Meridian Unplugged - Powered by MrTechGuy.co.uk Forums Discussion Meridian - Streaming External power supply for MC200

To stress two points already made above

1) Users should look at the volumes of information out there.......and compare the Orders of Magnitude difference between RFI Levels caused by Gigabit Networks and RFI potentially caused by PSU's.........obviously RFI resultant from Networks is much, much higher

2) Unlike Jitter which "travels" from one Component to another as part of the Data Stream, RFI does not migrate from one component to another alongside the datastream

Many Users here have familiarized themselves over the years with Jitter......i.e. What is it??.....What Causes it???.....how does it affect the Sound??......how can I Minimize it??

I think Users here should take the same time to familiarize themselves with RFI / EMI........especially with what it can do, as well as what it cannot do

[pedantic]Sooloos audio endpoints are 10/100 not gigabit. Core/Store (at least recent ones) are gigabit. [/pedantic]Not sure if that's going to make a difference to any arguments.

Understood Ian, but most User's Networks these days tend to be Gigabit and these volumes of data moving at high speed is what causes the high levels of RFI / EMI which can in turn effect ALL components on the Network, be they 10, 100 or Gigabit capable.

I should add that moving data around even at 10Mbps will also cause elevated levels of RFI / EMI just that Gigabit causes even more.

But a switch by definition isolates all network ports. So core to audio endpoint will only be active for the brief moment that data is transferring between the two. Can the RFI/EMI from other inactive ports still pollute the Sooloos ports?

Strength of interference from RFI/EMI >> from SMPS on a Cat 5e/Cat 6 is an acknowledged fact.

Strength of filtering on a Sooloos endpoint would seem to be better than that on a Meridian PSU, even in the 861v6.

That is my understanding; and it's on this basis that Meridian designed those special Cat5 Cables with Ferrite Pods for the Ruby Anniversary System.

Likewise the rationale behind the ID41 on the 818.

Please let me add this:

My initial Sooloos network configuration was reserved IP addresses by MAC address within the routers DHCP address range for each Sooloos component. It has been this way for years with no issues.

Prompted by reading some threads here, I decided to change the Sooloos components to fixed IP addresses outside of the routers DHCP range. To my total amazement the system sounder better..a lot better. My wife even asked what I did, it was that much improved. More width, separation, tighter bass. This was listening at the exact same volume level before and after. No other changes.

I know Ronnie and Ian will scoff at this but it was a eye opener.

I don't think scoff is the right work. Be highly sceptical yes.

I have freely admitted in the past that I do not understand the dark arts of audio. I have owned one piece of audio equipment where I could hear the difference in changing a digital cable, but never a mains cable. It was Meridian 200 series. Since then, I have never heard a fully functional digital (nor mains) cable make an audible difference.

Yes. I am a cable sceptic.

Extrapolating this out, until I hear a difference due to changes in network equipment and cabling, I too will be a network sceptic. There is no scientific reason why fixed IP will sound any different to dynamic IP.

However, I do have a scientific background, so do consider all alternatives and when I either hear a difference for myself, or read a discussion that is based on scientific fact, I will reconsider my position.

About 16 years ago, I circumnavigated the globe. I am a firm believer that the world is not flat and maybe one day I will be a firm believer that different network topologies, even when correctly designed to Meridians recommended specifications and reliably built, may have an audible impact. But not today.

From my limited knowledge of RFI/EMC/EMI issues I know 'noise' can be conducted on cables between equipment. I also know that EM waves are magic and resonance can amplify any small noise/signals picked up by the system.

This is not the same as noise being carried by the signal.

Of course a properly designed cable/system should minimise all of these issues. Meridian has already demonstrated the system is possibly susceptible to these issues by using ferrites.

But I do believe these effects are small and not likely to be relevant to Crion's and Nick's observations.

What I don't understand is what is actually packed in the UDP packets and so I wonder if that info could be affected the core having a improved power (and therefore lead us into the world of cores sounding different).

Chris

G997, the only thing I can think of is it adds additional processing to EVERY UDP packet on reserved IP but not static IP. But that would just be a crazy router implementation. Like it did SPI just on reserved but not static IP's

Also I would probably recommend having a separate Gbit switch for your Sooloos stuff and just an uplink to your regular router. That should take care of that weird behaviour.

I do have a separate Gbit switch...

Change it to 10/100 and see if that reduces EMI. Or as a test, create a playlist, start it playing then remove switch altogether and replace with single crossover cable. You won't be able to control Sooloos except via msr but it should remove any external network interference.