Uniticore - first impressions of sound quality and app interface

Posted by: J Saville on 20 December 2016

Last night I had the pleasure of being able to compare the newly arrived uniticore to a unitiserve SSD on a reference level Naim system. 

The reference system consisted of: 

NDS, 2x555ps, Statement S1 pre-power, S-800s, full fraim rig, unitiserve and core in rig, netstreams switch also power lined. External HDD with Chord Sarum Super Array USB, Chord Music Ethernet from Server to switch. 

So the first thing we did was power up the core, which immediately prompted us with a firmware update. This was fine, and after no more than 5 minutes the core was up to date and ready to database. Then we plugged in the external HDD, which was loaded with around 6000 albums. The core started databasing immediately, with music appearing real time. It should be noted that it takes a wee while with that much music, and it still hadn't finished databasing by the time that we had finished listening at around 10pm...

In terms of the interface, the core showed up immediately in the new app. It is very easy to configure, with local music storage as well as network shares/stores able to be configured from the iPad. We didn't jump too far into this however as we were mainly interested in how it sounded, so we just let it database what we already had available. 

I have to say that Naim have absolutely nailed the interface. You now get artist cover art. It grabs a cover from the album folder within, in the case of more than one album per artist it generates a collage from all coverart from every album! Very clever work here, really impressed.

So down to business with the all important listen. We started with the core, queued up a track by Angus and Julia Stone. I have to say it sounded absolutely fantastic. I haven't heard this system offer such a sense of rhythm. Naim have certainly build a Naim box here!

We then switched back to the unitiserve, simply unplugging the Ethernet and USB cables and switching them over to make testing fair. Hit play on the same track, and... It just felt a bit off. The unitiserve has always sounded better to me than other upnp servers running on other hardware, but now it just sounds a bit disorganized. The bassist is checking his watch to see how much of the gig he still has to play, the drummer keeps nodding off. Hmm.

After this the unitiserve was immediately unplugged and removed from the stack, and didn't get a look in for the rest of the evening. 

We did further testing on the core, and a big step up in sound actually came from removing the unitiserve PSU from the loop. Clearly the switch mode power supply wasn't helping things!

What I found most interesting about the whole thing was that suddenly I wanted to buy one, where previously I was content with minimserver running on my readynas. The unitiserve was always cost prehibitive to me, but the core is in my eyes an absolute bargain. A core is now most definitely on my list, right after a new unitinova, which I'm hoping are not too far away...

Well done Naim, 5 and a half stars out of 5.

Posted on: 26 December 2016 by ChrisSU

I think the point is that getting a bit perfect rip is achievable with most ripping software, and can be confirmed by checking it against an online database. If these devices sound different, it's likely to be caused by differences in the UPnP server, PSU quality, or network, but not by the bits themselves.

Posted on: 26 December 2016 by French Rooster

i think the ripping of the core vs unitserve is the same, about perhaps a better metadata and tags for the core. But is the sound quality of the core, as a nas, is better than the unitserve with a linear power supply?  i am very curious to here next sharings from users.

Posted on: 26 December 2016 by J Saville

Hmm, I was sure that I had read somewhere that they used 2-pass ripping, but it appears I was mistaken. 

From the hdx ripping engine white paper:

'The Naim audio servers use a superior technique called Secure Mode Ripping on a specially selected audio grade CD mechanism. Secure mode works by using a CD mechanism that does not cache data and reads the sectors multiple times. It also uses the un-correctable C2 error feedback information supported by the CD mechanism, so known problematic areas can be re-read and handled accordingly.'

Posted on: 26 December 2016 by ChrisSU

A ripper can read and re-read a CD as many times as it wants, but the point is that the Accurate Rip process will check that every bit has been read correctly so that the rip is identical to the CD. So a rip is either right or wrong, and the Naim ripper is not the only one capable of getting this right. 

Posted on: 26 December 2016 by Bart
ChrisSU posted:

A ripper can read and re-read a CD as many times as it wants, but the point is that the Accurate Rip process will check that every bit has been read correctly so that the rip is identical to the CD. So a rip is either right or wrong, and the Naim ripper is not the only one capable of getting this right. 

I'm guessing that we won't have a repeat of the UnitiServe discussions over whether the UnitiServe ripped "better" than a computer could.  But I've been known to guess wrong!

Posted on: 26 December 2016 by Simon-in-Suffolk

Indeed Bart, it does amuse me to think in what way data could be read from a CD incorrectly and not sound like random static... perhaps others might try and explain, or are we talking about the occasional tick sound or silence of complete data loss from a bad or marginal CD pressing?

S

Posted on: 26 December 2016 by CharlieP

J Seville,

Please don't let anyone talk you out of listening to new rips from the Core.  I have no idea why a rip is not just a rip, and many knowledgeable members here will put forward explanations for why they cannot be different.  I have personally been exposed to UnityServe rips which, on a high resolution system, sound more natural to my ears than my own XLD rips.  It appears that some at Naim (who are famous for leaving no stone unturned in their quest for audio quality) know something about ripping which is not "common knowledge."  I among others are curious to learn if Naim have any further improvements to this process embodied in the Core.

I am sorry if posing this question has derailed the original thread.  It would seem that speculation is of no use, and what matters is to listen and report.

Cheers to All,

Charlie

Posted on: 26 December 2016 by Mr Happy

 I also did a comparison of rips and they were slightly different. I tried a few before settling with xld, as I rip with a mac. I recently tried a rip with dbpoweramp, which sounded slightly different to xld. I personally prefered the presentation of xld. The other day I compared a rip from a friends unitiserve to an xld rip, this time I prefered the unitiserve rip.

The differences are subtle but in a revealing system they really do make a difference. I don't claim to know why these differences are there, but they certainly are there.

Posted on: 26 December 2016 by Simon-in-Suffolk

Oh I can sense a possible repeat of what a few of us did on this forum severeal years back... a few of us examined and checked some rip files that were shared... including from  iTunes, dbpoweramp, Unitiserve, and EAC... and apart from the iTunes rip which had a few mS chopped off the front of each track, the rest were bit perfectly identical, there was no digital or numerical differentiation at all.. absolutely identical... they key is to ensure you are ripping from the same CD pressing/master...

According to Naim some of the key advantages of the Uniticore lie in its library and media management as well as a tidier and cleaner digital output due to a better powersupply.

i suspect some people may  confuse ripped file differences with media server streaming differences, the latter can  cause a change to the produced sound when interacting with a Naim streamer, and here I would expect the Uniticore to have a 'signature' sound, albeit this  will have nothing to do with the produced rips.

Posted on: 26 December 2016 by Innocent Bystander
Simon-in-Suffolk posted:

Innocent Bystander, don't know where you got that comparison from.. certainly no one at Naim I have spoken to recognises that... they are completely different products, the only thing I have definitely heard is that there is an expected performance lift compared to the SPDIF out of the earlier Unitiserve.. through better decoupling and powersupply amongst other things.

IIRC Phil Harris stated that in his system, which includes NDX, Core through SPDIF  was indistinguishable, and, though rather vague, I also thought there had been a more general observation attributed to Naim to the effect that the Core was as good as any existing products. 

The point , of course, of my suggestions to compare, when people have their hands on Core in a system capable of revealing differences, is to find a general concensus as to how good it actually is as rendered source (player) direct into any DAC, to see if it is as good as any of the generaally well respected rendering sources including NDX, Melco, etc.

Posted on: 26 December 2016 by Bart
Simon-in-Suffolk posted:

Indeed Bart, it does amuse me to think in what way data could be read from a CD incorrectly and not sound like random static... perhaps others might try and explain, or are we talking about the occasional tick sound or silence of complete data loss from a bad or marginal CD pressing?

S

As evident in this thread, people can not-blindly a/b compare ANYTHING and hear differences.  People heard the green pen.  People will always hear everything.  It's ok . . . just the way it is.  

Posted on: 26 December 2016 by Simon-in-Suffolk

Indeed.. and if one is content on this basis then this is absolutely fine, I guess when it becomes an issue to me is when perceived differences are attributed to false assertions... to be fair to Naim I don't think such assertions are from them,  I think they are measured and responsible in their assertions.. there is no magic or outrageous claims used in their designs, just honest engineering design, and in the digital world they tend to implement and internalise well respected third party technologies to great effect whilst following true to their DNA. 

 

Posted on: 27 December 2016 by Simon-in-Suffolk
J Saville posted:

Hmm, I was sure that I had read somewhere that they used 2-pass ripping, but it appears I was mistaken. 

From the hdx ripping engine white paper:

'The Naim audio servers use a superior technique called Secure Mode Ripping on a specially selected audio grade CD mechanism. Secure mode works by using a CD mechanism that does not cache data and reads the sectors multiple times. It also uses the un-correctable C2 error feedback information supported by the CD mechanism, so known problematic areas can be re-read and handled accordingly.'

Indeed this is a benefit from Naim and many non Naim rippers, it gives a higher confidence to recovering data and reducing lost data  from a damaged CD. However if the disc is badly damaged beyond easy recover it rarely makes a bit perfect rip.. usually a tiny silence appears or a skip.. just like on CD play back.