Meridian Unplugged - Powered by MrTechGuy.co.uk Forums Discussion Meridian - Streaming LAN isolator anyone?

Ok, so I'm going to give some early thoughts on the GISO that arrived yesterday.

Does it improve the SQ? In terms of imaging (and in my opinion), yes, but nothing like as much as the step from a MC200 to an 818v2, which is hardly surprising given the GISO costs �200 and the 818 costs �8500! Of course the 818v2 is meant to be pretty well isolated from interference anyway.

If the 818v2 improved the SQ by a (notional) 200% (and one could rightly argue it does a whole lot more than this), then I would say the GISO probably increases this by another 10% (again, remembering this is on an 818v2). The dilemma I have is that, potentially, 10% is in the realms of human error and my own mood, which I've tried to eliminate by a/b testing - with the result that I feel the imaging has become more solid and I would still say it gives a 10% (perhaps 15%) improvement.

I have a more listening to do, but, if I feel the same way tomorrow, that would still make it a very worthwhile investment.


[side notes: Given what Meridian have said about magnetic isolation and the changes from the ID40 to an ID41 card, its use on an 818v1 might show greater results.

When unplugging the ethernet cable from the 818 during a/b testing, I found the music stopped less than half a second after the ethernet cable was unplugged

Just out of interest, the GISO sounds much better in my system with the earth side connected to the incoming cable, not to the cable going out to my 861. Is that the finding of others?

I will refrain from commenting about my experience of the GISO until my ears have burnt in further!

If the improvement in SQ depends on the GISO removing gubbins from the LAN, doesn't the improvement depend on how much gubbins was there in the first place?

I'm sure it does Tintin.

Spurred on by this thread, I've ordered, like ChrisLayer, a couple of S/FTP Cat6A cables. I'm using cheap free stuff at the moment, so have nothing to lose.

Same with my 861.

Shielded cables could cause the problem that the 'Isolator' makes go away.

If shielding is really necessary in the listening environment - it's probably not an environment I would want to be listening in - I am assuming you have no 3.3Kv / 11KV multi-Megawatt industrial motors or RADAR installations in your home.

The twisted pair part of the cable gives you noise immunity / cancellation, the ethernet endpoints give you magnetic isolation.

Adding shields into the mix brings the ground loop problem back - depending on how you split the earth will then determine whether you can get back to the unshielded performance. (Since you have added a massive antenna that defeats the point of twisted-pair)

Slapping in a plastic RJ-45 coupler (no metal tabs in the socket side) should give you the same effect as the 'heavily promoted' more expensive option when it comes to the (artificially introduced) ground problem.

The majority of the content running down the cable will be data - if the isolator is making a difference on shielded cable (which already has an isolator either end) - other than the ground loop, something is very very wrong.

To prove the isolator blind tests would be needed between a ground-break coupler and the isolator - not a test between an additional isolator vs built-in isolation - otherwise you are making multiple changes at the same time (fully grounded vs partial ground+external isolator). Presumably using some form of clamping mechanism to ensure the head+ears go back to exactly the same point in the room each time.


Matt.

And yet, my system sounds better, certainly no worse with shielded cable? Go figure...

Is the incoming cable connected to a grounded jack? I'm finding the opposite, mine sounds better with the grounded side connected to the MS600 than my Powerline adaptor. But I don't think the Powerline adaptor jack is grounded - it's just plastic.

I assume you're aware that Powerline adds a very significant amount of RFI / EMI to data transmissions... which is hardly surprising given the environment that the data has to pass thru??

My MC200 is in my wife's office and is connected to my office with Cat5e UTP running through conduit along the exterior wall of the house. I'm running Cat6 STP from where the Cat5e terminates in my office into a keystone on that wall to an unmanaged 8 port switch. I'm running Cat6 STP from an adjacent router/Modem to that switch and Cat6 STP from that switch to the 818v2. System sounds very good IMHO.

After following this thread I ordered and installed Cat7 SSTP from the router/Modem to the switch and could hear no change. I then installed Cat7 SSTP from the switch to the 818v2 and same result--no change. Based on all this I'm not inclined to try a GISO. Am I missing anything here?

Ken

Hmmm!
When I started reading this thread I was confused. However I am now confused at a much higher level. Can I be so bold as to suggest that Galvanic Isolation is either achieved or not in the dedicated network switch installed?
Hector