Innuos Zenith Mk3 or OpticalRendu + separate server?

[incus]

On 2/25/2019 at 8:51 AM, barrows said:

It is plausible, but unproven, that more error correction by the endpoint will generate more processor noise, and could result in a decrease in sound quality.  In any case, for that reason alone I advise folks to avoid WiFi for music distribution if possible.

 

 The other problem is the WiFi transceiver itself, this is by definition a source of noise

But you've just answered it!  What you say here applies perfectly to optical transmission as well! The more error correction the optical rendu will be forced to do, the worse it will sound. So, logically speaking, feeding it a better clocked signal will improve SQ. And the optical receiver in this case is by definition a source of noise too....

 

I realize these comments are a bit old now but as I read through this new thread, I cannot help but say that much of what you have said here is, to the letter, 100% false according to my extensive experimentation with streaming music in both my two-channel rig and my desk-top rig. I am, quite frankly, shocked to hear this coming from someone so deeply involved in manufacturing part of a cutting-edge digital streaming playback system. I certainly hope it's not because Sonore doesn't make servers (anymore) and has focused so much on the renderer. I'd be curious what your system is such that you don't hear difference between sources. There is a distinct, obvious and repeatable difference in sound and sound quality between my SGC i7 Roon server and my new SotM Roon server as fed through the exact same system. I have my understandings for why there is such a difference and J. Swenson and the folks at Uptone certainly have their understandings of why there is such a difference - so I could suggest you stop passing off subjective experiences/motivations as objective truth. Sonore wouldn't keep coming up with better and better products if the objective truth had already been discerned and implemented, right?

[incus]

On 2/25/2019 at 11:07 AM, barrows said:

And, again, if accumulated clock phase noise in an asynchronous Ethernet environment, would one not have to draw the conclusion that streaming Tidal and Qobuz would be hopelessly compromised by the hundreds of switches these streams pass through?  I guess this would be a good reason for the etherRegen and its "clock blocking", no?  and anyone who does not have this feature in their system would get absolutely unlistenable sound form Tidal or Qobuz (I use neither so have no experience...)?  Could the Internet even work if such problems really exist?

I think you are exaggerating the effects of phase noise to make a point here. But it's not that complicated. Tidal and Qobuz are clearly NOT good enough on their own or else no one would buy any of the products any of you sell! They would just stream into earpods directly from their iMac. Streaming services ARE good enough for that. Accumulated phase noise doesn't "ruin" the signal for that kind of playblack - which is probably the majority of consumers anyway. But it DOES degrade sound quality for those who have the systems, the money, the inclination, and the ears, to hear the difference. We are only ever talking about a tiny percent of music listeners. For us - I think I can say us - every little bit matters...

[incus]

4 hours ago, incus said:

 

 

4 hours ago, barrows said:

As mentioned, unproven.  But, to satisfy those who want a better clocked signal Sonore has solutions for that coming soon.  And I am looking forward to testing them, as I have also stated in this thread (perhaps you should read it all before becoming so critical).  I am skeptical as to whether this will really matter in a system with the opticalRendu, but I will test with an open mind.  And, if it does make an actual improvement, then great, even better sound!  I still have not received an answer to this inquiry from those who claim clock phase noise (upstream as it were) actually accumulates and degrades the audio performance, as you appear to claim much expertise in the matter, how is it that Tidal and Qobuz can deliver even decent sound quality, considering the hundreds of clock domains their signals travel through to on the way to ones home?

 

Your "extensive experimentation" does not include the opticalRendu, mine does.  My reports here are accurate to my experience, when you have experience of the opticalRendu in this context then you will you will have an opinion of what is "right" and "wrong" (or at least a sound quality that you prefer over another).

 

OK, either I misunderstand you or something is very wrong with the above statement?  Honestly, i am a bit confused here: All of your observations on sound quality of different set ups are based on measurements?  Or do you use "subjective" experiences to determine your "truths"?  For the record, i rely on both measurements, verifiable data, and subjective listening experiences in order to make informed decisions about how to improve system sound quality.

 

One more thing.  I accept that there is more than one way to skin a cat.  Multiple approaches to sound quality can lead to good performances and not everyone even agrees on what kind of sound quality is "best".  I have reported on my experiences, even if my experiences do not match yours, that does not make them "100% false". 

I only say 100% false to counter your repeated claims that absolutely positively nothing zero zilch nada makes it through optical. That is just as subjective as anything anyone else is saying, as you have no measurements to prove that *nothing* makes it through (and you dismissed some very interesting discussions about just these types of measurements from the What's Best interchange that was linked to.) You have come out firing with some big guns on the subject so perhaps you shouldn't be surprised that you are getting some edgy responses. I am saying that your "100% sure" claim that nothing makes it through optical is 100% false in my experiences with my system, which includes optical. I have optical isolation between server and desktop endpoint and changing servers has required many hours of note-taking to get a handle on all the changes I am hearing. This are not small changes between servers. They are large. 

 

So for the OP who is looking to understand where he should put his money, I will add my voice to the side that says he should absolutely look into the various server options and price them out and test them if he can because they do make a deep and obvious difference to the sound of a streaming playback chain - and not simply wait for the Optical Rendu and grab any old NUC as a source.

 

(The irony here is that I am pre-ordering an optical Rendu! Not just to test but to use because I have loved Sonore's products in the past and it fits perfectly with a need I have in another part of the house... I trust it will be better than all Rendus before it - all of which, however, benefited from a better source, as has been pointed out...)

[incus]

4 hours ago, barrows said:

I still have not received an answer to this inquiry from those who claim clock phase noise (upstream as it were) actually accumulates and degrades the audio performance, as you appear to claim much expertise in the matter, how is it that Tidal and Qobuz can deliver even decent sound quality, considering the hundreds of clock domains their signals travel through to on the way to ones home?

Reposting:

I think you are exaggerating the effects of phase noise to make a point here. But it's not that complicated. Tidal and Qobuz are clearly NOT good enough on their own or else no one would buy any of the products any of you sell! They would just stream into earpods directly from their iMac. Streaming services ARE good enough for that. Accumulated phase noise doesn't "ruin" the signal for that kind of playblack - which is probably the majority of consumers anyway. But it DOES degrade sound quality for those who have the systems, the money, the inclination, and the ears, to hear the difference. We are only ever talking about a tiny percent of music listeners. For us - I think I can say us - every little bit matters...

[incus]

6 hours ago, barrows said:

While I play only locally stored files and have no experience of streaming music from the Internet

Enough said.

[incus]

6 hours ago, barrows said:

The statement is that no electrical noise is transmitted through optical Ethernet interfaces

Because it depends what you mean by "transmitted." If you mean passed on to the endpoint/DAC then this is false because of the optical receiver and its power supply.

 

 

[incus]

Going back to the OP to try to keep things on track here, there are clearly different camps relating to the original question of single machine as server and endpoint versus a two box solution - and these camps have infiltrated many, many threads on this forum. Perhaps it would be helpful to look through some of those other threads covering this topic.

 

But in the meantime, the way I see it is this: you certainly can't go wrong with using a NUC (set up with AL in Ramroot or just using ROCK for now, assuming ROON is your thing) AND the Optical Rendu as renderer when available (OR another renderer that others might recommend, which for me, would be the SOtM sMS-200 Ultra SE with Master Clock connection - but that is another discussion entirely.)

 

That would be quite a fine system, for sure, and it would give you the flexibility of playing around with the settings on your server and/or of upgrading your server later - AND allows you to swap in different renderers to test/compare/upgrade as they become available over the coming years. Splitting costs between components makes upgrades easier and more cost/time efficient. Also, this way you can also hear for yourself what differences these components make.

 

In other words, this discussion has perhaps come too far too fast, robbing you of your chance to do your own experimentation.

 

I am a fan of flexibility and the problem with the one box solution as I see it is that you are married to that rather expensive box for a while (in order to get a return on your investment) and the only way to know if you can achieve higher SQ is to go out and start buying/comparing many of the other products you have already chosen not to buy! If indeed a NUC + renderer +power supply is roughly the same $ as an audiodphile server one-box solution (not to mention that many use these with a tX-USB Ultra so there is the cost of that as well!) then, as I have laid out, I would opt for the two box solution as your first forray into this world.

[incus]

5 minutes ago, barrows said:

Back to Qobuz servers: so if the "source" matters most, then clearly Qobuz would have to be using "audiophile" servers to produce very good sound quality, given the thesis, right?  So, what is the takeaway here?  Either all those audiophiles who find no problems with the sound quality from streaming services are mistaken, or really, the "source"is not what matters most at all...

I will try a third time to post my feelings about your repeated  return to this argument. Perhaps you could read it and react. It is about logic, not preference. Here is what I wrote, (twice!) with some changes this time:

 

I think you are exaggerating the effects of phase noise "build-up" to make your point here. No one is saying it ruins the signal to the point where, as you suggest, it "ruins" the internet (!).  Tidal and Qobuz clearly don't use audiophile servers. That's not in dispute, either.

 

But Tidal and Qobuz are clearly NOT good enough on their own sound-wise-- or else none of us would buy any of the products any of you are selling! We would just stream into earpods directly from our iMacs/iPhones and happy. Streaming services ARE good enough for that.

 

But they aren't good enough for people here. Because everything that comes before your modem DOES degrade sound quality for those who have the systems, the money, the inclination, and the ears, to hear the difference.

 

We are only ever talking about a tiny percent of music listeners. don't forget. For us - I think I can say us - every little bit matters...  So we go out and buy switches and servers, cables and renderers that mitigate the deleterious elements that Tidal/Qobuz server/ISP,  etc. introduce. I hope you can see the logic in this.

[incus]

11 minutes ago, barrows said:

I read it when you suggested the above previously.  While this may be your experience, it is not that of many other audiophiles I have spoken with.  There are plenty of audiophiles with systems the equal, or better, than yours, who enjoy listening to music files streamed from the Internet and find no difference in sound quality.  Certainly streaming from Qobuz does not require that one is listening via a cell phone and earbuds.  Indeed there are even manufacturers of very high end gear using Qobuz at audio shoes to demo their gear, do you think they would be doing this if the streaming service was hopelessly compromised for sound quality?

Okay so you clearly did not follow my argument.

 

1) "Find no difference in sound quality" compared to what?? Do you mean local vs. streaming? Then all that means is that they have optimised their streaming playback chain to achieve great SQ on par with local. That has nothing to do with what I am saying.

 

(And I will ignore the inherent dig at my system by your silly quoting of someone with a "better" system. Better based on what? Cost? How the F do you know what kind of SQ I am achieving in my home? I DETEST this line of reasoning - "better systems than yours." Ha.)

 

2) "Certainly streaming from Qobuz does not require that one is listening via a cell phone and earbuds." Whaaaaat? Did I say that? Please read again. Goodness my point is the opposite. Sound is good enough for a cellphone. So why not just stop there? Because "Audiophiles" don't stop there. Ever. They try to improve the signal...

 

3) And your last point IS my point! "High-end" gear can bring improvements to Qobuz. No dispute there. I use Tidal and love the sound I am getting through my system. No one anywhere ever said a streaming service was "hopelessly compromised" for sound quality by phase noise accumulation. Again you are trying to put words in our mouths that "phase noise accumulation" "hopelessly compromises" a signal. That is a straw man you have created for your own argument. BUT -- that signal certainly benefits from the kind of "cleaning up" that your Sonore products - and many others - do. That is precisely my point. There is something to clean up. Some things we don't entirely understand yet. But I suspect we will once JS is done...

[incus]

12 minutes ago, barrows said:

My only point here is to suggest that while you think Internet based file streaming from Qobuz is compromised vs. locally stored file playback, there are other audiophiles who do not agree with this assessment, and that there is no consensus on such at this time. 

I didn't say that. Read again. To me there is a small difference between streaming and local - they each have their own issues which I have tried to address in different ways - streaming primarily through reclocking and local primarily through power/filtering/stabilization of the hard drive. But each has its issues. And I would disagree with you, K-man, and say the exact opposite - if you DON'T hear a difference between local playback and streaming then there is something wrong with your system - meaning there is so much noise that it covers up the differences.

 

14 minutes ago, barrows said:

hey are not designed to "clean up" anything-instead they are designed to create a totally clean USB output in order for the DAC to produce the best performance it is capable of.

You contradict yourself here, no? Clean vs. clean? Anyway, you know what I mean. You have created a product you believe enhances the sound coming from a server be it from local playback or streaming from the internet. In other words, there is a "problem" you are presumably addressing with your product beyond mere convenience. So to say you are not "cleaning" the signal is splitting hairs. You are taking a signal and outputting a better signal or there would be no reason for your product to exist.

 

20 minutes ago, barrows said:

 Certainly you accept that some audiophiles have systems better than yours?

And no, I stay out of the ranking of systems entirely. Seriously. It's a terrible trap to fall into as it usually equates to cost. And cost don't mean shit.  Nearly none of us have heard each others' systems in our own homes, so there is absolutely no basis on which to rank these systems other than on perceived quality (based on what? reviews? people talking up their own systems? measurements of some parameters of some components taken in isolation.... so... cost? narcissism? some scope?  no thanks...) Plus the fallacy of the argument - "I know guys with killer systems who disagree with you.." I don't care. I am not talking about my system outside of trying to offer the OP an intelligent, cost-effective path toward achieving his goals. (And to refute the assertion - contradicted by so many anyway - that source doesn't matter in computer playback...)

[incus]

3 hours ago, vortecjr said:

My gear is quite good IMO. The Synology NAS, AT&T fiber service and router, TrendNet switch, cables are all quite good. I'm also getting stunning results.

I abhor system snobbishness but looking at this list of components I would hope you could at least entertain the possibility that this system is actually letting a lot of noise of various types through and therefore affecting your conclusions about what each component does or doesn't do. Each of these components could be replaced/upgraded/optimized in many ways along the lines of what many people here have been doing and posting results about. Regardless of whether you are getting "stunning" results. I am glad your new optical rendu has brought you that level of satisfaction, but all improvements are relative to where you started.

3 hours ago, vortecjr said:

My conclusion (audiophile euphoric type expressions side) is that it's sad that people feel they need to do more than they have to get stunning results.

Now here you really start to tread on thin ice. As one of the horde of people here and on all the various audio fora who continually strive to "do more than they have to" to optimize their playback chains, I'd like to point at the no one would be buying any of your products if we didn't fall into this category. Just because you have arrived at a place where you are satisfied with your sound (without going into the ethical issue of the fact that this involves the latest product you are about to bring to market!), this does not make anyone else "sad" for pushing further and staying tuned in to the latest developments from other corners. Many an audio product manufacturer has gone done the rabbit hole of extolling their product as end game, only to be eclipsed by the next "endgame" product - and the next and the next... Just be careful -- we sad people most likely make up a massive part of your clientele!!!

3 hours ago, vortecjr said:

BTW my opinion is independent of what the OP chooses as his solution.     

But it's not at all, is it? You make the very product he is considering and have just unreservedly endorsed it here AND you have just called anyone "sad" who feels the need to look elsewhere. There are no independent opinions when you are a manufacturer, sorry.

1 hour ago, vortecjr said:

For example a linear supply on a switch made no change at the output of the DAC

And yet when I swap out the LPS 1.2 powering my SOtM switch for an sBooster 12V, a SOtM sPS-500, a HDPlex 200W, or a battery pack, I got four very different sounds coming through my headphones. And you want to know why? Because the end of the the playback chain in ANY system is NOT a scope or piece of software measuring a signal - it is a pair of HUMAN EARS. A robot may "hear" no change in the output of a DAC, but a human being does. Rather than this being some unfortunate subjective artifact that must be thrown out in order for technical conclusions to hold, it is the ESSENCE of audio. The human ear is arguably the most important "component" in the playback chain. Removing it from your analysis is like removing speakers or headphones from a system and then trying ti convince someone of the purity and beauty of your signal by showing them a bunch of numbers on a piece of paper. Trust me this is the best sound you will ever hear. Wait, did I say hear? I meant see... I mean -- just look at the numbers...  In other words, instead of calling measurements "facts" and ignoring all other data, you need to take my conclusions - and those of countless others here - as facts also, facts that must be explained, not discarded. Otherwise you are just looking at one type of "fact" over another. And we know where that leads in our current political environment...

[incus]

I wasn't quoting you with 'endgame' but you more than implied that the opticalRendu is all that someone really needs right now because other 'upgrades' are not really worth it.  If that's not really what you meant then I would be careful with comments like the 'sad' comment. And I don't take it personally at all - just unpacking your  words -- with an admonishment to be a little more careful how you phrase things.

 

7 hours ago, vortecjr said:

If you can here the difference between various PSs on an endpoint or server then there might be a bases for it, but not on a switch.

 

What can I say about this? You say I can't possibly hear different power supplies on my switches and yet I do. Did another A/B over the weekend between Uptone LPS1.2 and SOtM sPS-500 on my SOtM switch and the difference is apparent and repeatable, even blind. I don't know what else to say except come on over and hear for yourself... But in the meantime I would suggest you avoid categorical statements such as "but not on a switch." Really? Unless you've tested them all, I see no basis for this statement...

 

7 hours ago, vortecjr said:

. The issue with relying on your ears is expectaional bios, preference, etc. mixed into your perception. The analyzer doesn’t have these human traits and is much more sensitive than your ears.  

 

Boy does that argument get old! It's the cudgel the objectivists always bring out in the 15th round against the subjectivists. BAM. Take that! My machine slays your ears!! You're just hearing things! Problem is that what your 'analyzer' (is that really the name?) is 'analyzing' is not music. It's something else entirely. Electricity in various forms. That's the only language it knows. Music is sound waves. Music is those analogue sound waves tickling parts of the inner ear (including the incus!) and in turn lighting up parts of the brain through neural pathways. That is where music is made. And we are talking about music, no? Your machines are sensitive, great, but they are shut out from this entire realm of actual musical reproduction. You may think this pseudo-poetic claptrap but it gets to the heart of this endless debate. You mistrust ears, you mistrust brains - BUT THAT IS WHERE THE MUSIC HAPPENS.

 

And when you wield this cudgel, you shut out people like me who are trying to respond to the OP with our own experiences with the components in question. I will leave this debate now because it never goes anywhere - only with one final shot over the bow: quality of modem matters, digital cable matters, switch matters, server matters, endpoint matters - it all matters with or without optical, galvanic and/or any other kind of isolation. Those are my findings, measured at the output... of my brain...

[incus]

40 minutes ago, barrows said:

The speculation that Ethernet timing somehow effects the timing of the audio we hear is false

What more precisely makes you so sure of this? Doesn't each ethernet clock have a certain tolerance that does allow for a certain level of timing error to pass through, albeit below the threshold of the precision of that clock? Could these in turn be perceptible along the lines of what @Nenon is saying? It has been the case in my experimentation that bettering the clocks upstream of the DAC results in and of itself in an audible improvement... Separate from power supply, cabling, electrical isolation and vibration-control upgrades... just a thought...

[incus]

15 minutes ago, barrows said:

There is no explanation here as to how this would be possible.  What are you proposing is the actual mechanism for any latencies in Ethernet effecting the output of the DAC?  As long as the USB interface gets the complete data, it is then clocked out in time as per the local clock, and no latency on the Ethernet side has any effect on this.

And yet subjective listening has shown that something like this is indeed having an effect on the signal... and that changes in switches and clocks upstream of the DAC do have an affect on the output of the DAC, simply one that is not yet measurable with conventional measuring techniques. I for one enjoy the speculation of what it could be more than simply saying it can't be over and over.

 

16 minutes ago, barrows said:

Additionally, if this latency is effecting music playback sound quality, via some as yet unknown mechanism, then what would be the implications for sound quality for files streamed from the Internet (Tidal/Qobuz)?  I would guess (please correct me if I am wrong about this) that latencies from an Internet streamed file would be orders of magnitude greater than that for locally stored files considering how many Ethernet switches and the like must be involved from the path of the server to one's home.

The implications are big. This is perhaps why our digital playback chains have become so complex with the advent of streaming audio - we are trying in various ways to mitigate these effects. Local playback has its own substantial obstacles to overcome as well, by the way. There are lengthy discussions about types of memory media and SATA cables, vibration control, etc. having a very pronounced effect on SQ of local playback.

 

Also, as John Swenson has hinted at, there is perhaps a build-up of latency/timing/phase noise issues over each upstream re-clocking and yet there ALSO seems to be either a limit to their effects or a fall-off of these effects whereby they diminish with more proximal reclockings... Many have found, for instance, that a master clock sounds best when reclocking the last device - in my case, and in many, a SOtM tX-USBUltra - before the DAC... this would imply that although distant clocks do have an effect, the closer ones have a more pronounced effect. This would simultaneously explain why far-upstream clocks do seem to effect SQ and yet closer ones like the last switch, a USB purifier and of course the DAC's clock make MORE of a difference...

[incus]

14 hours ago, barrows said:

 

Oh, I am sorry, I am not referring to streaming from the Internet, that would be another topic for me-I only play locally stored files.

My god after all this time... What a cluster... The entire conversation was about streaming via Internet... oi...

[incus]

46 minutes ago, Paul R said:

My engineering experience leads me to believe that the server part of the equation, as long as it is sufficently powerful, has very little or no effect on the sound at an endpoint

But what you need here is not engineering experience - you need LISTENING experience. And until you do, this is pure conjecture contradicted by many many MANY server users/buyers/builders here on these fora and elsewhere in the computer audio/digital playback world.

 

50 minutes ago, Paul R said:

When I say far superior, I am saying a much better processor, more memory, more storage space, a much better USB endpoint in the UltraRendu. 

This may be the case but this has nothing to do with superior SOUND QUALITY, which is what buyers of the Innuos are after, after all.

 

[incus]

And by the way, streaming music directly from my SOtM server into my DAC absolutely smokes the SGC i7 I was using before even though that server's specs are far "inferior" to the SGC machine's. Not apples to apples, I know, because SOtM includes their USB card in there, which is not off the shelf, but still - until you listen, you cannot say...