It’s trivial… the multicast groups are used all over the place for loads of services including mDNS discovery, UPnP, home automation, Apple devices, Airplay, Chromecast… etc etc … the multicast group data for Roon is trivial it hardly registers in the grand scheme of thing compared regular unicast data, such as regular data transfers.
If you have a 2960 switch, like most seem to have, then your multicast group data will be managed as IGMP snooping is enabled by default I seem to remember. A consumer un managed switch will be broadcasting all this group data to everything… not ideal, again another advantage in using a Cisco Catalyst switch… that is true network optimisation…
1 Like
Respect…
1 Like
Thanks for the explanation, Simon.
I’m guessing that my ND5XS -> nDAC is not classed as decoupled?
Woah, hang on there just a second, Mike. So, you’re now saying bits-is-bits but not WAV bits? Not that I’m necessarily disagreeing with you but…
Just checked… looked at my switch port connected to my NDX2… thanks to the power of the 2960 switches…
Played from Room some 48/24 music… and looked at the multicast packet count for discovery and status compared to the unicast media transfer data count. The multicast data was 0.01% of the media unicast data … hardly heavy traffic.
My switch is set to IGMP snooping so performs multicast group control on the ports… this should be the default for the 2969 switch and will work optimally if you have a Querier on your network such as from your quality ISP router, like a BT Smart Hub 2. This ensures only the relevant multicast group data goes to your streamer and it’s not being bombarded with data from your home network, such as printer discovery, home automation, status update, etc.
Yep is it decoupled… but the first generation streamer transport clocks were not as well decoupled as the current streamers… so yes despite decoupling my NDX benefited from LPCM being sent to it when acting as a transport, probably like your ND5XS.
The improved transport architecture and significantly improved internal decoupling with the later streamers made a significant difference here, and sending FLAC has a lot less impact when streaming the output via SPDIF to a decoupled DAC on my NDX2.
1 Like
Not at all, & I don’t actually understand what you are referring to.
I know what I hear on my player on my system
I read somewhere the theory of why it might sound different with some streaming players is the players OS has less to do to process WAV.
Hi… yes I find the best lay tool for identifying RFI so you deal with it is the AM radio test… if minimal disturbance to a very closely placed untuned AM radio on MW or SW then for our purposes here one has dealt with most of the potential issues from RFI.
Just teasing 
1 Like
I notice that you have an ND5XS, and I suspect that is why you have formed the belief that WAV is ‘better’ (I assume you mean sounds better) than FLAC.
I used to have an ND5XS a long time ago and to my surprise I could hear a subtle difference in favour of WAV. However, I no longer have the ND5XS - I now have 2 main systems (Linn Klimax DS/1 and Sonore microRendu/Mytek Brooklyn+) and on these systems there is no audible difference between WAV and FLAC source files.
I continue to transcode from FLAC to WAV on my Synology NAS simply because I am too lazy to switch transcoding off, and of course my Listening to Tidal (via Roon) is carried out using FLAC source files.
I think this ‘WAV is better than FLAC’ thing may well be a legacy Naim streamer phenonemon.
1 Like
Hmack, and indeed Roon converts FLAC to LPCM
Ah - of course it does. Thanks Simon.
1 Like
What makes me laugh about this thread is that if Ethernet cables don’t’ make that much difference to the sound then why the witch hunt with AQ/Chord cables & audiophile switches?
It’s almost like folks are bitter 
2 Likes
I’ve got some Wranglers coming Monday. It will be entertaining. I hope…
1 Like
Stonewashed ???
Is that full loom or just to the ND5 ??
2 Likes
FULL loom - but with ONLY a splash of Vodka in the other corner.
1 Like
I am lost what are you referring to ?
Wranglers = a type of blue jeans.
2 Likes
Because AQ and Chord are conning people by saying
The more you pay for their cable the better the cable sounds (even though as pointed out previously: digital cables can’t have a characteristic sound of their own, they just interact with the characteristic sound of the connected player in unpredictable ways).
or
The more you pay for their cable the better will be the sound of the system despite the fact that there’s no systematic correlation between the digital cable and the resulting overall sound of the system (also as pointed out previously: digital cables can’t have a characteristic sound of their own, they just interact with the characteristic sound of the connected player in unpredictable ways).
Just because you pay more for a cable, it means nothing for the sound - a more expensive cable may make the overall sound better, but equally likely it may make the sound worse.
And the explanations given by these companies (sophisticated screening schemes and superior dielectrics affecting the ‘sound’ of the cable) are complete twaddle where digital transmission in a domestic environment is concerned.
3 Likes
I don’t think many are saying mains leads or ethernet leads don’t make a different to the audio from the connected equipment - its quite interesting how a buying a range of leads for a few pounds can make a difference…
I think there is some intrigued curiosity of why some people are spending more than 10 to 20 twenty pounds, sometimes significantly more where the same effect can be had by spending only a few pounds albeit with some trial and error - but as they say there is nought as strange as folk - and it certainly keeps up the entertainment factor of the thread you say
4 Likes