Roon sound quality related to drift?

Hi,

I have 2 Roon server installations: one on Synology NAS and one one a ROCK device. The playback chain is Roonserver -> Ethernet CAT6 1Gbps network wired -> SoTM SMS-200 Ultra -> Mytek Brooklyn DAC+ -> Amplifier -> Speakers.

I find that when playing from the Synology Roon server there is much better sound quality, cleaner, more depth, smoother highs than when playing from the ROCK server.

I am puzzled as to why this sounds better via NAS than ROCK. The ROCK device has much better hardware (core i3) while the NAS is a Intel Atom device (DS1813+).

I have had a look at the log files in both machines. I found that the drift differs:

Synology:

Trace: [sMS-200 living] [zoneplayer/raat] sync SOtM sMS-200 (): realtime=2309989355751 rtt=500us offset=1293278355us delta=17us drift=-9086us in 879.1145s (-10.336ppm, -37.208ms/hr)

Rock:

Trace: [sms-200] [zoneplayer/raat] sync SOtM sMS-200 (): realtime=101336024969793 rtt=500us offset=37670983969us delta=135us drift=-5885us in 152.191s (-38.675ppm, -139.228ms/hr)

Could this be an explanation why the sound quality differs and if yes, what can be done to improve it?

These drift numbers are used by the core when multiple zones are grouped together as part of a method to keep different devices in sync with each other. For single-zone playback (as you’re using it) they don’t have any meaning.

It’s not uncommon for the sync messages (the first ones logged during a playback session) to show what appears to be a lot of drift. That has to do with how this part of RAAT works. There’s no concern there.

I’m not sure why you’re perceiving a difference in sound quality as there should be no difference from core to core, regardless of hardware. I tend to run multiple cores and have never experienced a performance difference between them even when very different hardware was in use (Synology vs ROCK vs Nucleus+)

2 Likes

Strange. The difference is not subtle to my ears.
Could the power supply be of any influence? With the Rock the PS is built in (Rock installed on PC hardware) while the syno has an external PS.

What core do you perceive as better sounding?
I also find the sonic difference between different cores is not subtle

Andrew, please let me tell you something off topic here:
dCS is not my cup of tea sonically but you are an absolute top bloke.
Always extremely helpful (also in non dCS matters) and VERY knowledgable. Thank you very much for that :+1:
I just wanted to tell you already some time ago but never got around to.
So I thought it’s time now :sunglasses:

3 Likes

In my case I like the core on Synology much better than the core on ROCK.

It shouldn’t be a surprise that a device that is specifically designed to serve up media is quieter than a NUC or general PC designed for desktop or industrial use.

Has anyone had a similar experience i.e. Synology core sounding better than ROCK (or maybe also NUC)?

Nope. I’ve tried Windows, ROCK and NAS and they are indistinguishable, as they should be IMHO.

2 Likes

Thank you @Christoph, that made my day!

2 Likes

Honor to whom honor is due :+1:

I used to use Synology, then moved on to a NUC. The performance of the NUC is much, much better but the sound quality is the same and my sound system isn’t exactly forgiving. Two things that I did do though was make it fan-less and get a decent power supply, and I think that especially the latter may make a difference.

Perhaps the difference is due to the location of the data (i.e. music stored on the Synology NAS). The location of the NUC in your network might involve restrictions. But I’m no network specialist, maybe someone could elaborate.

It’s some time ago now, but I felt that sound quality improved when I switched from an smb mounted music folder (NAS) to a direct (USB) connected HD on my core. Core is Debian Linux, not ROCK. I know this should make no difference, but that is what I perceived at the time.

Good suggestion. I’ll give that a try.

Hi Benny,
I share your experience- SQ in my room is clearly better when rendered via Synology than with NUC. I too am scratching my head as to why? And believe me, I really wanted to embrace the NUC which deserves high regard, despite this. If you learn anything insightful, please share. (I am running Synology 718+, Qobuz, Roon, dCS Vivaldi, Ayre 20-series, Rockport loudspeakers- brief summary- hardly a “cost no object” system but resolving enough to make such comparisons as these.)

In the meantime I purchased a “real” NUC. Previously I used Roon Rock on a standard PC with builtin powersupply. Suggestions were that a better, external power supply could improve things.

The new NUC is model NUC7i3BNH. I also purchased an audiophile power supply, the SBOOSTER BOTW ECO 18, 18,5, 19V.
Now this would potentially produce the best sound quality as a Roon server?

Unfortunately, no improvement noticeable. The Synology NAS still is the clear winner in terms of clarity, instrument separation, deep tuneful bass, microdetail and ambient sound.

I take it your music was still on the Nas when using the NuC and ROCK? If so then I would try music directly attached to NuC. Using a Nas and NuC you having two network hops before it gets to the endpoint as ROCK will pull it from NAS and then push to endpoint. When just using NAS you have one hop to endpoint.

I noticed an improvement with local storage on my ROCK server.

I forgot to mention that I also put the music on both an internal and external SSD, but I can not hear an improvement over NAS retrieved files.

And most of my test was done with music streaming from Tidal, so that takes
the location of the files out of the equasion.