Nuc sound differences

And this gives a “better” digital stream ?

Computations that lead to a bit perfect stream that are altered by using a different powersource/chipset in a computer?
All this independent of programs or algorithms processing the data and sending it out on the network.

1 Like

That would be the end of computer science and networking as we know it…

4 Likes

Ok Peter…Ryzen X300 on a LPS…with fairly small standard…connection cord. Sounded good everything working…bit perfect no. Added a thicker braided psu cable to pc…and it was omg better… How can this be explained… computer is supposed to be bitperfect? or is it dac sensitivity I don’t know. Which leads us to why do LPS improve sound quality…it does…but you may not believe it which is fine…and I am perfectly happy with that. I wish I was… I am really inerested in your findings and what works well…

Please let us keep the discussion within this thread civil and respectful.
Thank you for your patience and understanding

Sorry…I mean absolutly no disrespect to anyone…appologies if any of my posts have been taken that way…

I build Roon servers for a living and I have found quite a few things that can improve the sound.

We use our own mother boards for the sonicTransporter so I have not done listening tests with standard NUC board. I know other have and there are differences in sound. I looks like other Roon community member in this tread are working on finding good ones.

These things will make an improvement with any board.

  1. remove the FAN. There are lots of fanless NUC cases available.
  2. use a linear power supply to power your NUC.
  3. If you have a USB DAC use a good USB cable, a USB filter, or even better and network player. The noise from the power supplies in the NUC can travel down the USB cable and get into the analog side of the DAC. If you can eliminate this sound will improve.
2 Likes

Re amd Vs intel chip architecture it is noted the USB interface to the core is much tighter/closer on amd than intel. This may explain the difference in sound. Note the comparison was made on high end gear including holo may dac / gustard x26 Pro, t+a pre amp, avm mono blocks, gauder akustik darc 200 speakers with rel subwoofer array.

The processors were mounted in streacom fc5 fanless cases.

Spot on. But hey, this thread will weirdly drift away from any logic anyway. It’s like a law of nature :rofl:

Regarding the ones and zeros, in my computer, a Z440 server, the ones are more onezie and the zeros are more zeroy. Yes my zeros have perfectly round curves on them and the ones are the most straight up and down of all ones anyone has ever seen in a computer. I think this is why my DAC, (like almost any modern DAC) which has specs well beyond any human discernible difference, sounds the best. But seriously, if you hear a diffrence that is not a cooling fan or the diffrence in speakers and room setup you could have a defective amp or DAC. Computers just make ones and zeros. Amps also have specs beyond any human discernible difference. It is rooms and speakers you hear and if you have adjusted the bass and treble knobs.

4 Likes

Cool thankyou :slight_smile:

As other replies have indicated, a pcm stream is a digital data stream, not an audio signal. With plenty of encouragement from the manufacturers of streamers (which are just general computers), audiophiles have bought in to the idea that computers are audio devices. As a claim to hear the difference between 2 computers is a clearly falsifiable conjecture, one would imagine that the web would be awash with the results of proper blind tests showing this. You can take the total absence of such evidence to mean that no one can hear a difference, what ever the number of unsupported anecdotes to the contrary. Of course we don’t need to do such tests anyway. as an an analyser can show that there is nothing to be heard.

Here’s a good example of a serious investigation of the issue:

3 Likes

Modern DACs can handle any noise that is likely to be thrown at them over a usb connection, not that noise artefacts from even the least suitable pc (such as a gaming i7 running on full CPU load) would be audible anyway.

The points you make are easily testable:

Noise Graph Composite LOG - USB - Coax - TosLink-i7

To be fair, I can see that you build some very nice computers, but they are computers, not audio devices, and as such have no impact on the resultant audio signal whatsoever.

5 Likes

That is correct we also sell Sonore Roon Ready player/streamers. These are audio devices and make quite a difference in the sound of a USB DAC.

Thanks for your reply. I’m curious though.

“The noise from the power supplies in the NUC can travel down the USB cable and get into the analog side of the DAC”.

Do you have evidence of this? The graph above shows that at no point does electrical noise reach audible levels (remaining 25db lower than best case hearing thresholds) on either a 10 year old i7 board at full CPU load or and RPI, both over a long, generic usb cable lacking any special noise mitigation. I’d also be keen to know which DACs you believe are failing as described above, as they should be avoided (not that I can see a reason to worry if no noise is going in in the first place, but belts and braces and all that😀). My RME and Topping DACs are proven to be audibly transparent on a state of the art ap analyser; its possible now to get instrument grade DACs for less than £100 in the UK). I realise that there are a few boutique DACs out that take a state of the art akm chip and add loads of noise, but they are quite rare thankfully and usually so expensive that people can’t afford the really bad ones.

I guess what I’m saying is that if you are going to recommend things like linear power supplies etc to deal with noise, you ought to provide some measurements showing the benefit as all of the published evidence says that they aren’t needed, or at least a credible technical argument based on sound engineering principles.

Wow people are very precious over their ones and zeros…all good stuff. Nice to see a debate.

‘Or even better a network player’. You mean quality improves if you insert a network player into the chain, with any USB cable? Please explain. Thx!

1 Like

I have found through personal experience and testing this guy to be right…GTT Audio - Episode 35 - How to Maximize Roon Performance. - YouTube except that I have heard improvements with Ryzen and X300 as well.

Only to the suggestible.

4 Likes

There my friend we will have to agree to disagree…I am no longer responding to this thread…

Computers, unless they’re misbehaving for some reason, are bit perfect, so this can be explained as a belief rather than fact. You believe that things like braided cables can make a difference, so it’s no surprise that you do hear a difference when you switch one component for another. As has been mentioned in countless other threads, auditory memory is quite poor, so unless you’re doing fairly rapid AB or ABX testing, any perceived differences should be taken with a pinch of salt.

What seems to be missing in a lot of these discussions is anyone saying, ok, I guess I could be mistaken, but they don’t say it because they’re privileging their own experience over and above commonly believed facts. Why, when the evidence points to the much simpler explanation that they’re just mistaken?

When I was in my early 20s a friend of mine died unexpectedly. For ages afterwards I kept thinking I could hear his voice, particularly in crowded rooms. I guess I could have assumed he was there, or maybe talking to me from the ‘other side’, but I didn’t do either - I just accepted that I missed him and that my brain was playing tricks on me as a result. Human perception is frail, flawed, and prone to error.

9 Likes