Mosaic-GUI < > Roon

Agree with you there! And just so no one misunderstands me, I would never tell another human how to seek their own happiness. If Roon—or anything else for that matter—isn’t satisfying someone, no amount of what my granddaddy called “splaining” will change that. Follow your heart. (Though curmudgeons like me will always ask “why the heck did you do that?” :wink:)

1 Like

Well, the Rosinni’s streaming board has a pretty powerful ARM Cortex A8 CPU which has plenty of computing headroom. Neither Roon/RAAT nor Mosaic/UPnP (or even FLAC decompression for that matter) is going to “stress” it at all :smile:

In a proper setup, meaning if there’re no bottlenecks on your server platforms/networking, then objectively, streaming via Roon and Mosaic will sound identical.

1 Like

Hmm, that means there is no difference btw. a standard Qnap vs Melco or standard LAN cable vs. Tellurium Q or standard switch vs. Ansuz Power.
I think my set-up for streaming is pretty optimized…Melco, dedicated roon core+Sboost PSU+Cisco 2960+Tellurium Streaming cable.
Anyway, I hear what I hear and I checked some other forums- many claim that roon has an impact on SQ.

There is lots about ethernet cables in other threads (try the search facility) . The main point re-iterated often is that for use with dCS equipment the cable should be CatX certified and unshielded. In that list I think that all or most of them are shielded which could make an audible difference.

1 Like

We all hear what we hear. But this industry is full of induced audiophoolish paranoia. I’ll put my stock in my own ears and the people who build the $36k DAC I bought, rather than the people trying to convince me that dCS doesn’t know how to optimize Ethernet playback. There remains plenty to learn about digital musical reproduction, I am sure, but that doesn’t mean we should stop using our brains.

Still waiting for that rational hypothesis for why compliant cables and switches should sound different into a buffered input of asynchronous data. :cricket:

The funny part about this test is that people say it‘s impossible to audit any difference though they didn‘t demo any switches in a/b tests, nor cable nor nas systems.

This is a report of a live and blind test of 7 switches for streaming audio:

Thanks Erno, that matches my findings as well. I could easily hear the differences btw. switches.
I tested cheap dlink, Cisco 2960 and Melco S100 with Naim ND555 and Rossini.
My winner was the Cisco, the S100 adds too much “edge/harsh”-ness in music flow.
Naim uses an advanced (async) buffer as well- but even here the differences are notable.
Anyway…my question simply was: will there be an improved GUI for Mosaic :slight_smile:

Am I reading this right? :face_with_monocle:

For the test we pulled a – fully solid core, copper – SFP-CAT6a cable of about 16 meters from the server room to the Metrum Acoustics Ambre. Each switch to be tested is connected to a Netgear GS724T v3. This switch is connected via fiber to our core switch.

Does that mean?

Server --[core switch] --[Fibre]-- Netgear GS724T Switch -- Test-Switch --[Cat6A]-- Metrum Streamer -- [i2s] -- Metrum DAC -- Bryston PreAmp -- PassLab Amp -- Focal Speakers

Looks like such a convoluted setup!! :rofl: :thinking:

1 Like

Test setup looks like this. Necessarily convoluted…

Not really. If they wanted to really know if the Switch contributes to a sonic differences, keeping it super simple would have been more ideal. Plus why a Raspberry Pi based i2s streamer of all things?? (And after going through so much trouble, why couldn’t they include a basic physical connectivity diagram? :roll_eyes:)

1 Like

I guess the blind testing needed the possibility of rapid changing of switches. This test succeeded, as far as the same original datastream is offered to the whole set of tested streamers, so differences between them were tested, not so much if a switch contributes to sonic differences.

Spoiler: Conclusion: yes… a switch does make audible difference. However, the differences are small. The basis of the set and the acoustics must be right. Otherwise it’s a shame to invest so much money. But the fact is: it is audible. That’s what the data in this test shows.

Thanks for sharing that. It’s intriguing. I applaud their efforts if not their methodology. Purely from a statistical point of view, how seriously do you want a skeptic to take this?

1 Like

Not :grinning:

1 Like

That’s funny.

I agree. The GUI could be better. I am not familiar with Roon but used Aurender’s Conductor for 3 years. It would be nice if Mosaic could use the whole screen with icons in ‘browsing’ mode like Conductor.

1 Like

I can confirm. Upnp SQ is better than RAAT.

2 Likes

May I ask how you were able to confirm that? :grin:

1 Like