Mmm, no, that is not a correct assumption. I use a Roon Nucleus primarily for ease and convenience and “assuredness” of adequacy for the task of handling a library of about 40k tracks. I do not believe the Nucleus+ improved the SQ over my previous Roon deployments on a Mac mini, an iMac, a NUC, and more than a couple of NAS devices. The SQ, as opposed to performance reliability—my early NAS deployments were on underpowered devices—has been steady. The reliability and simplicity of the Nucleus are quite distinct from the SQ.
But I am unclear on how this is responsive to my previous post. If one’s DAC and Roon Core are on a network, and one is not using Roon DSP, what would be the explanation for a change in SQ for bit-perfect data between 1.7 and 1.8? I’m not saying that you are not hearing it. In what way is the SQ better, and why should that be the case? Note: implicit in the Roon statement is that non-DSP-Roon over network should sound no different.
If it’s connected to your DAC over USB or other non-network connection, I could see that making a difference. If it’s connected via a proper network, how will that make a difference?
No DSP, no USB but ethernet. I cannot explain why I have better SQ now. More air is what I am hearing. And no expectation bias, because I was not expecting any SQ improvement.
Well, that’s not exactly the definition of expectation bias, and you cannot eliminate the cognitive awareness that you are using an updated piece of software, but no matter. It should still cause you to ask “why should that be?”. And where would “more air” come from? It’s not a timing/jitter improvement, and no musical information has changed. The DAC performing the digital to analog conversion is the same, and is receiving the exact same data the same way as before. SNR is almost certainly the same, and your computer and other network devices are likely introducing the same insignificant quantity of parasitic noise into your Ethernet as before.
Because of the way Roon works, it’s very difficult to go back to conduct any sort of reasonable comparison, to test whether one was listening to 1.7 the same way that one is right now attuned to listening to 1.8, to switch between the two without knowing which is which. Someone on the Roon forum proposed a phase-reversal-null test. That’s a great idea, but I suspect it wouldn’t change any minds.
The answer to the question in this hobby—“Do I really hear that, or do I only think I hear it?”—seems to always be “yes.”
It’s like thinking that a package of refined sugar tastes sweeter when delivered by a nice person driving a Prius rather than an old beaten up gas guzzler. It definitely may be a more pleasant experience, but it’s the same sugar.
On my NAS, running Roon 1.8, the CPU load is 2%, RAM load is 60%. It is a basic Qnap NAS, I also run Minim server 2 on this NAS.
If there is a change in the SQ I don’t think it is related to CPU load…
No, I did it now because I read this discussion about potentiel change in SQ, but otherwise, when listening I cannot notice any change in SQ, but I can imagine that if you run server, core, player on the same computer, may be the SQ is negatively impacted.
Yeah, but that bit from Andrew proves nothing, except that he wants you to enjoy whatever you are hearing. And no, it wasn’t science that proved the earth was flat; that was a religion and superstition thing.
But I love that! That’s perfect. You enjoy it Erno!
If Roon outputs bitperfect, and over ethernet a DAC receives these music data bitperfect, and converts these bitperfect data to an analogue signal, then any DAC would be perfect. We know about a forum that promotes this…
Yes we do, but I’m not saying that. In between that extreme position, and audiophoolish magical thinking, there is a reasonable middle ground for understanding real world phenomena and psychoacoustics. If all we changed was some software code that didn’t alter the bit perfect delivery, and there are no other changes in the system, then there is not much reason for believing SQ has changed. Audiophiles are amazingly susceptible to the “I can’t explain it, but I can hear it” mantra (which is a valid statement); but accepting that mantra as proof of something in the absence of a reasonable engineering hypothesis is not science. Since last Sunday, the only thing that has changed in my system is Roon 1.8. If someone tells me that the same is true of their system, but the SQ is better, then I think it is a fair question to ask why. The Release Notes of 1.8 offer no explanation for why a network-based system should sound different. The Roon tram offers neither a claim that it should nor a reason that it should. So, what would be the hypothesis to explain it?
I agree with Andrew: Enjoy! But if one cares about understanding, as I do, no one has offered an explanation of how SQ could change other than psychoacoustic. What’s interesting is how much intelligent people resist the suggestion that some cognitive bias is the far more likely culprit. It is significantly more powerful than a few CPU clock cycles operating below a -115db noise floor.
One of the beauties of the human mind is that it can understand that it doesn’t know everything, and that the pursuit of more knowledge is an infinite quest. I accept that there are things about the human perception of and reaction to music are not 100% understood. But the incompleteness of that understanding and the infinitude of that quest do not require us to adopt fantasy as a framework.
Information is not knowledge.
Knowledge is not wisdom.
Wisdom is not truth.
Truth is not beauty.
Beauty is not love.
Love is not music.
Music is THE BEST.
Ordinarily, Roon will tell you almost immediately when an update is available to be installed. This time for me, I had to restart the core and at least one client to get Roon to recognize the update’s availability. All good so far.
