it is due to audio chain and changes etc, experiences might change. Last weekend I tried to test volume control of Vivaldi DAC without the preamp again, so I directly connected Vivaldi DAC to Constellation Hercules II monos. I was surprised with the results. Transparency and resolution were audibly improved and while the presentation has been more detailed, it was not on clinical level, actually far from that, music had more texture it was more natural. If anything to criticize, it might be regarded as slightly dry. Well I need to say I am used to tubes at least in some parts of my audio chain for years so that slight dryness might be due to lack of tubes in this case as well I am not sure. Anyway, bottom line is it was easier for me to prefer no preamp option at the weekend. Given my respect for Ref 10 and my experiences with the unit which I used to test some other fine preamps previously, the situation now is a bit unusual. One might be curious about the tubes of Ref10 unit that I have, let me tell I used to replace the old tubes and newer ones have worked over 150-200 hours so far and they are from ARC factory so in that case I rule out the tube factor. I understand the volume control within the DAC performs within digital domain so it is not an analog preamp.

I am not trying to come to any conclusions or questioning the necessity of a preamp. Besides, listening to analog by itself is a major factor to necessitate a preamp. What I am after is, in case of optimal performance for DCS Vivaldi DAC, does a preamplifier -even a very good one- slightly degrade the sonic performance of the Dac? For sure it is hard to conclude as musical assessment is fairly subjective. At least for some Vivaldi users who have tried the DAC w/o preamplifier might give their opinions. Is digital domain safe enough to have a volume control as good as best analog preamps?

number95 said:

Is digital domain safe enough to have a volume control as good as best analog preamps?

Click to expand...

Funny you should bring that up today... just last night I emailed a friend why digital volume controls SUCK, including the one in the Vivaldi 2.0 that I had in for evaluation (though to a lesser degree - I suspect they interpolate more and more as the volume is lowered, up to a point). I will be posting soon about this in my system thread, but suffice it to say that I have proven unmistakable and repeatable loss of resolution with complex material (Mahler 2nd) with simultaneous increase in glare, as the volume is lowered - yes, I feel digital volume controls and the inevitable loss of bit resolution are the predominant cause of these issues, with jitter second. Basically, once you start losing bits, it's impossible to properly reconstruct the signal - I now perfectly understand why Spectral, Schiit and others never did and never will offer digital volume controls.

On the other hand, finding a transparent preamp is equally challenging, and only with my 30SV (in RCA-in mode) have I been able to get the full RBCD resolution coming out of my modified Alpha... and my modifications are mainly responsible for enabling me to bring the digital volume control all the way up, to full 24 bits.

So with respect to your comparison, be sure to compare with the exact same digital volume control setting, and see if a preamp is still veiling things. I personally would never drive amps directly from a DAC, and though I have heard wider dynamics driving direct, pre-30SV, the sound was way off, because I was forced to lower the digital volume quite a bit.

ack said:

Funny you should bring that up today... just last night I emailed a friend why digital volume controls SUCK, including the one in the Vivaldi 2.0 that I had in for evaluation (though to a lesser degree - I suspect they interpolate more and more as the volume is lowered, up to a point). I will be posting soon about this in my system thread, but suffice it to say that I have proven unmistakable and repeatable loss of resolution with complex material (Mahler 2nd) with simultaneous increase in glare, as the volume is lowered - yes, I feel digital volume controls and the inevitable loss of bit resolution are the predominant cause of these issues, with jitter second. Basically, once you start losing bits, it's impossible to properly reconstruct the signal - I now perfectly understand why Spectral, Schiit and others never did and never will offer digital volume controls.

On the other hand, finding a transparent preamp is equally challenging, and only with my 30SV (in RCA-in mode) have I been able to get the full RBCD resolution coming out of my modified Alpha... and my modifications are mainly responsible for enabling me to bring the digital volume control all the way up, to full 24 bits.

So with respect to your comparison, be sure to compare with the exact same digital volume control setting, and see if a preamp is still veiling things. I personally would never drive amps directly from a DAC, and though I have heard wider dynamics driving direct, pre-30SV, the sound was way off, because I was forced to lower the digital volume quite a bit.

Click to expand...

Thanks for the sincere feedback. Interestingly, in terms of glare/noise, I am quite sure without my preamp involved there was less noise/grain, perhaps tubes at the chain has been contributing somehow, who knows. I succeeded to push the volume w/o preamp much higher than normally the levels I prefer with Ref10. The only point that I criticized so far has been a bit dryness which it seems it might be related to your findings. I take your feedback as a strong opinion, will try some other preamps in near future as if my findings are particularly related to Ref10 or not.

number95

Our systems are very similar and I too prefer the Vivaldi dac as the preamp. I will say that the Vivaldi changes dramatically based on what support it is on and what feet/cones are under it, more so than any other component I have had.