Servers

[Mr Underhill]

Question

I am interested in whether Linn DS users have found much difference in the quality of their systems based on the server software and hardware platforms they have used.

History

I have just bought a Linn Klimax DS/1 renew.

Having built and played with a number of systems I ended up buying a Naim NS01 which I used as a big DDC, this was before Naim started producing their 'streamers'. This gave me a good service for a decade before it died. I sent it back to Naim and bought a Sonore Microrendu in the meanwhile, this opened my eyes to what was possible. I then spent time and effort honing my digital front end and got some excellent results for not too much money, in HiFi terms. My last pre-Linn front end included:

Audiolinux + NUC backend (LMS), powered by LPSUs via LT3045s >> Audiolinux + NUC backend (Squeezelite), powered by LPSUs via LT3045s >> SingXer F1 >> Audionote 2.1 DAC

NUC = Next Unit Computing. Small Intel boxes - not all of them work well in this context.

The issue was that I had a system that was spread over three shelves (just the front end), had two 6A LPSUs running 24/7 and a spaghetti of cables.

The DS reduced all this to one shelf.

Sound Quality
The Klimax is more detailed. I suspect the AN is slightly richer, not hugely but rather subtly and with respect to vocals especially. But, BOTH setups give me sonic thrills; if in rather different places.

An issue with the DS/1 Renew is that it is a technological dead end, but would I spend the money needed to do an update if I had spent twice as much for a full 2nd hand Klimax? I think not.

I would have been equally happy with the dCS NB I think, but that would have cost me an additional £500, second hand; and force me to buy into Apple.

It is interesting to listen to familiar tracks and both pick up on added details, but also find other things are hidden deeper in the mix than I think they were.

Sound Optimisation

Sound Optimisation is a nice trick. The effect is noticeable, and the sound staging is very precise. With studio recordings the music sticks rather closer to the speakers than with the NUC/AN, but with natural acoustic material it is both precise and well spaced.

I have played with v1 & v2, having read through the SO Thread. I bought the Renew from an ex-dealer who kindly bought it over and did the v1 settings, which I saved and then used as a base from which to tweak. I will read Paul's thread and then dig deeper.

Last night I logged onto the Linn site and set up a few v2 profiles, which i think I prefer at my beginner/superficial level ......BUT, this is a long journey which I am sure will take time and effort.

Servers

With Audiolinux/NUC I found that I got a real step up with different server software, and where files are hosted. Currently I have three sources:

1. Qobuz;
   
2. Synology NAS running Minimserver; and
   
3. Audiostore Prestige 2 (think NUC) running Audiolinux, files on 4 x 512GB USB sticks.
   

Before going into the variation I have tried I have found the Linn to be REMARKABLY unresponsive to server and file changes, all rather small beer.

Minimserver: NAS vs NUC/AL = No real difference.

NAS Minimserver vs NUC/AL Logitech Media Server = No real difference.

With the Naim NS01 there was a definite uptick using wav over flac. I read that potentially Linn sounds better with flac ......not to my ears.

With Minimserver I DO find a nice uptick in upscaling files to 24/192; NOT something I have enjoyed historically.

I had hoped that flac would work well as with Minimserver you can then flexibly upscale based on the file input, i.e. 16/44.1 > 24/176 & 16/48 > 24/192.

In fact I found that what sounded best was ANY wav file upscaled to 24/192.

Any thoughts or advise here gratefully received.
 

 

[Nestor Turton]

I use Mac Mini with both Linn & Chord systems and delighted with the sound. I’ve used a NAS and various server software, but never heard a difference between them unless I use DSP. I didn’t get on with Synology so converted it into a door stop.  I prefer Mac Mini running MinimServer or J River MC: swapping between them depending which system I’m listening to. I found putting various audiophile gadgets, PSUs and wires in my cupboard and going with manufacturers’ basic recommendations gave me the best results. I was not impressed with expensive products from Innuos and Melco, neither of which seemed to have any sonic benefits when I heard them and neither is as versatile as what I have.

After selecting speakers, I found Space Optimisation made the most significant difference for my Linn system and MScaler with my Chord system. Many years ago I had a Naim system and found it more finicky than either of my current systems and could easily imagine it would have sounded different with a change of server (from what I remember Naim strangelely didn’t play FLAC files very well, but they may have improved their software). I really like Linn because, it sounds great without having to tweak it and plays all lossless formats with equal aplomb. 

Of course different listeners may hear different things as evidenced by various reports I read of recent Bristol HiFi show. Best thing I heard at the show was Falcon Acoustics speakers. Very dissapointed that one high end manufacturer still seemed to think its expensive ripper produced better rips than a computer especially when its flagship streamer sounded nothing special to me (I preferred its entry level player). 

So I would not expect you to hear differences from different servers as long as player and server talk to each other properly across a fully qualified Cat5e or Cat6 network. 

 

[timster]

All I can say is that a Jitterbug connected to my NAS makes a difference. I can't say Asset or Minim makes any difference.

 

[Mr Underhill]

Thx Gents,

As I said above I am finding my normal tweaks are making minimal difference, although the Linn sounds good so I am not in any pain.

I am playing with SO, which DOES make a nice difference.

Timster, I do have Jitterbug and so I will try that on the Synology.

Nestor, I bought the Synology as I wanted a 10 bay NAS and it was available at the right price. Interestingly, with some more 'interesting' files, that is ones that make my system shout, I find the NAS better than the LPSU'd NUC.

I think Linn have a very built Qobuz into the DS very well, the sound quality is excellent.

M

 

[Nestor Turton]

M, I’m very glad your NAS gives good results; I suspect I was unlucky. I had a Synology DS413 & after a month it developed a problem with its fans. When I tried Synology support, I found it lacking and there was an expectation I would carry out the hardware repair myself. Synology sent some parts with no instructions, I’m not good with hardware, but did manage to fit the replacement fan. It failed again after 6 months or so. Synology were not especially helpful implying I had not fitted the replacement fan properly. I’m used to customer service levels I enjoy with Linn, Chord, Spendor and Apple so perhaps my expectations were too high. 

I’m sure there are many happy Synology customers, but I preferred a safer route. I know my way around a Mac being a long time Apple user. I have a 2012 Mac Mini with 8TB Thunderbolt Attached Storage connected over USB to my Chord system, which doubles as an OpenHome server for my Linn system. I use my home office Mac to rip CD, download music, tag with metadata and embed artwork then synchronise my local copy of the music library over my LAN with the Mac Mini. I have two offsite backups. 

Whereas, I think many users run their NASs around the clock and indeed have a permanent Internet connection and keep their system powered day and night, I only power on devices I’m using. I don’t currently subscribe to Tidal or Qobuz, but I do stream Sky Sports to TV which has HDMI to Linn DSM. Definitely, not advocating my setup is better than anyone else’s or it couldn’t be improved. However, my attempts to improve it with better switches and music servers were fruitless and I’m more than happy with the sound and system stability. I did find a couple audiophile patch leads I tried made a difference, for the worse, and discovered this was because they were dropping bits. So I stick to Fluke tested patch leads from Blue Jeans Cables. All my other cables are Linn or made by Mark Grant using Canare. 

Space Optimisation is wonderful in my view as it not only works a treat with my Linn system, but it generated the set of parameters I use with my Chord system courtesy of J River MC’s DSP Studio (poor man’s SO). Out of curiosity, I measured in room responses with REW and it closely followed my ideal curve. 

That’s about it really. 

All the best from Nestor

 

[Nestor Turton]

All I can say is that a Jitterbug connected to my NAS makes a difference. I can't say Asset or Minim makes any difference.

I read about the AQ Jitterbug and can’t figure out where it would connect in my Linn system as if I’m reading it correctly it is USB based and there are no USB connections. I don’t understand AQ blurb as it says it eliminates dropped packets, but I doubt anyone with a wired LAN would have any dropped packets (unless they were using the AQ patch cable I tried, which I found did drop packets. I discovered this when trying to update a Linn DS, replacing it with BJC Ethernet immediately solved the problem). 

I’m not saying you can’t hear a difference with Jitterbug or if I could as I’ve never tried one. I was puzzled by how it connected and what it did. Old fashioned curiousity, which I know can be felicidal, but I’ve always been able to figure out how to use iFI & W4S devices and what they do (they do make measurable differences, but whether they’re audible is down to the listener as with all things hifi), but AQ is a mystery to me. If it sounds better than it is better is a good way to view these tweaks. 

 

[timster]

I read about the AQ Jitterbug and can’t figure out where it would connect in my Linn system as if I’m reading it correctly it is USB based and there are no USB connections.

It's connected to any device's USB busses. So not on the DSM, but the NAS that serves the music.

It doesn't have anything to do with dropped packets - that's a network thing. It does something to the USB bus circuitry to eliminate jitter. And no, I don't understand it either. But it does have a +ve effect.

 

[Mr Underhill]

Hi Nestor,

Synology has been a happier path for me. I do find tinkering with grounding, power, cables and widgets has previously led to better sound quality; but, as always context is King, and the Linn appears to be remarkably self contained.

I will follow your advice and test the room with REW.

I have read Paul's v1 SO thread and followed his advice, but I suspect it is more geared for Linn speakers rather than the Focals I now use; and I sold my Kans a couple of years ago having kept them in the Loft for the best part of a decade; seemed such a waste to leave them unused.

I now have three v1 and four v2 SO profiles as I have been experimenting; I am getting better results with v2 currently.

Overall I THINK that I get marginally better sound quality using the NUC + Audiolinux as a server & a tad better using upscaled WAV over flac, but I would hate to do it blind; methinks I would fail that test! This is a great testament to the Linn engineering.

I was very tempted to double my outlay and get a Klimax Renew DS/3, equally I played with getting an ADS Katalyst second hand, but I think you can carry on down those paths into amounts I simply won't allow myself to spend. My feeling is that the KDS Renew is an excellent music making box that is more detailed, has greater dynamics and is better than my previous setup in all respects .......apart from some tonal richness, especially with female vocals and the bass, which I do miss a bit - but then I am getting much more elsewhere.

My LP12/Aro/Geddon has gotten much less use for the last couple of years.

Cheers,

M

P.S. Got some friends coming over who know my system well, I will be interested in their thoughts.

 

[MagicRat]

It's connected to any device's USB busses. So not on the DSM, but the NAS that serves the music.

It doesn't have anything to do with dropped packets - that's a network thing. It does something to the USB bus circuitry to eliminate jitter. And no, I don't understand it either. But it does have a +ve effect.

As in it’s plugged into a USB port on your NAS, but the NAS is connected to the network via Ethernet? 

 

[timster]

As in it’s plugged into a USB port on your NAS, but the NAS is connected to the network via Ethernet? 

Yes, but that doesn't mean it has anything to do with it. It's purely to reduce jitter in the device it's plugged into.

 

[Tendaberry]

I just bought a Jitterbug and will test it in my Qnap (and during CarPlay in the car  )

 

[MagicRat]

Yes, but that doesn't mean it has anything to do with it. It's purely to reduce jitter in the device it's plugged into.

Thanks. Might try sticking one in the back of my Roon Nucleus...

 

[Nestor Turton]

Hi Timster

Thanks for the reply. I remember being advised to use a dedicated USB connection on my Mac Mini for my Chord DAC, which is not easy as MacOS maps ports dynamically. You can see what it’s mapped in SysInfo and it isn’t only externally connected USB devices using the USB bus and juggling devices can achieve the desired result, However, I never thought about USB affecting my Linn system. I’ve always assumed using OpenHome (UPnP) I was pushing streams in to my DS’s buffer so it didn’t really matter. I can see my Mac Mini must poll its USB ports to check if I’ve plugged in something so it’s got me wondering.

Nestor

 

[timster]

Hi Timster

Thanks for the reply. I remember being advised to use a dedicated USB connection on my Mac Mini for my Chord DAC, which is not easy as MacOS maps ports dynamically. You can see what it’s mapped in SysInfo and it isn’t only externally connected USB devices using the USB bus and juggling devices can achieve the desired result, However, I never thought about USB affecting my Linn system. I’ve always assumed using OpenHome (UPnP) I was pushing streams in to my DS’s buffer so it didn’t really matter. I can see my Mac Mini must poll its USB ports to check if I’ve plugged in something so it’s got me wondering.

Nestor

My guess, and it's only a guess, is that a USB bus creates noise in the system which can be manifested as jitter. The bug has a pass-though so you can connect your DAC through the same port.