
August 1st 15, 09:31 AM
posted to uk.rec.audio
|
|
Ethernet cables matter now?
As if trying to sell people hugely expensive mains and audio cables
wasn't enough, it seems that computer cables are now fair game:
http://arstechnica.com/gadgets/2015/...no-difference/
(from comp.misc)
A vaguely interesting point is made in the test report however,
referring to cheap ethernet cable not meeting LAN spec. Quite what that
means in practice, I'm not sure.
--
Cheers, Rob
|

August 1st 15, 04:24 PM
posted to uk.rec.audio
|
|
Ethernet cables matter now?
Well over long runs if the cables internal twisted pairs have capacitor
mismatches between them then you get a non working network, but over the
couple of metres most use I'd imagine any old stuff would do.
I believe its driven at a fairly low impedance, or should be, to make it
more immune to issues like pick up or attenuation.
Most of the failures I've seen on bought leads have been poor plug
attachment or the actual connectors in the plug are crap.
Brian
--
From the Sofa of Brian Gaff Reply address is active
"RJH" wrote in message
...
As if trying to sell people hugely expensive mains and audio cables wasn't
enough, it seems that computer cables are now fair game:
http://arstechnica.com/gadgets/2015/...no-difference/
(from comp.misc)
A vaguely interesting point is made in the test report however, referring
to cheap ethernet cable not meeting LAN spec. Quite what that means in
practice, I'm not sure.
--
Cheers, Rob
|

August 1st 15, 06:09 PM
posted to uk.rec.audio
|
|
Ethernet cables matter now?
There is a difference in cables that can carry Gigabit Ethernet
(10/100/1000Mbps) as opposed to just Fast Ethernet (10/100Mbps) in
that IIRC Gigabit Ethernet requires an extra twisted pair. Thus you
find that if you use a Fast Ethernet cable to connect two Gigabit
ports, the connection will only be Fast Ethernet.
As far as any other hype is concerned, ethernet connections either
work or they don't, though it should be remembered that if a faulty
connection causes a high error rate, then a proportion of the data
sent across it will be corrupted and will have to be resent, and
therefore the net data rate across the connection will be reduced.
On Sat, 1 Aug 2015 10:31:43 +0100, RJH wrote:
A vaguely interesting point is made in the test report however,
referring to cheap ethernet cable not meeting LAN spec. Quite what that
means in practice, I'm not sure.
--
================================================== ======
Please always reply to ng as the email in this post's
header does not exist. Or use a contact address at:
http://www.macfh.co.uk/JavaJive/JavaJive.html
http://www.macfh.co.uk/Macfarlane/Macfarlane.html
|

August 1st 15, 09:37 PM
posted to uk.rec.audio
|
|
Ethernet cables matter now?
On Saturday, 1 August 2015 10:31:43 UTC+1, RJH wrote:
A vaguely interesting point is made in the test report however,
referring to cheap ethernet cable not meeting LAN spec. Quite what that
means in practice, I'm not sure.
ICBW, but copper-clad alumin(i)um conductors (CCA) aren't Cat5 or Cat6 rated, possibly because they don't work well with insulation-displacement connectors.
The latest CPC flyer quite carefully refers to Cat5 or Cat6 cables with copper conductors, but CCA cable is just called 'networking cable' without any Cat claim.
Owain
|

August 2nd 15, 11:34 PM
posted to uk.rec.audio
|
|
Ethernet cables matter now?
In article , Java Jive
scribeth thus
There is a difference in cables that can carry Gigabit Ethernet
(10/100/1000Mbps) as opposed to just Fast Ethernet (10/100Mbps) in
that IIRC Gigabit Ethernet requires an extra twisted pair. Thus you
find that if you use a Fast Ethernet cable to connect two Gigabit
ports, the connection will only be Fast Ethernet.
You sure about that?. I thought its down to the number of pairs used in
the cable, normally Green and Orange but plus the Brown and Blue in Gig
working....
As far as any other hype is concerned, ethernet connections either
work or they don't,
Not quite that simple, length plays a large part in thruput...
though it should be remembered that if a faulty
connection causes a high error rate, then a proportion of the data
sent across it will be corrupted and will have to be resent, and
therefore the net data rate across the connection will be reduced.
On Sat, 1 Aug 2015 10:31:43 +0100, RJH wrote:
A vaguely interesting point is made in the test report however,
referring to cheap ethernet cable not meeting LAN spec. Quite what that
means in practice, I'm not sure.
--
Tony Sayer
|

August 3rd 15, 12:17 AM
posted to uk.rec.audio
|
|
Ethernet cables matter now?
On Mon, 3 Aug 2015 00:34:50 +0100, tony sayer
wrote:
In article , Java Jive
scribeth thus
There is a difference in cables that can carry Gigabit Ethernet
(10/100/1000Mbps) as opposed to just Fast Ethernet (10/100Mbps) in
that IIRC Gigabit Ethernet requires an extra twisted pair. Thus you
find that if you use a Fast Ethernet cable to connect two Gigabit
ports, the connection will only be Fast Ethernet.
You sure about that?. I thought its down to the number of pairs used in
the cable, normally Green and Orange but plus the Brown and Blue in Gig
working....
Isn't that what I said, only you've given more detail?
As far as any other hype is concerned, ethernet connections either
work or they don't,
Not quite that simple, length plays a large part in thruput...
though it should be remembered that if a faulty
connection causes a high error rate, then a proportion of the data
sent across it will be corrupted and will have to be resent, and
therefore the net data rate across the connection will be reduced.
Well again, an overly-long connection would tend to behave as I
described for a flakey connection, so IMV that's six of one and
half-a-dozen of the other.
--
================================================== ======
Please always reply to ng as the email in this post's
header does not exist. Or use a contact address at:
http://www.macfh.co.uk/JavaJive/JavaJive.html
http://www.macfh.co.uk/Macfarlane/Macfarlane.html
|

August 3rd 15, 04:48 AM
posted to uk.rec.audio
|
|
Ethernet cables matter now?
On 01/08/2015 19:09, Java Jive wrote:
There is a difference in cables that can carry Gigabit Ethernet
(10/100/1000Mbps) as opposed to just Fast Ethernet (10/100Mbps) in
that IIRC Gigabit Ethernet requires an extra twisted pair. Thus you
find that if you use a Fast Ethernet cable to connect two Gigabit
ports, the connection will only be Fast Ethernet.
As far as any other hype is concerned, ethernet connections either
work or they don't,
Well, as I say, the test report suggests that certain (cheap) cables
don't work very well. Not a 'fault', just poor materials and sloppy
manufacturing.
I recently wired up my home with cat6. Nothing too fancy, just a feed to
each living room. A lesson learned for next time is to use trunking, but
also to pay a bit more attention to the quality and source of cable.
though it should be remembered that if a faulty
connection causes a high error rate, then a proportion of the data
sent across it will be corrupted and will have to be resent, and
therefore the net data rate across the connection will be reduced.
On Sat, 1 Aug 2015 10:31:43 +0100, RJH wrote:
A vaguely interesting point is made in the test report however,
referring to cheap ethernet cable not meeting LAN spec. Quite what that
means in practice, I'm not sure.
--
Cheers, Rob
|

August 3rd 15, 08:47 PM
posted to uk.rec.audio
|
|
Ethernet cables matter now?
On 03/08/2015 01:17, Java Jive wrote:
On Mon, 3 Aug 2015 00:34:50 +0100, tony
wrote:
In , Java Jive
scribeth thus
There is a difference in cables that can carry Gigabit Ethernet
(10/100/1000Mbps) as opposed to just Fast Ethernet (10/100Mbps) in
that IIRC Gigabit Ethernet requires an extra twisted pair. Thus you
find that if you use a Fast Ethernet cable to connect two Gigabit
ports, the connection will only be Fast Ethernet.
You sure about that?. I thought its down to the number of pairs used in
the cable, normally Green and Orange but plus the Brown and Blue in Gig
working....
Isn't that what I said, only you've given more detail?
It's _two_ extra pairs.
Andy
|

August 4th 15, 10:37 AM
posted to uk.rec.audio
|
|
Ethernet cables matter now?
I was going to retort: "But one in each direction, surely?" as I was
considering data moving in one direction, from server to playback
machine as in the original article I was commenting on, but, in fact,
a search seems to suggest from the wiring names that the pairs in
Gigabit cables are all bi-directional, which I didn't realise before.
For example (diagram entitled 'Ethernet Connector Pin Assignment'):
http://www.bnoack.com/index.html?htt...bles/CAT5.html
On Mon, 3 Aug 2015 21:47:25 +0100, Vir Campestris
wrote:
It's _two_ extra pairs.
--
================================================== ======
Please always reply to ng as the email in this post's
header does not exist. Or use a contact address at:
http://www.macfh.co.uk/JavaJive/JavaJive.html
http://www.macfh.co.uk/Macfarlane/Macfarlane.html
|
Thread Tools |
|
Display Modes |
Linear Mode
|
|