The best source for describing the engineering that went into the Nucleus is the whitepaper written by Roon Labs. That makes it clear that the Nucleus uses standard Intel NUC boards. The engineering went into mechanical (the design of the case) and software engineering (the design of the OS).
1 Like
Thatās hilarious Pim! I hope your pancake breakfast worked outā¦! 
hey Anders, what type of audio components and speakers are you running - and how do you connect your Nucleus? are you using an endpoint, or connecting Nucleus directly to your DAC?
Have you observed any changes to the sound as youāve progressed through the various ROON setups?
cheers, T
CMy main rig is Meridian 818v3 + DSP8000SE speakers. It has a built-in network connection (Sooloos, which Roon supports).
It also has a USB port which I can feed directly from the Nucleus, or with a MicroRendu (with LPS1). Switching between inputs, I have not noticed a consistent difference. (The speakers are active and include DACs.)
I also have a Hugo 2 in the same room, and again, I can feed it directly from the Nucleus or through the mR, again no difference. (I even feed it from an iPad, out in the backyard ā a great portable zone.)
My theory is that with modern, well-designed DACs, and with a reasonably designed server like the Nucleus, the direct-connect-noise bugaboo isnāt really a problem. (Another theory is that I have 16-bit ears, of course.)
(I am not making any statements about noise from a big old desktop. I donāt remember the sound of my old desktop, but it was network connected and in a separate room, never tested with USB.)
In other locations I have other gear, some of it much older, but I havenāt bothered carrying them around to make the same tests, they are all network connected.
I have a Benchmark headphone rig (DAC3B + HPA4) in another room, driven by the MicroRendu ā glorious. This week I will also get another modern headphone setup. I will carry them around and do the systematic testing of direct-to-Nucleus vs. MicroRendu, just to satisfy curiosity and to contribute to the debate (Iāll write it up). Afterwards they will be permanently in different rooms.
Btw, it isnāt just a theory that a modern DAC is immune to noise and jitter, it is my expectation and demand. Why should the designer get away with a lazy design and force me to clean up the signal? And AC cleaning: itās as if the designers say, āIāve read the specs for AC, itās a steady 110 V, 60 Hz, low impedance feed. No other frequencies. What, you say there may be noise in the AC? Thatās not my fault, the spec says no other frequencies, if you deviate from that there will be noise and distortion, youāve got to clean it up.ā And USB: āthe spec says the power line is 5 V DC, no non-zero frequencies, if you feed in noise on that line, not my fault.ā A designer has to build defensively and look at the contract for any service he uses, the contract for AC doesnāt promise zero noise.
1 Like
I agree with you in theory, practice is unfortunately still somewhat different. Of 4 > $1K DACs Iāve owned (still own 3 of them in different systems): one met your USB requirements from the beginning (the cheapest one, as it turns out), one does now after an upgrade, two do not. Only reason Iāve put up with meh USB is that there are other DAC qualities I seek that are not fully aligned with the best USB receiver designs. That is, designers tend to focus on some aspects they are passionate about and do blah work on other aspects. USB audio started as a low-cost consumer PC feature, and most high-end audio designers just shrugged and bought cheap USB>I2S parts from mass producers for their DACs ([sarcasm]who in their right mind would connect a fancy DAC to a noisy PC, anyway?[/sarcasm]), while focusing on the more traditional studio digital connections (S/PDIF coax and AES/EBU). When I (re-)started in this hobby 8 years ago, even top-rank manufacturers like Naim and BelCanto had awful USB audio implementations. Itās now changing, fortunately, as you describe.
1 Like
Your Hugo DAC was designed by Rob Watts, and if it is anything like the Mojo in this review, then you are wrong in expecting it to be immune to the source.
Paul Miller shows that the Nucleus improves the s/n performance of the Mojo from 104db to 115db and reduced the jitter from 50psec to 35psec compared to a standard USB output.
Rob found that a fifty dollar (US$50) Audioquest Jitterbug made a difference with his ten thousand dollar (US$10000) state of the art Chord Daveā¦
Rob on Head-Fi forum:
"Now Dave is galvanically isolated from LF and RF noise, and this makes a huge difference in SQ. But the galvanic isolation is not perfect as there exists a 2pF coupling capacitance. Now you may think 2pF is nothing to worry about; but at 1 GHz it is nearly 80 ohms impedance, so RF noise in the GHz will couple through - and I know that GHz noise is significant in causing noise floor modulation and hence changing the sound quality.
I used my design lap-top, an MSI Intel i7 machine, as it will be noisier than my music lap-top (HP Pavilion) which is very power efficient. This is so it would be easier to hear any changes.
So adding the Audioquest Jitter Bug did indeed improve SQ - it was fairly easy to hear it, with it sounding smoother, warmer, with less sibilance and glare. Instrument separation and focus was better. These are exactly the kind of change I hear with lower RF noise, so it was not unexpected."
Hugo2 doesnāt have USB ground isolation that Chord Dave has eitherā¦
Iām not sure @AndersVinberg has the expertise to say what his Chord DAC is or isnāt immune to in the objective technical sense, only in the subjective senseā¦ i.e. what his ears can and canāt hearā¦ Anders certainly canāt be more expert than the designer of his DAC?
Then again, Iāve seen people argue they know more than the designer of their own gear, so nothing surprises me these days.
2 Likes
My post explicitly said it was subjective. I make no objective measurements, and I make no claims to know the electrical engineering of the Chord or any other gear.
But subjectively, I believe itās possible the difference between 50 ps and 35 ps jitter is not audible (for me). What about 104 vs. 115 dB SNR? I was once in a jazz club and used my iPhone to measure Ron Carter at 85 dB SPL, so maybe not (subjectively).
If you want objective, Amir wrote in his Mojo test report
I was getting SINAD numbers jumping from 99 to 105. The above snapshot randomly captured 101. I usually see some variations but not 6 dB. Regardless, using a 102 dB SINAD puts the Mojo solidly at the bottom of our tier 2 DAC performance.
He hasnāt tested the Hugo 2, but he has tested the Qutest which is like a Hugo 2 sans headphone output and batteries. It measured at 114 dB, which put it the top of his DAC measurements. And he saw no jitter whatsoever.
And I stand by my statement that designers should protect their products from cable-born noise and distortion: if Rob Wattsās $13,000 Dave is indeed improved by a $50 Jitterbug, why didnāt he include the Jitterbug circuitry in the Dave?
(That said, and Robās magical ears notwithstanding, I bought a Hugo TT 2 + M-Scaler. Very nice. Direct USB from the Nucleus.)
Itās one thing to trust an audio designerās engineering, itās another to trust his subjective listening impressions (the good thing is that the former is only partly dependent on the latter). Some audio designers have extremely bad listening rooms (in which you get completely distorted results). Or they do their listening with mediocre headphones only. What is more, some designers have really dubious test methods when they do their listeningā¦ (Iām not talking about Rob Watts. Iāve never met him and I havenāt seen his listening room etc.)
BTW, I can confirm that the Audioquest Jitterbug does make an audible difference with my Chord Dave (and my other DACs), but I wouldnāt call this difference an SQ āimprovementā (thatās completely subjective, of course). Neither would I use expressions such as āsmootherā or āwarmerā to describe this differenceā¦
4 Likes
The quote I shared above discusses the technical mechanisms the designer believes is at play (30+ years design experienceā¦) and hints at the RF and IM distortion measurements he has done. I donāt have the energy to find all of his additional posts where he shares even more info on the objective info he has but itās all there on Head-Fiā¦
Your subjective impression is as valid as anyoneās on the planet of courseā¦ but my ears would agree with Rob on this one.
One personās better is another personās worse and vice versa - this is absolutely fine of courseā¦
Iāve spoken to Rob and I can assure you his subjective impressions on this topic have objective data behind themā¦ whether he publishes everything publicly or notā¦ Rob has RF and IM distortion data (and more) to make the comments aboveā¦
Actually if you speak to the designers of your other DACs, you may find a common pattern in what Rob is sayingā¦ that lower RF results in lower IM distortion which results in a āwarmerā soundā¦ not everyone will like the result of warmer sound howeverā¦
There are lots of examples on the Chord threads on Head-Fi forum of people disagreeing with Rob on what sounds ābetterā or āworseā.
One additional quote which was easy to find:
āAs you know, RF noise creates noise floor modulation, as the intermodulation distortion from random RF noise is a white noise modulated by the wanted signal. This then results in noise floor modulation, and is very very audible. It accounts for the things sounding brighter and less smooth; additionally, when you reduce RF noise, things sound considerably warmer and darker, and one consequence of this is perception of tempo - more midrange gives the impression of a slower tempo, as individual instruments have much more body.
Now if somebody prefers the brighter sound from more noise floor modulation, then fine - thatās their taste and preference. But itās not accurate.
ā
I think is very important to remember that these are not just words on a screenā¦ there is 30+ design experience and objective data behind these wordsā¦
Just personally, Iām not interested in arguing with the designer of my own gearā¦
Just for others reading this, I also have tremendous respect for other designers like Ted Smith, the late Charles Hansen, Andreas Koch, John Westlake, Daniel Weiss, Bruno P and othersā¦
This is a fair question and one that came up over 2 years ago, when this was discussed on the Chord Dave thread on Head-Fiā¦ because itās old news (and not breaking news) Iād personally rather not go into the full discussion again, over 2 years later.
I found Robās honesty and transparency at the time very refreshing and not something you find oftenā¦
But there is a good chance your Hugo TT2 incorporates more extensive RF filtering on itās USB inputā¦ probably even Hugo2 (even though it lacks digital isolators)ā¦ both products came after Dave, so it wouldnāt (shouldnāt) take a rocket surgeon to guess these later products may do some things better than Dave, with knowledge learnt along the wayā¦
Itās the first Google resultā¦
1 Like
When Rob says āitās not accurateā, he probably doesnāt mean peopleās preference for what he calls ābrighter soundā (if he did, his conclusion āthen fineā would just be plain sarcasm). He very probably means the ābrighter soundā itself, because he thinks itās caused by RF noise/noise floor modulation etc.
As a professional musician (25+ years), my ideal has always been a kind of music reproduction that enables me to re-live what I experience on stage or in the recording studio. We all know that RF noise etc. is ALWAYS there (to a certain degree) when people use amplified instruments. The sound I get when I use devices like the Audioquest Jitterbug is NOT closer to the āoriginalā, does NOT sound more like ārealā music. Thatās why I wouldnāt call this sound more āaccurateā, Iād rather call it āsterileā or ālifelessā (but thatās entirely subjective, of course; and the differences are much smaller than such metaphors might make you think)ā¦
Iāve never even talked to any of the designers of my own gear. I donāt mean that in a derogatory way. I let the designers do the designing and I do the listeningā¦
I also have the greatest respect for people who make great products. However, sometimes even extremely successful and internationally-admired DAC designers canāt live up to their own claims, as I pointed out here. 
5 Likes
Noā¦ lower measured RF resulting in lower measured noise floor modulationā¦ The result of these two objective measures resulting in a warmer sound (subjective but also based on his 30+ years design experienceā¦).
These ideas are not exclusive to Rob Wattsā¦ as I mentioned, spend enough time talking to enough experts (if one is interested in learning) and you may find some common thinkingā¦
Talking to a few well known mixing and mastering engineers and spending time on Gearslutz forum ( a forum of recording, mixing and mastering engineers and others in pro audio), there is differing opinion on what gear best results in āmusic reproduction that enables me to re-live what I experience on stage or in the recording studioā .
Even among those involved in live performance and recording on a day to to day basis, I have seen there can be just as much debate among professionals as hobbyists, maybe on different topicsā¦
Ask 2000 other experienced musicians like yourself if a change in the source makes something ābetterā or āworseā (in the context weāve been discussing above) and I donāt believe youāll get all 2000 of them in agreementā¦ So your 25+ years experience is noted and commended but even then, your subjective opinion is just one data pointā¦ Respectfully, Iām not yet convinced that you have the ultimate set of golden ears, for determining a system of gear capable of the most accurate performance/recording reproductionā¦
To emphasise this from a different angle:
George Augspurger has been one of the worldās top studio acousticians for over 40 yearsā¦ involved in over 100 recording studio designsā¦ and even he will tell you, even with all that success, there are engineers and artists over the last 40+ years that simply donāt like the sound of his custom monitors and/or his custom designed roomsā¦
Itās important to note the obvious - that your ears are not directly wired to the DACās output (in the context of what weāve been discussing above)ā¦ you are listening to headphones and speakers/monitorsā¦ itās very possible that reducing RF throws the overall balance of the overall playback system off, to the point where you donāt like it subjectivelyā¦ Rob hints at this above.
As I mentioned, I respect your experience but it is only one datapoint - I know from personal experience there will be differing viewpoints from others that also have a lot of performance and recording experienceā¦
I talk to designers because I enjoy learning about the products I ownā¦ it genuinely interests me. Thatās just a personal thingā¦ I fully appreciate that some may not give a t0$$ about such things and thatās ok too.
I think I know who that DAC designer was - I had narrowed it down to 2 (in my mind) when I first read that post a while agoā¦
This is stating the obvious - nobody has described any night and day differences in any above discussion - but I wanted to bold this because itās probably the most important thing in this friendly discussionā¦ just in case others reading this get the wrong idea that weāre talking about night and day differencesā¦
The fact that a $50 Audioquest Jitterbug made any difference at all to Chord Dave was definitely not even remotely the reason I sold my Dave
I didnāt even buy it, it was given to me for free. (Long storyā¦)
Sean, I donāt really know how to reply to your post. Sometimes when you quote me, I have the feeling you write about things that have little to do with what I originally wrote.
Just one example:
If ānoā is what you really mean (in reply to what you quoted from my post), your first sentence says that lower measured RF resulting in lower measured noise floor modulation is what Rob Watts considers ānot accurateāā¦ But lower measured noise floor modulation is the goal heād like to achieveā¦
Then you keep stressing that ābetterā or āworseā are highly subjective concepts. But I honestly canāt believe you havenāt realized that is exactly the point I was making, too.
I know that you know I ALWAYS emphasize the subjectivity of my personal listening impressionsā¦
5 Likes
I have to be honest - I get confused by a lot of what and how you write alsoā¦
Mixed in with that, itās probably very obvious to all, that Iām not the sharpest tool in the shed 
I take blame for some of the diversion but itās probably best for the OP that we let the discussion go back to ROCKā¦
1 Like
Sounds like a good plan!
1 Like
In my experience, this really depends on your DAC. My Auralic Vega G2 seems to be more immune to this kind of problem than any other DAC Iāve owned or demoed. No audible difference (to my ears) between direct USB connection and Ethernet connection etc. ā¦
1 Like