NAS us connected to router, PC and DAC are connected to router via switch. All LAN, no WiFi.
Hi NOA,
I through I got what you mean, as Iām also in similar progress.
I start from MinimServer on NAS, JPlayStreamer on a fanless NUC with LPS (Win10), connect to Meitner ma-1 DAC through USB Regen.
Install Roon server on the NUC and very happy with the SQ and UX. Then install Roon server to my DS713+ (with DX513), and stream to my NUC. In my case, I notice virtually no change of SQ. And last night I uninstall the Roon server from my NUC, and install Roon bridge, which is good enough as a output endpoint. The SQ seems the same so far, will do more serious listening later.
I will live with this meanwhile, and also interested to replace my NUC with a endpoint device in future hoping to simplify the setup and hopefully will improve SQ as bonus.
FYI, I have 2 JitterBug on my NUC, and a Acoustic Revive LAN isolator at the LAN input, which could contribute to the result.
If moving to a linux endpoint, I also worry about the performance diff between the DAC driver, where Iām using Windows ASIO driver provided by my DAC vendor now in my win10 NUC.
So in your case is how we setup the ācoreā, and my concern is how we setup the āoutputā. Both could contribute to the resultant SQ.
More experience sharing are mostly welcome.
Given that you are using totaldac, I remember they recommend and sell SSD fanless QNAP NAS. Also seen user report that NAS with more spindle motor on RAID config could deteriorate SQ as there is more noise. So on my 2 bay DS713+, I use only a single disk without RAID to store music file.
you really need to read Dannyās post referenced above.
It made a lot of improvement in SQ if your NAS is powered by a linear power supply instead the one that come with the NAS. Iām using Synology DS-214SE(2x3TB) powered by KECES DC-116 power supply. Because it has two output supplies, one is used to powered the NAS and the other, the NUC which runs the RoonServer.
www.keces.com.tw/5_english/index_5_english.htm
They saw you comingā¦running a linear PSU on the NUC I can understand, doing so on the NAS makes no sense at all unless youāre connecting directly to a DAC.
Yes, it does make a different running off the NAS on a linear supply. Thereās more clarity and background is more ādarkerā than before.
To you, yesā¦I guess thatās all that matters.
I was also going to say it is a scientific impossibility for the NAS PSU to make any difference as well, but as we all know science has nothing to do with music reproduction.
And the Bits it sends over his ethernet line are more sharpened.
Sorry, I canāt hold this back, but that is total voodoo.
I cannot understand why each little thing has a big influence on a digital signal, if this signal tansports sound. On every other data you never hear such arguments. Or did a linear supply prevents your NAS to send your Wordfile with a lot of typos?
Bits are Bits and a normal functioning NAS is capable to send them without any change to the receiving device without any overpriced extras.
I thought the current thinking was that ānoiseā in general within your network - be it from switch mode power supplies on NAS, router/switch, mains, computer RFI, etc, etc, were all believed to be detrimental to SQ if they make it into a DAC that may not be able to deal with removing 100% of it? Same for jitter. And this is the reason something like a Ā£10K+ server like an Aurender W20 or a Melco N1* or whatever allegedly sounds far superior to a bog standard computer? Or you go down the NAA route to try and avoid both things by making a ācleanā signal as close to the DAC as possible.
Or have I been brainwashed too and weāre actually saying a standard PC and standard USB cable and power supply, or a NAS or whatever all sound the same and can not be bettered - any change is imaginary? I used to think the latter, but more and more Iām leaning towards the former, and occasionally have a āhave I been brainwashedā moment.
It seems like thereās a hell of an industry built completely around expectation bias etc if its all totally untrue. I really wish one day weād actually reach consensus on itā¦.
Interesting! My NAS has an internal power supply. How did you bypass this?
When you are talking about file transport (file copy, file move) I 100% agree. The problem is that streaming files is different from file transport.
Yes Iām actually only talking about streaming from the core, not the rendered output.
Sorry, but no. Streaming files IS file transport. Nothing else.
After the DAC when the signal is analog there can be measurable differences in signal quality depending on the equipment (the discussion if audible or only measureable I wil not start). But not on the digital transport way.
Look at all the tiny (smaller than a hair) connections inside your NAS and look at the ten thousands of kilometers of different cables (and thousends of switches/routers inbetween) when I stream something other internet from TIDAL. And then I should believe the power supply of my makes a difference?
Sorry, but no. Thatās against every scientific education. But the believe in such things brings some companies a lot of money. That I believe. And as long as people has the hope some snake oil will improve their live, there will be a snake oil dealer.
Or do you really believe a thousand euro/dollar USB-cable will improve anything other then the purse of the company who sells it? The audio scene is deeply infested by voodoo and snake oil dealers.
2 Likes
Itās turtles all the way down, isnāt it?
1 Like
To verify my point, you can make an experiment:
Start playing some music from your NAS and while playing remove the cable from your NAS.
If your endpoint is not the cheapest one, you will see, that the music plays along for a while (on my system ~ 1 minute).
And your NAS is not longer connected to the system. So it can have no longer influence to the sound.
Did the SQ increase?
No? So if it has any overall influence, it must have already altered the one ore two megabyte of data the endpoint has in his memorycache. And that are pure digital data, bitperfect from the source.
As every today data transmission is asynchron and cached in a memory at the endpoint before beeing proceeded, there canāt be such things like jitter or noise on a digital signal. Only succesful or corrupted transmissions.
4 Likes
So, if you look for improvements, donāt waste your money on the digital side of music.
Better invest it in better loudspeakers or in increasing the overall experience in music listening with something like Roon 
2 Likes
Mine is a 2 Bay DS-214SE Synology running on 12VDC plug in. It comes with brick wall switching power supply. I just replaced it with the linear ones.