Vince, perhaps you want to give https://github.com/nihilux-org/roon-web-stack a try; it lets you control Roon from a Web Browser, without the ability to change any of Roon’s settings (I believe. I haven’t actually tried it out myself).
3 Likes
I have been using Qobuz connect on various Volumio devices recently and I am very happy with the SQ of connect vs the volumio app (which sounds not bad in itself, meaning better than Roon).
So I am impatiently waiting for dCS to implement Qobuz connect in their devices. Stream Unlimited is a connect partner, so that should make things easier.
Rudi, Qobuz Connect should make no SQ difference compared to Mosaic. Both get music data direct from the Qobuz server .Where it does make a sonic difference is for other systems which use the device app to receive the music . Mosaic app does not carry music and is only a control point.
Qobuz Connect will, however offer an alternative GUI.
thanks, Pete
Yes I’m aware that it SHOULD not make a difference, and everything I learned at ETH in Zürich during my studies suggests that it SHOULD not. Call me a sceptical objectivist.
However there are configurations which in my (subjective) view let the music flow and others that do less so.
I know that DACs that measure identically should sound identical (at least if one believes the folks at ASR). But we all agree that dCS DACs sound quite different from Topping of SMSL (which measure identically) or else we’d all be listening to Topping or SMSL (I have one at home so I can hear the difference whenever I chose) and save a lot of money. Can’t get more transparent than transparent, can you.
I have no idea why a Melco N1A should sound different from an Odroid N2 (bloody good for its price) or a Raspberry Pi (to my ears pathetically bad). They all transport bits in the correct order.
So, I have no idea why to me Qobuz connect sounds different, but it does. Anyone who is happy listening to Mosaic or Roon, I am happy for them, I have no objections whatever.
No doubt we’ve seen this picture before 
(unfortunately any analogy comparing analog outputs is not going to be very useful in the digital world)
This is one of those situations where I could literally show you, with Tidal Connect for example, that the way a track is pulled down from Tidal’s Servers is identical regardless of whether you use Mosaic or Tidal’s Application as the Control-point.
At the TCP flow level, there’s absolutely nothing different, no new technology invented or involved, in both cases the DAC’s Streaming board issues a HTTP GET Method to pull down the music stream. The packets are still received asynchronously, buffered, re-assembled, and decoded (FLAC) all on the DAC’s Streaming board in exactly the same way, and then played out synchronously to the processing/filter FPGAs and eventually makes it’s way to the Ring DAC.
There’s absolutely nothing different as far as the DAC is concerned. Nevertheless folks will still “hear” a difference 
3 Likes
Thanks, @Anupc
I am well aware of all that. I am also very aware of expectation bias. And if it is expectation bias that lets me prefer Qoboz connect, then I am happy to accept that. I still prefer it, no matter what the cause.
Why should I fight my preference and force myself to consider all streaming devices equal, if it does not add to my enjoyment of music? Just to be scientifically correct?
2 Likes
Agreed……however interesting you left Roon off this list (although I know where you stand on this )
PS - add JPlay to the Connect/Mosaic list as well. It also simply pulls from the Tidal/Qobuz servers
1 Like
Wholly agree with your sentiments; it’s everyone’s prerogative to subjectively prefer whatever they want to 
, no argument from me on that.
BUT, on a forum like dcs.community, I think it’s also important to clarify how things actually work, because there are many less technically inclined who will believe there’s an underlying technical difference causing a sonic improvement when in fact there’s no such thing.
Well, if A = B, and A = C, then by the transitive property of equality, B must = C
(A is Mosaic, B is Tidal/Qobuz Connect, and C is Roon 
)
3 Likes
I think that’s where the difficulty starts. A system playing through mosaic is clearly not identical to a system playing through Qobuz connect. The fact that the sequence of bits being sent to the dac is identical does not consider that the EMC environment of the DAC is different due to different processor loads, different spectra of the CPU radiation, different load on the PSU etc.
These differences may be minute, but there is clearly a different electromagnetic environment depending on the way the software performs its task. You may well argue that does not matter, because the differences are too small, but it is still not an identical situation. So I think we need to be a bit careful with the mantra that there is NO underlying difference, when obviously there is. Whether is makes a difference to SQ or not is a different question that would need to be established through double blind ABX testing.
Its the same absolute argument that the ASR folks use. If Amir cannot measure a difference between an SMSL DAC for 100 USD and a Rossini, then there is no difference to be heard (which if true would make this forum superfluous).
You never step into the same river twice. You never listen to the same reproduction twice.
You’re forgetting that it’s very easy to objectively prove that it makes no difference. I would refer you to my post on Roon vs. Mosaic, and leave it at that.
You will fail any such ABX test, I’d put money on it
Don’t get me wrong Anupc, I am not trying to troll you here. I am trying to point to a flaw in the argument. I have led that discussion over at ASR, because the folks there fall into the same trap of drawing a general conclusion from evidence of limited scope. It may well be that you are factually correct, but the reasoning you are using is flawed and does not prove the conclusion.
What your test proves is that the sequence of bits going to the Vivaldi DAC is identical in both cases. It does not prove that the sound the chain produces is identical. If indeed the sound were identical, why use a Vivaldi Upsampler at all? Any cheap streamer device will be able to feed an identical sequence of bits (which is what you have measured) to the Vivaldi DAC. There seems to be consensus in this forum that the presence of the Vivaldi Upsampler (and the Clock for that matter) does indeed change the SQ vs. a DAC only setup. This should not be the case if your experiment actually proved what it is claiming to prove.
The issue in the argument is that the setup of the experiment determines its outcome. Chose a different setup (ABX test) and the outcome may be different. You have chosen a very limited scope of the experiment and the result of the experiment is only valid within that scope.
I know this is just (maybe pedantic) semantics, but in order to use absolute terms like ‘there is no objective difference’, one needs to be extremely careful about the setup of the experiment, as the burden of proof that the absolute claim is correct lies with the designer of the experiment. One successfully passed ABX test is sufficient to prove the hypothesis of ‘no objective difference’ wrong.
I totally agree that I would probably fail a double blind test under pressure. But me failing the ABX does not prove the (not stated but inferred) claim that there is no difference in the analogue signal output by the DAC.
Sorry, but the flaw is in your argument here Rudi.
You mention there’s no evidence the same input data will produce the same output (sound quality) as it does not on different devices (Vivaldi vs Vivaldi + Upsampler + Clock).
BUT comparing apples to apples, same input data transmitted to a same audio device is expected to always produce exact same output… unless there’s kind of flaw or RNG somewhere.
Respectively, sending same input data to different kind/brand of DAC devices is totally expected to produce different quality of outputs.
Therefore same input data on a single Vivaldi VS a complete Vivaldi+Upsampler+Clock stack will produce different outputs…
@Anupc is right.
Cheers,
Erwan
2 Likes
Context: This is not about differences in sound quality, but about wordings.
I am not disputing that Anupc’s claim (that Mosaic sounds the same as Roon on dCS) MAY be correct. I am only pointing to the fact that his experiment does not prove this. It only proves that the Upsampler outputs the same sequence of bits using Mosaic or Roon.
Your argument is equally flawed as Anupc’s. You are making the assumption that I quoted above. The phrase ‘expected to always produce the exact same output’ is a subjective assumption, not a scientific argument. If A=B and B=C then A=C only holds true if the exact conditions under which A, B and C are defined / measured are indeed identical. In mathematics this is easily done, in physical reality this is extremely difficult.
My argument is NOT about the question whether Anupc’s conclusions from the experiment (that Mosaic and Roon sound the same) are factually correct or not. It is about the validity of the argument that leads to his conclusion. It is about having to be careful how one words a reasoning if it contains phrases like ‘no objective difference’,
Again, if Anupc’s experiment proved what it claims to prove, there should be no difference between a chain of Upsampler → Vivaldi DAC and a chain of cheap streamer outputting the same bitstream → Vivaldi DAC (which is exactly your argument in your post about the same input producing the same output).
I didn’t think you were trolling, I’m sure you were serious 
Just to be clear though, my posts are not general commentary about what makes a difference or not in other systems or posts on ASR, etc. etc…
My posts in this thread are specifically about dCS systems when comparing Mosaic vs. Roon. vs. Tidal/Qobuz Connect, and whether or not there’s a difference in SQ.
On that point, to suggest that a difference in control-points can cause EMI differences that impacts sounds quality you hear, is quite frankly absolutely ludicrous. It’s typical of the rubbish talking points from snake-oil vendors like JCAT/JPlay.
Sorry, your points above tells me you really don’t understand what the test was about, what it does, and what it proves. If you do have a technical background, I suggest you re-re-read that post of mine carefully and think about it. Keep in mind I’m not interested in generic setups, I’m specifically posting about dCS setups and what makes a difference or not.
Have I logged on to the ASR forum in error
There’s a reason I avoid that site.
2 Likes
Without at all entering the debate here, I just want to say that absolutely LOVED the explanatory power in your caveat statement on the ABC equivalency.
The simplicity of your argument that the measurement protocols for when one equates A to B must absolutely be the same when one equates B to C to make that A equivalence to C, just blew my mind! Such a clever, yet straightforward, way to expand one’s thinking beyond the obvious mathematical equivalence.
Great stuff!!
BTW, we have sure moved far afield from the Qobuz Connect headline, haven’t we?
1 Like
Thanks for clarifying this, Anupc. Limited to a dCS setup and your dCS Vivaldi setup specifically, your conclusions of course hold true, as your are changing only one variable (Mosaic or Roon processing) and everything else stays the same.
I had originally posted that I found a difference between Qobuz connect, Roon and the generic Volumio app on a volumio based system. An RP3B with Hifiberry PCM5122 HAT. That is a completely different environment and I wouldn’t even be sure whether the idea of CPU usage and EMC influence is so ludicrous in such a non audio optimized and low cost system. Of course in the case of the Vivaldi it would indeed be ludicrous to assume these things would make a difference. Would be disastrous if they did.
Our discussion is a case to my point that scope of consideration matters. I was basing my arguments on a general view including my volumio Raspberry, you were approaching the discussion from your dCS Vivaldi perspective. No wonder we disagreed.
1 Like
Ahhh. Glad thats cleared up. I may have posted more than once; I find the disaggregated nature of the dCS Vivaldi System makes it easy to make precise objective measurements, but the results are really only meaningful w.r.t dCS systems.
That said, there’re still plenty of regular audiophile beliefs which can be quite readily debunked. Sites like Positive Feedback and forum sites like WBF are full of absolute batsh*t crazy rubbish, which frankly is what elicits the likes of ASR which goes to the opposite extreme
3 Likes
Is the Qobuz connect update going to be available to discontinued model? I own a Network Bridge
Probably yes as the Qobuz Connect update will be to Mosaic, not to the unit’s hardware (so as per Tidal Connect).AFAIK Network Bridge is still supported in this respect.
1 Like