is it worth to upgrade to new 24bit/192kHz streaming board

Yes, I mistyped. I am getting a ND5-XS. I just seemed a better idea than upgrading the Qute, since I have NAC/NAP separates now. I would be surprised if it isn't more than a inconsequential improvement than my Qute. I'm also looking forward to streaming my iTunes ALAC library.

RaceTripper,

It is a great convenience to be able to play the ALAC files. But if you have music that you love only available in that library, you might not be hearing the best sound. I have done this several times, most recently with MIdnight Oil's Blue Sky Mine and just today with Dread Zeppelin's Un-leaded where I have a copy of the music in my iTunes library from before and then drop the same CD into HDX or UnitiServe for naim style WAV rip and then marvel at the difference between the two versions. 

I should rip with newer versions of iTunes on newer hardware to see if the difference is still as great as these few year old iTunes rips. Maybe things have changed?

My Qute is off being 24/192 upgraded at the moment, primarily to provide native ALAC support without transcoding (I hear no difference between WAV, FLAC and ALAC on any system I have tested on and certainly no difference between WAV and FLAC on my Qute).

Looking forward to seeing whether this does indeed provide 802.11n, whether the UI is any more responsive and whether general SQ has changed in any discernable way.

Will report back once I have it back.

I don't have a HDX nor UnitiServe (and no plans to get either). I don't have a CD player either. So ripping CDs on my Mac is how I roll. I have a bunch of them ripped to FLAC using XLD and also ripped directly to iTunes as ALAC. Once I have the ND5 I'll see whether I hear any difference between them. In any case, most of my serious listening is vinyl.

Please let us know what you find between the FLAC and ALAC(luster) rips. Does the server you are using transcode from FLAC to WAV before sending? Musically useful to do so.

I have been intending to try XLD sometime but time I don't have much of and the Naim rips prove themselves so musical and the UnitiServe such a great sounding UPnP server compared to my ReadyNas Duo that I haven't been compelled to do a lot of research. You should try to borrow a UnitiServe sometime just to see what it adds to the picture. Even if you just get a few Naim WAV rips of your favorite albums from it to compare to your XLD Flac rips and iTunes ALAC versions, that would be useful.

Hah, the last thing I want to do is borrow a UnitiServe and get tempted to buy it. 

If I spend any more money on digital it will be for a PSU for the ND5. But, I'd much rather put that money into upgrades to my analog front end, like going from a Rega RP3 to a P7, or comparable, then upgrading the phono (cartridge & preamp).

Ok - I had the upgrade done - new 192 board and software update to my NDX . The short answer: NO COMPARISON to the original!  Sound is more open transparent and in particular, more analog sounding...relatively. To  be clear - the original set up was excellent, however the update definely yielded some significant benefits. This assessment does not factor in the added benefit of the sound of 192 tracks!  Is it worth the cost?  ABSOLUTLELY! if you are in the window of a free upgrade, as I fortunately was, it's a know brainier. 

Regarding the discussion unfolding about WAV v FLAC Files - I can tell you definitively that WAV sounds clearly better than FLAC, however it Is all relative to how high you are on the Naim chain. I listened to an A-B comparison on a top of line Naim system and it was an obvious difference, on my system it's better but less obvious.  My system is pretty good but not top of line by any means.  If you can't hear the difference on your system - forget it perhaps - but know that it IS better on a higher resolution system.  Hope that is helpful?!

I'll be interested to find out how the upgrade changes the sound of the UnitQute if at all. I wasn't going to bother upgrading, but if it yields better sound quality for ripped Cds then I could be tempted.

I just had my UnitiQute upgraded (by Connon -- THANKS!).

There is a significant improvement in sound quality -- something significant changed (for the BETTER).  Everything is more open and there is more air in the music.

While I was at the shop, we also listened to the system with a Nserve (rather than just a Mac).  Even with the new ALAC support on the 'Qute, it sounds better with the Nserve doing the transcoding (I can't explain this unless it has something to do with heat or power usage -- maybe one of the Naim design folks could chime in?).

Connon also had some Cat-7 ethernet cables on hand (vs the normal C5-e)... The difference was subtle, but there was an audible different -- Cat-7 has more shielding (so less noise)... These cables were also a high-end variety with some silver wire (don't remember the brand) -- didn't have computer-grade Cat-7 to compare... But, there was a difference with the Cat-7.  (Much smaller change than the 'Qute upgrade).

Thanks again Connon.  It really was a sonic upgrade -- more than just playing ALAC.

Scott B.

Scott, are you actually saying that different network cables produced improved sound? I find that difficult to understand...

That would suggest some form of electrical interferences between the ethernet signal on the cable itself and the unit, unless this particular network connection had a large amount of ethernet packets being lost, which the new cables reduced, but that would be a big not a subtle improvement... At the end of the day 0s are 0s and 1s are 1s, are'nt they? Also the actual signal on the network cable is in the 10 to 100 mhz range, isn't it? I mean 10 mbps (basic ethernet speed, fairly slow by modern standard) is 10 millions 0 or 1 signal per second, and the Cat5 specs require certain electrical characteristics to be met at frequencies up to 100Mhz anyway...

I am quite happy with this phenomenon if somebody actually comes up with a scientific explanation for it - but not seen one yet. if one tells me that there are measurable radio interferences from the ethernet signal and that different cables are not equal in this respect, then I would accept this can impact the sound on the analogue side of the DAC... Is this what people assume is the case?

Herve:

Not just suggesting it... I heard a difference... and, it wasn't just different, the Cat-7 was better (fuller, richer, better secondary harmonics, etc -- sonically, it was in the direction of adding an XPS to an CDX).

I can't explain it.  Honestly, I am disappointed -- I thought that with the digital file streaming of 0s and 1s (rather than a bit-stream), we would have moved beyond the influence of cables on the sound quality.

Before I heard the different, I would have said that there is NO way that there can be a difference -- it is just 0s and 1s over an error-correcting Ethernet link... Right?

Cat5 vs. Cat7 may explain a bit of this because the Cat7 has better interference rejection (which could decrease the number of re-trys, and may in some way impact things (but the buffer was full...?).  If this is true, any Cat7 should improve things (and have the same sound quality)...  I didn't have the opportunity to play with other Cat7 cables, so I don't know if it is the difference between the Cat5 and Cat 7 cable spec or something else at play (what else could it be?).  [If it is more than just shielding, it would also call out that the Ethernet switch is also an opportunity for audiophile-tweaking (I haven't heard of an audiophile version of a switch yet)].

Naim (and the 'Qute is no exception) is known for PRaT, so anything that impacts pace and tempo in the signal (micro timing alignment of the bit-stream (HOW does this playout?  what is the science behind it?) would be audible with Naim equipment (more than some other brands).

I don't know enough about the uPNP standard/protocol and its relationship to Ethernet to know about the enforcement of check-sums, etc... what I heard, I can not explain.  And, since it is about a file transfer (accuracy of 0s and 1s, not sound-shaping, etc), I don't understand how something like silver or gold in the cable would change the signal (shielding, yes, but metal...?).  Along this line, if silver really helps out the file transfer, then why don't we use silver Ethernet cable for the backbone links between high-performance servers in the computer rooms?  what am I missing?).

As I said, I don't know enough about uPNP and I can't explain what I heard.

Are there any computer network specialists and/or electrical engineers who can shed some light on this cable issue?  What is causing the sonic difference?  Is there any science here?

Thanks.

Scott B.

There is a lot written about this, and is all down to interference, both common mode and internal (through the transformer effect) and external induced interference in the shield if present.

The cables themselves have no sonic difference, but the interaction of the conducted / generated RF interference with the receiving DAC or equivalent does. This does mean every implementation will be different to some extent. No Ethernet cable at our speeds for streaming  and short domestic lengths unless in an extremely industrial noisy environment is going to develop frame errors, if it does, it's faulty.

A good quality Cat 5e cable should be fine and generate little RFI, as the twisted pairs are accurate and radiate or induce little. ( admittedly in can be difficult to get good quality 5e, which is sometimes exploited by the audiophile cable merchants) Cat 6/6a, has more tight specification on twisting and therefore cable cross talk  but uses thicker wires to reduce voltage drop over longer distances.

Cat 7 can be shielded but one must take care to used compatible earthed connectors on the network equipment or loops can occur or the shield goes high impedance and cause more problems.

Scott, you mention, '0s and 1s', I am afraid that doesn't apply to cables. Technically cables are at layer 1 in the OSI network stack and it's all analogue signals/pulses of light etc at this point, and the regular physics of transmission lines and other signal transmission parameters  apply.

However if you are hearing a significant sonic differences with Ethernet patch leads, I suggest you have an RFI issue, either you are a very electrically noisy environment (unlikely domestically unless you use Ethernet  over power-line in your house) or your connected switch is noisy. Either way I recommend clamping some ferrite chokes over your Ethernet lead near the point of entry to your audio equipment. It's what I do, I simply add until I hear no sound improvement ( I currently use a large  RG213 style clamp, with Blu-Tak ) and I use regular 5e from a Netgear switch, and I take care to route my Ethernet lead away from mains, speaker leads and interconnects. Try and look for 'green' switches that lower the Ethernet signal current on shorter patch runs, again these will create less RFI, I have yet to experiment with these yet..

Simon

Thanks simon, your explanation makes sense to me - and I agree loss of packets is very unlikely these days. RFI is the only logical connexion and also an area where the different types or quality of cable should probably have an impact... How much of that is a "placebo" effect in the end is probably a matter of personal experience/opinion, I guess! :-) To that effect, I just ordered some clip-on ferrites, to give it a try and see if I notice anything... dead cheap, anyway.

If using cables causes RFI interference, is wireless potentially better?

I suppose switching on the wireless receiver within the streamer may cause additional load on power supplies etc and cause a decrease in quality, but I'm wondering if the best solution maybe an airport extreme connected to the streamer with a short ethernet cable so as to minimise the effects of RFI.

I have to say I've tried everything from ethernet over mains, airport express as a wireless bridge and I now just use the wireless of my unitqute and can't say I've noticed much difference. Obviously with more revealing equipment this might not be the case. 

Hi, I guess a reputable component vendor such  as RS Components or similar, but if wiring your house you might need to get a special crimp tool to add your own connectors.

If you are planning infrastructure wiring you might want to consider Cat 6a. It has a high frequency bandwidth. It's down side is that it is not flexible and not ideal for patch leads, but great for infrastructure.

If your network equipment supports grounded leads, you also might want to consider shielded variant of Cat 5e or Cat 6a. They should commonly have the term  FTP for a common foil shield or STP for shield around each twist pair. The latter is probably over kill, theformer will help stop the cable radiating or picking up interference where the twists are less than perfect. 

Simon

purchase 192k clare martin new aldum from linn and the get rid of 16 bit can come sooner than later

Unfortunately (and if i have understood the above communication correctly) not even 192k can make Claire Martin sound good.