Mu-so, audio drop outs...A cure ?

Posted by: Super on 18 March 2015

Hi all, this has been spoken of elsewhere on this forum, by me mainly. I've had a lot of trouble of audio drop-outs on the internet radio on the mu-so whether it was on wi-fi or ethernet. I changed channel settings on the router which helped a little but it was still there. One of the problems is that i live near a taxi rank a couple of doors down and i know for a fact that when they cue the mic it interferes with my m/c cartridge on my LP12, so it's possible it could do the same with the mu-so. When using ethernet i was using a 5 metre CAT 5E connected to extended wi -fi router and i was still getting drop outs, so i bought a 3 metre CAT 6 cable. Guess what, no drop outs and had it on for days now and still going good. I thought i would let anyone know who has had a similar problem, now i wonder what a CAT 7 would be like ?

Posted on: 18 March 2015 by ChrisSU
Sounds like a good result. I'm pretty sure Cat 7 is all shielded, so it should work too. (If not, I'm sure someone who actually knows about this stuff will be along to correct me soon enough ????)
Posted on: 18 March 2015 by Huge

The Ethernet cable category is based on transmission standards, not construction.

 

However, in general,  Cat 6 may or may not be shielded, Cat 6a is usually shielded (but can be unshielded), but as far as I know there are no unshielded Cat 7 cables being made.

 

Shielded cables are described as STP, SSTP or SFTP; unshielded cables are described as UTP.

Posted on: 18 March 2015 by Super

I'm no expert on this, but all i know is since i've put the cat 6 cable in i'm no longer getting drop outs, so it's doing something right.

Posted on: 19 March 2015 by Simon-in-Suffolk

I suspect if it was to do with the ethernet cable its was more likely the original Cat5e cable was faulty and only allowing operation in half duplex mode or similar... any way you have replaced your cable and the problem has gone..

But a proper Cat5e cable is more than adequate.. And even if there was the occasional data corruption the more likely impact would be sticky UPnP updates or discovery issues rather than media drop out.

 

Also if your RFI exposure was intense it would impact your mains wiring more than ethernet and you would get distorted break through into your Naim audio equipment irrespective of source, TV, radio and even phone. Ethernet cable uses twisted pairs and therefore rejects to a high degree ingress or egress RF interference and cross talk... Mains wiring is not... And often cartridge wiring is not...

Simon

Posted on: 19 March 2015 by andarkian

When used with the tv i.e. the optical connection, the Muso drops out at least once a day which drives my wife to distraction. This does not happen when, as right now, I am streaming music to the Muso. Hopefully the next patch will fix this. 

Posted on: 19 March 2015 by GregU

What do you mean patch.  Is there some way to update or is it automatic   Some weeks it's fine and some times I get 10 drops a day

Posted on: 19 March 2015 by andarkian
Originally Posted by GregU:

What do you mean patch.  Is there some way to update or is it automatic   Some weeks it's fine and some times I get 10 drops a day

If you look at the first page of Streaming you will see that Naim are looking for volunteers for the new release of firmware for the Muso. As i said, I only get drops on the physical TV connection to my Muso not on the WiFi streamed music.

Posted on: 19 March 2015 by GregU

Thanks

Posted on: 19 March 2015 by Super
Originally Posted by Simon-in-Suffolk:

I suspect if it was to do with the ethernet cable its was more likely the original Cat5e cable was faulty and only allowing operation in half duplex mode or similar... any way you have replaced your cable and the problem has gone..

But a proper Cat5e cable is more than adequate.. And even if there was the occasional data corruption the more likely impact would be sticky UPnP updates or discovery issues rather than media drop out.

 

Also if your RFI exposure was intense it would impact your mains wiring more than ethernet and you would get distorted break through into your Naim audio equipment irrespective of source, TV, radio and even phone. Ethernet cable uses twisted pairs and therefore rejects to a high degree ingress or egress RF interference and cross talk... Mains wiring is not... And often cartridge wiring is not...

Simon

I have two of those 5m cat 5e cables and they were both the same. I suspect due to the length they were acting like an aerial in my bad polluted RF area. The cat 6 is better screened so blocking out that RF.

 

 

 

 

Posted on: 20 March 2015 by Super

Simon, that cat 5e cable wasn't shielded so i guess there was the problem. I have been looking on the internet and you can get shielded and unshielded and there is not much difference in price. My question is.... why would anyone want an unshielded one, i know i did, i just picked one from the shop not looking and showing my ignorance at the time. I thought all ethernet cables were all the same, guess i know more now. I have just ordered some high performance cat 7 shielded (STP) cable.

 

Posted on: 20 March 2015 by Simon-in-Suffolk

With ethernet leads you don't need shielded leads in most environments, infact shields can cause more troubles than they solve in some scenarios with creating ground loops. Shelding in my book  only offers benefit in extremely electrically noisy environments or where you are wanting to minimise crosstalk between bundles of cables in long runs  where they are in close proximity to each other and/or carrying very high bandwidths over long lengths.

Ethernet cable use twisted pairs and use differential balanced drivers, thereby theoretically avoiding the need of screening. In the real world however twists are not perfect, and the screen gives another layer of protection to these slight imperfections. The shields can be around each twisted pair or around the bundle of 4 twisted pairs or both.. Each construction type addresses a specific need for a specific envionment.

Simon

Posted on: 20 March 2015 by ChrisSU
Cat 7 can carry a lot more data, but that's unnecessary for even hi-res audio files. Having said that, all 'audiophile' ethernet cable as far as I'm aware is shielded Cat 7 (Chord, Audioquest).

One practical consideration if you have a long/awkward cable run is that some of them are rather stiff, and can be damaged if you need to bend them around tight corners.
Posted on: 21 March 2015 by Simon-in-Suffolk
Originally Posted by ChrisSU:
Cat 7 can carry a lot more data, but that's unnecessary for even hi-res audio files. Having said that, all 'audiophile' ethernet cable as far as I'm aware is shielded Cat 7 (Chord, Audioquest).

It is worth noting for those that are thinking of using Cat 7 for thier ethernet LANs that is not a universally ratified standard by the IT industry.. It has not been ratified by the TIA.

The current highest spec ratified cable standard for Ethernet is Class EA or Category 6A.

it appears the next ratified standard may be called Class 8 (simply so as to avoid confusion of two competing and differently specified 'Cat 7' standards.)

 

By all means buy a patch cable including 'cat 7' boutique types because they sound better; but I advise against buying if relying on the often, in my opinion,  shameful marketing info ( thinking of Audioquest Vodka here), and I advise against buying Cat 7 if you are doing so for future proofing reasons, as it is not universally ratified and it is unlikely to be. If you want shielding and low crosstalk for your main ethernet wiring I suggest audition Cat 6A now or wait for Cat 8

 

Simon

Posted on: 21 March 2015 by Mike-B
Originally Posted by ChrisSU:
.........   all 'audiophile' ethernet cable as far as I'm aware is shielded Cat 7 (Chord, Audioquest)

I believe the higher end Chord cables are bundled mini-coax's, not 4x TP ethernet  & as such do not qualify as Cat anything.   I'm not saying anything about SQ,  & it might well be they perform to the same data transfer rates & bandwidths of Cat 6 or 7,  but Cat they ain't.

 

Agree Simon on "shameful marketing"..........  pure BS & liberal applications of rarefied snake oils.   

Posted on: 21 March 2015 by ChrisSU

Mike, I believe you're right about the higher spec Chord cables: I didn't feel the need to to elaborate on that, in the context of this thread, as some of these cost as much as, or more than a Muso.

 

As regards dubious marketing, yes, but you're into a price bracket where you should be able to audition and decide for yourself before you drink the snake oil...

Posted on: 21 March 2015 by Huge

Simon,

 

Thanks for the info.

I didn't know that Cat7 is unratified, in the circumstances I feel it inappropriate for any manufacturer to label cables as Cat7 ("meeting the proposed Cat7 standard" or "Cat6a+" would however be acceptable).

 

As it happens I have done as you suggest!

 

I mostly have Cat 6a SSTP (avoiding multiple earth points on any one 'section'!), since where I live there's a lot of RFI and without the shielding it gets into the audio system via the Ethernet; but one link is Chord C-Stream as it sounds better!

 

 

Posted on: 21 March 2015 by Simon-in-Suffolk

Huge, unfortunately it appears not quite as black and white as I would like. The ISO/IEC 11801 defines class  F and class Fa cabling specifications that correspond to Cat 7 and Cat 7a. It's simply it would appear the TIA has not ratified for use with Ethernet..

My instinct would be to stear clear, as without official industry support at least with the TIA, there is the risk that Cat 7 joins the realm of snake oil merchants for use with ethernet.

Simon

 

Posted on: 21 March 2015 by Huge

Simon I suppose any form of clarity in the situation is too much to ask.

 

Most people find much to do with computing to be complex and confusing, and standards tend to be written in legalistic terms.  So, given the quote from 'Mr Bumble' "...then sir, the law is a ass"; we have a complex problem written in asinine language.

 

I rest my case!

Posted on: 22 March 2015 by Simon-in-Suffolk

Indeed, another way of looking at it is that for the intend ethernet uses Cat 7 is seen by many in the industry (ie TIA) as in appropriate. The average consumer is fine with 5e technically in thier low speed home LANs. ( or 6a if they have some sort of industrial RFI issue.. Which personally I would be concerned about from a health point of view). I guess however there is room in the boutique market to create exotic cables.. I have no issue with this, my beef is when companies try and do this on pseudo technical grounds.. If however they said, 'we tried this, it worked and it sounded better when used like this' then they would probably gain my respect.

Simon

Posted on: 22 March 2015 by Mike-B

I read a report of a number of tests on Cat 5, 5e, & 6 patch cables (lengths of made up RJ45 terminated cables) from Blue Jeans Cables, USA that showed an alarming low number failing to meet test compliance. 

My concern is that if that is the norm, then it could be many of our home cables assy's will drift out of spec due to a bend or crush or poor termination. I have assumed - in theory at least – a higher spec cable such as Cat 7 will meet & exceed the spec for 100Base-T.   Also not all Cat 7 are in the boutique category.  (no pun intended)

I would be interested on your comments on the Blue Jeans tests

Posted on: 22 March 2015 by solwisesteve
Originally Posted by Mike-B:

I read a report of a number of tests on Cat 5, 5e, & 6 patch cables (lengths of made up RJ45 terminated cables) from Blue Jeans Cables, USA that showed an alarming low number failing to meet test compliance. 

My concern is that if that is the norm, then it could be many of our home cables assy's will drift out of spec due to a bend or crush or poor termination. I have assumed - in theory at least – a higher spec cable such as Cat 7 will meet & exceed the spec for 100Base-T.   Also not all Cat 7 are in the boutique category.  (no pun intended)

I would be interested on your comments on the Blue Jeans tests

There are several white papers on the Blue Jeans site. Which one do you refer to?

Posted on: 22 March 2015 by Mike-B
Originally Posted by solwisesteve:
There are several white papers on the Blue Jeans site. Which one do you refer to?

This is hijacking a Mu-so thread - I'm starting a new Cat thread 

Posted on: 22 March 2015 by Simon-in-Suffolk

Mike, I expect many cables will fall out of compliance if bent too tightly or crushed.. I would have thought one  really need armoured cabling if you want to protect yourself from this totally.

For RF I use hardline.. And you need a large hacksaw to cut it.. But that will keep in impedance with quite a lot of abuse and you can bury it.

 

Cat 5e has 100 MHz bandwidth, and is compliant to 1000BaseT at the required 100 metres. You can obtain Cat 5e in shielded varieties such Cat 5e (ScTP) and (STP). In the industry I know data centres have a recommendation that for connectivity  Cat 6a or better is used, but in those environments one is sometimes going above 1000BaseT in electrically noisy environments.

 

Finally I had understood for Cat 7 to achieve its performance it could not use RJ45: EIA/TIA 568 connectors.. It required its own special connector?