Best linear power supply for Nucleus now?

Needed?

There are 4 answers to this, as far as I can tell

Not tried: not needed
Not tried: interested (you?)
Tried: not needed
Tried: needed (me)

2 Likes

I always use the power supply the equipment comes with. I figure the product engineers have the best insight into what’s needed. And with Roon – well, we know they aren’t shy about charging a markup, so if it was useful, they’d have shipped it, and charged any increase in price to the purchaser.

1 Like

UK brands Naim, Cyrus, Trichord (I name these from experience, but there are many more…) offer their amps/DACs in basic form but with PSU upgrades

All made a +ve difference in my experience

I’m not sure about the amps, can’t remember, but the Pulsar One DAC PSU created two separate supplies for different parts of the system (I think, it was a long time ago!) I.e. it wasn’t simply a replacement PSU, so may be a slightly different story

I have SOtM gear that comes with a wall wart and the option for a PSU. The latter makes a clear improvement that’s good VFM IMO

I think the designers have technical and commercial objectives; keep the price right to get the sale, offer the upgrade for further revenue - and performance - if you think it’s worthwhile

This holds true in any range of product; cars, white goods, food etc etc

1 Like

This assumes one is using just as a PC…when you start adding audiophile type DACs to the PC connections then such things like power supply matter. but in the end trust your ears as to if you have a preference one way or the other.

1 Like

Any product engineer worth his place will design kit with a ‘good’ power supply. And any accountant worth his place will remove it altogether or make it a high cost upgrade option. Unfortunately product engineers don’t have the final say about what goes to market.

2 Likes

Yes, very true. But the Nucleus isn’t the ordinary kind of product, and Roon Labs isn’t an ordinary kind of company. The box is carrying a significant markup, and is clearly designed for installers who are doing a price-no-object audio room for customers who are too rich to care (as is the lifetime subscription). So the normal considerations of price elasticity in the market wouldn’t seem to apply.

On the other hand, Roon Labs does clearly indicate a preference for network-connected endpoints. I think there are several reasons for that, not necessarily including this pervasive (and to my mind somewhat superstitious) worry about electrical interference travelling on the wires. But we just had that argument.

@richard_a, why don’t you search for this topic on the forum?

I wonder how many Nucleus (Nucleii?) are bought via installers?
Mine wasn’t
I’ve built many PCs and I fully understood the premium I was paying

Roon is a for profit company, no problem with that
All for profit co’s assess costs, ROI, target market, volume, growth, competition, their USP, etc and consider how much money they need to break even and how much they want to profit - and price accordingly. If they get that wrong, they reprice or fail

Roon software is expensive, but its clearly attractive enough to sell
I certainly thought long and hard before buying - but no regrets here
Nucleus is expensive, I did use an existing PC to trial it, I helped my friend build a NUC, I still bought one - still no regrets here

1 Like

Yes, that would be interesting to know.

I think the product was intended for installers, though. Nice packaged box you could put in a system. Sometimes these things break out into the wider consumer market, though.

The white paper is interesting

Covers a lot of points raised in these dialogues

Nucleus wasn’t designed for installers. It was for people who wanted an independent device for Roon but did not have the inclination or time to build their own servers. And Roon are on record saying that unless my powers of recall are failing me.

1 Like

Indeed. The Nucleus web page explicitly markets it this way:

Nucleus is the easiest, fastest way to get your Roon system running. No computer or networking skills required. Just take it home, plug it in, and download Roon apps for your mobile devices and computers. Nucleus keeps itself current with automatic updates.

1 Like

Yes, I know that’s what the customer-facing marketing page says.

It isn’t just what the page says. It is also what Danny said at the time. They were built because there was a demand from end users, not intermediaries like installers.

Moved to the existing topic, lots to read and catch up on.

On a side note, there’s often very detailed discussions on the forum so it’s worth searching to do some initial research.

1 Like

Yes, that’s actually pretty common with DACs to have individual supplies for various functions within the DAC. My latest design has supplies for the receiver chip, and another supply for the asynchronous sample rate converter, and a special low noise supply for its clock, then supplies for the DAC chip itself. This is typically for the digital side and also the analog side. Then supplies for any analog filter and buffering circuits (I use a transformer output so I don’t have output stage supplies).

Whet I’ve seen in external supplies for DACs is one or two raw supplies that are then regulated down on the board with the regulators and filters close to the circuits that need the power.

The Trichord Pulsar One was quite a thing, in its day

From a layman’s perspective, I found, building my own pair of JPlay connected PCs, power supplies - and shielding in particular - made a huge improvement to sound. I mean :open_mouth: differences

I ended up with 7 LPS, to mobo, SSD, CPU etc in each

Also did a lot of sorbothane damping and solid brass passive cooling

Did it work?

I had a Mark Levinson 31.5 Reference transport and 360s DAC - l sold them both

I think it’s coming back, but not for consumers any time soon (if ever). Seemingly great for advancing neural networks and advanced data streams — and could be sweet for audio but too costly to research I’m betting.

This is where I’m at mentally with my NUC. Since it’s only a computer feeding my endpoints via Ethernet or WiFi I can’t see why a power supply would change the performance. Assumed the original power supply is well constructed in the first place. I also have challenged this with a quality UpTone power supply. The result was no perceived change of the sound quality produced (at the end of the chain).

1 Like

At least you tried - and found no appreciable difference, which is commendable IMO :+1:

Why bother…?
Centuries ago, Ivor, the guy who started LINN, used to sell the idea that the source is all important ie. if you can’t get the music off the LP, the rest of the system can’t create it. It’s true, but you also need decent kit downstream…

Following that logic, ensuring your PC is as effective as possible, so it sends a ‘clean’ signal, all helps - less work for downstream components means less degradation to the sound

What does that sound like? In my system, it’s all about soundstage depth, width and the outline/realism of instruments and voices (yes, audio w@nk word bingo Full House! coming up :nerd_face:)

Is it night and day? No, of course not. But when you start pushing up the increasingly steep slope of diminishing returns, everything matters - since we are all here, we are all a fair way up that slope…

Something I’ve learned with these things is that A/B comparisons are of limited value. My best evaluations come from at least a week of listening post-change. It’s when you swap back to the original part that any differences are often glaringly apparent

Regards
Andy

3 Likes

I agree. AB listening is not helpful. Listen for a decent period and then try to go back. Then you know.
The Brain compensates for a lot to make things sound correct but it’s all work. The cleaner the signal, the less work your brain has to do, the less fatiguing things become and more satisfying.
Dare I suggest, MQA works this way for me. (I’m not trying to start an MQA fight here)

1 Like

I’m not sure that this is a thing, your brain has no concept of the “correct signal” so doesn’t work to clean up the “dirty” version. Whatever psycho-acoustics go on there’s no tangible concept of a known target that your brain is working to meet, it just does its thing.