Ethernet cable comparison, basic and mid-price

My findings based on what I have heard with my new NDX

1. Wireless streaming via Belkin Surf wireless access point - bass rich however quite wooly

2. Wireless streaming with TP link router / access point - much the same as above with more detail

3. 10m Length Cat 6 from Novatec direct to NDX, much better details a bit bass light for my tastes

4. 10m Cat 6 cable as before but with a Netgear gigaswitch in place and a 0.5m Cat 6 cable from Novatec, exactly same results as 3

5. 0.5m cable Cat 6 replaced with audioquest Forrest, too much bass weight

6. Forrest replaced with Cinnamon.  Sounstage opened out, great separation and no weight given to any particular frequency, very nice cable

Hence why I have settled for option 6

I also  agree that the forest give a more basi sound, better in general with the cinnamon, but prefer at least 1 meter cable.

I tried both the AQ Forest and Cinnamon. Better than generic Cat6, but nowhere as good as Chord C-Stream.  Now I use a combo of C-Stream and Sarum TA with my NDS/UnitiServe/NAS drive.  All lengths are 1.5 meters.  C-Stream is not much more expensive than the AQ cables and is significantly more musical.  

AQ Vodka was replaced ny Chord C-Stream into the NDS from the switch. 0.75M. It's swings and roundabouts. I think I prefer the Chord for its less forward presentation. So many variables, not the least being ears. I think you can read it anyway you want.

OK, I suppose it's about time I posed my results!

System:
Synology Diskstation -> Cat6a SSTP + ferrites -> Hauewi Router -> Cable under Test -> ND5 XS -> 'lavender' interconnect ->Nait XS-2 -> AudioQuest FLX/SLiP 14/4 + ferrites -> Spendor SP2s


Tests:

1   Cat5e UTP: the baseline

2   Cat5e UTP + ferrites:
small improvement to resolution*
some improvement to treble - reduction of 'harshness' on some recordings.

3 & 4 Cat6 UTP
Same results as Cat5e UTP.  At one time I thought there may be an improvement but then on retest I decided it was either not there or insignificant.

5   Cat6a SSTP
more solid bass
greater increase in resolution*
noticeable improvement to treble - significant reduction of 'harshness' on many recordings.

6   modified Cat6a SSTP (with ferrites and other modifications.  This cable is currently undergoing a wider assessment)
further improvement to treble 'harshness' on many recordings.
further slight increase in resolution*


* = the increase in perceived resolution may actually be a reduction in background clutter.


I think my next steps are to try the same (unmodified then modified) with Chord C-Stream and possibly Supra Cat7+, each time comparing it to the basic Cat6a.



It's also notable that we've survived five posts without the naysayers who've never tried changing the cables telling us that we're deluding ourselves!

N.B. this wasn't an invitation to them.

However for those people who have tried different cables and not found a difference. then that is an equally valid result and would be a useful contribution.



Thanks to those who have contributed, or will contribute, their experience of these cables so that others can have more information from which to judge.

Thanks for a useful post, Huge. For what it's worth, my experience so far.....

Unitiserve in kitchen, connected to Airport Extreme with short cheap Cat5e.

Superuniti in next room originally streaming UPNP and iRadio by wifi. Sounded good when it worked, but was shockingly unreliable despite a rock solid wifi signal. (Naim people, if you're reading this, SU wifi is not fit for purpose, please sort it!) Hi-res files not playable due to constant cutting out and buffering.

Then added Airport Express to pick up wifi signal and connect to SU by short cheap Cat5e cable. Brilliant. Problem solved. Hi-res FLAC files now playable, general, all round improvement in sound quality, even the dreaded n-serve app loaded faster (although it still doesn't stay connected when you ask it to.)

BUT........like everythig else about the Superuniti, this was too good to last. All of the old problems reappear sporadically. I would guess my chances of being able to listen to what I want, when I want, currently stand at about 80%, which makes my opinion of Naim's streaming audio products very slightly higher than my opinion of the contents of my cat's litter tray.

Finally, in the interests of staying at least slightly on-topic, I replaced the cheapo Cat5e cable connecting to the SU with a 1.5m Chord C-stream. The reliability of the connection is no better, but there is definitely an improvement in clarity and detail of the sound.

WARNING: Off Topic Reply

Chris,

"Sounded good when it worked, but was shockingly unreliable despite a rock solid wifi signal. (Naim people, if you're reading this, SU wifi is not fit for purpose, please sort it!) Hi-res files not playable due to constant cutting out and buffering."
High-Res FLAC is right at the practical limit of the capability of streaming over WiFi, it's not just a matter of bandwidth but also of latency and the availability of WiFi bands that aren't being shared with other people in your local area.  If conditions are good, High-Res FLAC will stream fine, however if local conditions are not good then you'll get trouble with drop-outs.  The Naim manual does actually state that Wi-Fi isn't suitable for high sample rate streams - it's not a Naim problem.

"Then added Airport Express to pick up wifi signal and connect to SU by short cheap Cat5e cable. Brilliant. Problem solved."
"All of the old problems reappear sporadically."
I believe that Airport Express also shares bandwidth (i.e. uses contention management) if there are multiple devices in the neighbourhood all requesting service.  My understanding is that Airport Express is better than standard WiFi, but the fundamental problem of contention still exists.

"The reliability of the connection is no better..."
As the Airport Express link is still there, the same potential problem with drop-outs will still be present.
The only way to guarantee a clean, continuous (i.e. all data packets arriving in good time) uPnP feed is to use a hard wired connection all the way from the streamer to the router.
There is another possibility that may be causing you problems; that is the number of Broadcast packets on your network.  Check out Simon-in-Suffolk's post on the subject (unfortunately it's quite a technical area, but he's a real expert at networks, and we're very lucky to have his advice.)

"Finally, in the interests of staying at least slightly on-topic, I replaced the cheapo Cat5e cable connecting to the SU with a 1.5m Chord C-stream...   ...but there is definitely an improvement in clarity and detail of the sound."
Yes the observation that a Chord C-Stream improved the sound quality even after an Airport Express link is actually most interesting, even though (for reasons I give above) it can't improve the reliability of the connection.

Pedro,

That could be because a longer cable allows you to get a greater distance between the Router or Switch and the streamer.  Network devices are inherently electrically noisy, with very high frequencies - it's best to keep them well away from audio equipment and signal cables.

I was told by a guru at the Cable Company that digital cables should be at least one meter long.  I took his advice and replaced my Cardas .5 meter digital cables with one meters and I could hear a difference.

Bud

I would bin the airport extreme and get a decent router. Something like the awesome Asus RT range.

I am probably the biggest apple fan bois here, but their routers leave a lot to be desired even in an all apple environment.

 I had my UServe and NAS in my office and streamed music through an Airport Express plugged into an NDS.  I moved the UServe and NAS to the equipment cabinet and connected everything through a switch using Cinnamon cables.  The improvement in sound was dramatic.  I have the Airport Express connected to the switch so I can control things remotely from an iPad.

I can't say the AE was the problem or simply streaming music via a wireless connection but I firmly believe that things should be hardwired through a high speed switch where possible.  Why stream music through a $99 device like the AE (or any other one) when it's feeding 10's of thousands of dollars worth of equipment?

Bud

I'm following this post with interest as I use Supra Cat7 & I might need to get a longer length & a friend swears Chord C Stream is audibly better than Supra.

When I chose Supra it was audibly better than a no-name cat5e, very similar to AQ Cinnamon so much that it was price that swung the deal.

I've added ferrite & found it appeared to smooth some aspects of iRadio but could not detect anything with NAS.

So watching & reading in anticipation & getting home sick for my NDX while travelling USA west coast & deserts

Not a Naim owner (Simaudio Moon) but don't hold that against me! I have been following this and the other threads on ethernet cables and have found other threads that have helped me set up my system for the streaming age. I have just installed C-Stream (in place of Belkin Cat6) between router/NAS and living room switch/380d MiND with unnamed Cat6 from router to living room (under floor and through walls so hard to replace). All I can say is wow! It is not a small change. Everything is cleaner, weightier & better separated. Biggest difference is in the bass which is much tighter but top end is much cleaner as well.

I guess I was a bit sceptical on ethernet cables having an influence on sound and am gobsmacked at the effect of this change. 

As I needed 30 ft and didn't want to have to sell a kidney to finance it I went the sensible route and got the following:

Blue Jeans Cable - BJC C6AP (floating shield)
Bonded Pairs, Cat 6a, 500 MHz; Rated CMR for installation
Every cable individually tested, with report.

Sounds great and I use a second one for streaming large video files to my Oppo player.

Sorry I forgot to mention I'm using a SB Touch with FLAC files loaded on my PC. Wi-fi is disabled on my router and on SB Touch. Files are being decoded at the server (PC) not by the Touch.

For video files I'm using Mezzmo from my PC to the Oppo.

Everything works flawlessly.

I've been using a Sonos Connect for a few weeks with my DAC V1 and tried a few ethernet options yesterday.

1. 10mtr Belkin cable from router to Connect, Coax to V1

2. As above, but a female to female coupler with 1.5mtr of Audioquest Forest (£29.00 from Amazon last week).

3. 1.5mtr Audioquest Forest from router into Sonos Bridge and then wireless to Connect.

1. Been using this option since obtaining the Connect - nice balanced sound.

2. This improved things somewhat, I noticed the bass was better defined, less bloated sounding.

3. Sounds the same as 2.

So currently now using option 3, nice to be wireless and no drop outs!!

Anyone else using wireless over cabled?

Cheers,

Bill.

Has anyone compared their wired Ethernet results to a USB memory stick?  Would that not be a better baseline?

I found that the USB stick sounded better than any wired option but I could not tell them apart in a true A-B blind test.  Now I know blind testing is a controversial topic on this site and many posters think that it is flawed and is much better just to make a change and sit and listen to it.  I don't, and the fact that I could not tell the two apart surely means something?

I'm hoping to set that up later.  Currently I'm optimising my network streaming set-up; until I've done that there's little point in comparing it to USB.

As you know, the choice of appropriate control (i.e. baseline) depends of what you are trying determine.  At the moment I'm still in the network system optimisation phase hence comparing each cable system under test against the previous perceived best option, and using Cat5e as the control reference.  Evaluating the result against USB comes later.

And no I don't discount blind testing when it can be set up and carried out properly.  However, for practical reasons, I can't do all my tests using a blind protocol with full controls.

Other people will be at a different stage to me.

I totally disagree.  The thing about a USB memory stick is that there is no Ethernet so that goes away and a baseline does not have to be the poorest performer.

The other thing about using a USB memory stick as the baseline is that we can all easily do it, does that not count for anything?

What protocols need to be setup for blind testing?  All I am saying is that if you can do the tests blind then do it, so please do not try to belittle the exercise with the use of techno babble.  Do the tests blind because otherwise they are meaningless.

When you say the Mac DAC I assume you mean the 16bit DAC in your MAC PC???  You prefer this to your NAIM DAC??????

Why do you repeat this allegation?

Is it you simply not understand what I've written?

So that other people will know why I will no longer reply to you and,

1  know that I haven't conceded your unproven assertions,

2  know that I'm not the one being rude

the reason is given here...

Resorting to personal insult rather than rational discourse illustrates that your comments are no longer worth any further reply.

Should you apologise appropriately, I will reconsider.

Sorry when you said the Mac DAC I got confused.  The Lenovo probably has really Naff Speakers like my iPad or my wife's smartphone.

What allegation have I repeated?  I have alleged anything about you.  Please tell me what allegation I am repeating.

I also have not asserted anything aside for the requirement to do blind testing.  Why does the idea of blind testing frighten you so much?

I have also not insulted you and I never said you were being rude.  The thing is you cannot stand being questioned about your posts, we have to accept them as gospel or you get annoyed.

You go on about protocols and controls for blind testing but you won't say what these protocols and controls are and why they are so impossible for you and us to carry out.  While you are at it explain your method of testing and why it is so much better than a simple blind test.  If you don't like my use of the phrase 'techno babble' then explain some of these things so we know what you mean.  Don't just drop them in to a discussion without explaining what you mean. 

That is why I refer to it as techno babble because it is words without substance when you say things like: ' I can't do all my tests using a blind protocol with full controls..

Exactly where did I personally insult you?

Don't hold your breath waiting for an apology from me.

I prefer USB to Ethernet by a small margin. I haven’t been able to do a blind test because it is impossible. A USB stick cannot be made to look like an Ethernet cable and they plug into different sockets on my streamer.  I can say with confidence that they both work significantly better than a placebo, which not surprisingly produced no music at all.

A USB stick on my UQ sounds better than streaming (AQ Cinnamon all throughout) though I haven't compared since adding a V1 to the mix. But streaming from a NAS using nstream is a heck of a lot more navigable with around 1000 cds in my collection and I don't plan on having a computer near my stereo any time soon. Waiting on some linear power supplies for my Vortexbox and switch and modem so will see what that brings....