Rerouted Ethernet Cable

nigelb posted:

.... Also we are all capable of convincing ourselves to varying degrees that something is better (define better!), particularly when we have invested a fair amount of dosh in the 'new best thing'....

Interesting though!

except in my case, right now, i definitely prefer the much, much cheaper Maplin cables than any of the other more expensive ethernet cables that i have tried so far -- and i have tried a few. But i the case of Meicord -- i think a true test is hampered by the fact that i havent found a good route for it to the NDS -- if i do, i will give it a good run...

but i agree with the general drift of your post -- that we of course hear differently -- though i suspect there is probably quite a significant overlap in the way we perceive 'better' given that we seem to like naim kit anyhow...

enjoy...

ken

ps: as a matter of interest -- i emailed Nigel Finn of Chord about this whole issue os shielded vs unshielded. a rather significant part of his response was that they are looking at fibre seriously...

Ken, fibre is straightforward (especially in low speed use) and the cables themselves are cheaper... A lot cheaper than many so called audiophile copper cables... The only downside is that not much consumer kit currently supports fibre.. But that might change of course. In the commercial world many devices use GBIC type sockets which allow either fibre or copper transceivers to be inserted as required... as fibre or copper Ethernet is identical... I would not be suprised to see similar GBIC type sockets coming to serious consumer hifi soon... It makes sense.. and all talk of cable RFI,, common mode noise, cross talk, shield ground loops is gone in a stroke  However such is the way of the audiophile community and market, I guess there will be 'premiums' on special 'audio grade' GBIC transceivers.. which are modular and might be treated like the turntable cartridge.... 

Simon 

Mike, I liked your post, I see  your thinking and understanding has evolved on cables and I think you and I agree on most of this now 

To my mind a lot of what we 'hear' with Ethernet and to a lesser extent SPDIF and USB cables is the effect of coupling and reflections caused by cables and terminations acting as imperfect transmission lines. This in turn causes a degree of EM modulation on the transmission line source. This would account for the variability that is observed and why many different cable geometries and designs cause an audible change...  The digital abstraction aspect of this is the side show, it is the fundamental physical analogue voltages and frequencies which contribute to this.

Simon

Mike, Simon,

+1

With one additional caveat, digital cables can also carry RFI into a device
(and hence ferrites are useful as a protection against CM RFI, even when the better construction helps to improve rejection of DM RFI).

Between these various effects, I believe we can account for at least the majority of the sound differences
a)  between cables in any given installation
b)  the variability of what appears to be the best cable for similar hardware arrangements installed at different locations.

Simon-in-Suffolk posted:

I see  your thinking and understanding has evolved on cables and I think you and I agree on most of this now

Indeed Simon, I was struggling with digital - bits is bits -  I think the light bulb moment came when I read up on Manchester Code & understood this is volts & frequencies, it has resistance, impedance & phase angles with lag/lead are in play here. It might be called skew & propagation delay, return loss & attenuation, but I can relate.  Crosstalk & all the fancy named NEXT derivatives  is simple mutual induction.   Now I am able to envision how sub-optimal connections & cable damage can unbalance the relationship between the twisted pairs & why it can affect data transmission - & maybe SQ.

Mike,  Food for thought...  Too much bandwidth can be a problem in digital transmission systems.  If a transmission system has too much bandwidth, it can carry higher levels of energy above the bit rate frequency.  This energy can appear as as transient voltages above and below the digital transition thresholds, causing false transitions which then have to be removed (if you're interested, Schmidt triggers followed by monostables are one technique).

We live in an analogue world: The only real physically digital systems are those of quantum mechanics, and that gets even more bizarre!

Simon-in-Suffolk posted:

Ken, fibre is straightforward (especially in low speed use) and the cables themselves are cheaper... A lot cheaper than many so called audiophile copper cables... The only downside is that not much consumer kit currently supports fibre.. But that might change of course. In the commercial world many devices use GBIC type sockets which allow either fibre or copper transceivers to be inserted as required... as fibre or copper Ethernet is identical... I would not be suprised to see similar GBIC type sockets coming to serious consumer hifi soon... It makes sense.. and all talk of cable RFI,, common mode noise, cross talk, shield ground loops is gone in a stroke  However such is the way of the audiophile community and market, I guess there will be 'premiums' on special 'audio grade' GBIC transceivers.. which are modular and might be treated like the turntable cartridge.... 

Simon 

Simon, many thanks.

it does sound like fibre will be a real saviour  for reasons that have been the subject of a lot of threads here --- and i would not be surprised if this becomes the norm in due course -- that would be absolutely brilliant !

other than the fact that it needs different connections/interfaces -- are there any downsides to fibre compared to the copper cables that we know and 'love'? otherwise come the time, i would be very happy to junk all my ethernet cables...

enjoy...

ken