So I started with the Nucleus and Sbooster LPS going USB into my DAC. I made several statements on this site that I was satisfied with the setup and even asked the question, are streamers necessary when you can go straight out of the Nucleus. Everything was fine until (does this sound familiar to anyone?) I wondered if the sound would be better if I plugged in an Audioquest Jitterbug and it was. Then while doing some reading about Nucleus I learned that it has no galvanic isolation, reclocking, error correction, femto clock, FPGA, etc. (if I’m wrong here someone please correct me). So I bought a USB galvanic isolator to see if that would improve things. Well it never worked because IMHO the “isolator” prevented Roon from seeing my DAC and wouldn’t play anything. My solution was to put the Nucleus into the LAN only and buy a great streamer only (Lumin U1) to use with my DAC and I am currently testing the USB Isolator to see if it has a benefit. The Lumin U1 has its own LPS and some very good crystal oscillators and it sounds much better (IMO) than the Nucleus.
Also Roon recommends using Nucleus only in the LAN. If you are happy with the Nucleus USB to DAC so be it, just don’t try to get much more.
I have ROON on ROCK on Intel NUC. I used to connect NUC to DAC over USB, thought SQ would be better with the direct connection, I did believe that for a while until I saw the KB. My DAC has Ethernet interface it did not cost extra for a try. I keep this topology until now. If it does not cost you extra for a try, you should.
When going Nucleus USB to DAC I heard a sonic benefit using the Sbooster linear power supply. Quieter, less strained, blacker background. Nucleus comes with a switch mode power supply ie, wall wart. I used the wall wart when I quit using the USB on my Nucleus.
My point above was that anything you can do to create a quieter signal before the DAC is still beneficial.
I’d argue that if an UltraRendu inserted between your Innous and your high end DAC improves the sound, then it is addressing a deficiency in the Innous’ or the DAC’s USB implementation (or both). I.e. something to do with jitter, electrical noise isolation, etc. And there lies the rub. People talk about USB vs. Ethernet but both can be very good, and the relative sonic superiority depends on the details specific products, not the technology.
Not sure the need for inserting UltraRendu between a nucleus and a tt2 chord dac connected via usb. Seems it’s more of a streamer, which may be fine on its own. Adding so many layers for signal to travel thru at some point would seem to weigh things down.
40 years of trying to make digital sound as good as it is supposed to do, and still we get “bits is bits” arguments against every possible improvement. The only good advice is to try out various options and see what works for you in your setup.
I ended up with Roon on iMac, via ethernet to a dCS Network Bridge, then to the Chord DAC. That sounded best to me after several years of trying other combos – and the difference was worth paying for. But every situation has its own considerations, not least your own ears and what you listen for in music, so an audition is the best route.
On balance, having read many posts on this topic, I would say most come down in favour of Ethernet making a difference and providing a better source than USB. If the wisdom of the crowd can be believed. And that makes auditioning the ethernet solution at least worth the investment in time and energy.
I have a Sonore UltraRendu acting as an endpoint with the Roon core on my iMac. It works very well and sounds great versus other set ups I’ve had in the past, or at least it did and then I had a few problems with the connection to my Devialet D Premier via a HiFace USB/Coax adapter (which I think was the problem) but I managed to rectify this eventually. In the meantime I decided to buy a separate DAC (Chord Qutest) and use the Devialet only as an amplifier. The DAC in the Devialet is good, if a little old now but the difference is sound between it and the Chord is amazing; usually for me, any change in kit requires some hard listening to reveal the differences (if there are any at all) or else I resort to confirmation bias to prevent post purchase dissonance. Anyway, I wouldn’t say one is better than the other but the Chord is an easier listen, less dynamic and precise and preferable with certain types of music. The Devialet is more detailed and arresting so when I want this sound I switch DACs. The point is that sometimes you find big differences when you least expect them.
Bill_Janssen
(Wigwam wool socks now on asymmetrical isolation feet!)
49
The thing is, that argument gets stronger with every passing year. Gear designers who only knew analog learn about digital, and start doing the right things in their products. Good solutions to common problems are disseminated through the literature and packaged into standard products like chips. Suppliers start making parts that eliminate various sources of interference. Engineers start using those parts in their designs.
Human ingenuity, combined with the ability to “write stuff down”, leads to astonishing decade-over-decade improvements in almost every field of engineering and science. And audio engineering is no exception.
Of course isn’t it tru that new products introduced into the complex chain of sound signal flow need a breakin period? Maybe at least several weeks. Speakers need to be broken in and then a new electrical device like a dac needs a breakin period.
Could it be that we may be over reacting in passing judgements while not giving time to a reasonable breakin period?
I thought the DEV resampled all incoming Analog and didn’t have an Analog pass-through. Or is that only on the later models?
Bill_Janssen
(Wigwam wool socks now on asymmetrical isolation feet!)
52
Well, speakers are a physical device with moving parts. In general, the whole idea of break-in period is another myth, with respect to modern audio electronics.
In support of @Robert_Westcott’s post, and @danny’s original response, I’d just like to mention that in a prior thread, DCS has recommended Ethernet over USB to their DACs/equipment.
E.g., Roon to Ethernet, Ethernet to DAC, done.
I understand this is only an option if you have an Ethernet-enabled DAC.
I’m suspicious that over a certain line in the sand, when we say something is better or improved as a result of a new bit of “arms race” equipment, we really mean it’s just different. Im discovering after getting Roon, then joining this forum and getting my first high end system, that my line in the sand is moving like a little itch I just can’t reach - for years I had iTunes and Spotify and thank god im off them now but oh man upgrade to a Statement, or dsc network bridge, Vini Rossi amp or that 7000 buck Crystal Ethernet cable… I use to listen to music only while travelling, now I insist on a daily dose at home in my new listening room of Mahler 6th or whatever at full blast
I blame Roon and this forum for turning me into a music audiophile addict haha
H
Yes, galvanic separation is good. However, clock and noise over USB are still to be considered.
I am not familiar with what jitterbug does exactly. However, if it does noise filtering (for the electrical voltages of 0s and 1s be less influential by electrical noise) then potentially your DAC receives better quality 0s and 1s and this will have to work less harder to recognize 0s and 1s by electrical signals and thus better SQ potentially. A quality USB cable with good shielding is also worth considering.
I do not know enough to comment on the level of SQ influence with clocking over USB, however, I can say that anything related to clock has direct influence to SQ due to jitter.
Welcome to the club if this is the level of SQ that you are looking. Otherwise, just relax and enjoy the simplicity and already high SQ over USB.
This is not how electrical noise over the cables works. If the noise changes the bits, you will hear a click or pop, not fidelity loss. The only way electrical noise can impact digital audio is in the analog stages of the signal path (the DAC and beyond…)
I agree. Perhaps I did not make clear earlier. I am referring to noise that interferes how digital signals are received and processed at the receiving interface of DACs before the analog domain. Digital signals are transported as electrical voltages over copper and are not noise immutable.
I wish we could move on from trying to explain digital transports in analog terms. They’re two different things. There are a number of mechanisms to ensure the 0’s and 1’s arrive intact, and they do.
Anyway, I wouldn’t say one is better than the other but the Chord is an easier listen, less dynamic and precise and preferable with certain types of music. The Devialet is more detailed and arresting so when I want this sound I switch DACs. The point is that sometimes you find big differences when you least expect them.
I use to listen to music only while travelling, now I insist on a daily dose at home in my new listening room of Mahler 6th or whatever at full blast
