Roon adds SQ

I thought you were living in Paris….

Thanks for asking. I would say remote streaming Qobuz first with Roon: ( through PN without Statement, vs, through Statement without PN).
In any case if the result is different with the Naim app it will be good to know just in case…

Work in Seoul, family vacations in Paris…

1 Like

Ok. I’ve done a little messing around in between juggling various client panics. I was expecting to be fairly clear that most of the benefit with Qobuz comes from the PhoenixNet, but now I’m not so sure. The Statement makes more of a contribution than I had expected. Here’s what I wrote in reply to a similar question by Gazza, having done a little comparative listening for both of you: ND555 Current Wisdom - Simple set up

1 Like

Ok in your post you mentioned the greatest bang for the buck with the Phoenix net. Since it is also less expensive than the Statement, this is probably where I will go…it will at least free up space and I will repurpose the older stuff!

But my greater issue is that ideally I would like to buy these boxes or at least one of them from NAIM!
Any chance they would come up with something? Basically Naim is losing a piece of the sound supply chain to new players and that is a pity.

I do not think Naim is losing anything, the only thing that Naim need to do is to bring up the Naim streamers to the next level of performance (if needed). Innuos and the likes just capitalize some weaknesses of the current lineup of the Naim streamers, these won’t be needed if Naim can incorporate some better re-clocking technologies, and / or bring in optical ethernet, USB input with high memory cache.

Yes. So it looks like we are saying the same thing?

Nope, your comment carries some pessimistic tone :). Maybe it is just my own understanding?
I am sure Naim R&D must be on something. It seems unreasonable to me to spend $$,$$$ on a streamer and then $$,$$$ on some re-clocking devices.

It is not pessimistic. I would say it is sad to see this piece of business go somewhere else just because Naim have a gap in their product offering; I really would prefer to buy a Naim box, if only they would share their plans.

From a SQ point of view Roon is transparent (assuming you are not processing the audio such as via eq) just like UPnP streaming PCM. However local implementations regarding connectivity and parameters / configuration used in underlying network transport protocols between server and audio player host might provide a sound signature through differences between extremely low level noise interaction.

If your system is playing in an environment where you experience speaker reflections and certain filtering effects in higher frequencies… possibly likely in most un treated domestic listening environments, you may be more susceptible to disproportionately noticing these sonic profile changes. Some people seem to confuse this with overall system ‘SQ’… thinking sometimes certain changes in the audio chain ‘improves SQ’… where in fact it is simply shifting the sonic profile. You see it a lot when some people exclaim ‘firmware X produces better SQ’ for example… where as in reality all that has happened is the resultant low level noise profile has changed its characteristic.

Roon streaming Tidal and/or Qobuz will likely result in a lower noise floor generated in your audio host as the audio being rendered is uncompressed PCM compared to natively streamed FLAC which requires extra host processing to decompress.

5 Likes

I think that makes sense, but I’ll try and find the time for a quick naked 555 vs 555+Statement vs 555+PhoenixNet comparison today and let you know. Was hoping to manage it yesterday but work was too hectic

1 Like

I always went with that with whatever NAS I was using and then later with Roon, setting them to do the decompression. AFAIK the Innuos servers don’t decompress FLAC files, however, so the ND555 has to do that work but it still sounds better…

Sure, whatever sounds best with you. A myth on many Hi-Fi forum is all noise must be bad, some digital and analogue noise when very low level can be advantageous sonically from a preference point of view. In music production adding low level analogue or digital noise is a technique that can be used to make a track sound ‘better’ or more appealing. It tends to be anharmonic noise or unnatural resonances we find more objectionable.

1 Like

I think it could simply be that the lower noise levels overall are much greater in difference than the “added noise” of getting the ND555 to do the decompression…

Sure, the ND555 is a very accomplished player… I suspect the ‘noise’ matter is a red herring.
Likely the digital to analogue reconstruction between your players is very different.
Remember D to A is a lossy process and many design decisions on which compromises to use are made in a renderer. For example Naim Audio and Chord Electronics go about analogue reconstruction quite differently… and the resultant audio has slightly different characteristics. The more detailed and lower analogue noise floor the audio chain is, the more you can notice and appreciate these design considerations.

2 Likes

That sounds fantastic. Looking forward to your feedback. Hopefully it comes out supporting the lowest cost option (Phoenix Net)

I’ve done a little more faffing around for you and basically the results surprised me a bit (again!), though on reflection there’s no reason why that should have been the case.

Reverting to the ND555 playing Qobuz straight from the router and its far from shabby (nice to be reminded of that!): a bit more indistinct and fuzzy around the edges, with less tonal and textural complexity and a smaller soundstage with less presence overall.

Add the PhoenixNet and Statement into the mix separately and you basically find they do different things.

I tried the PhoenixNet first in each of these comparisons and it did what I expected: opening out the soundstage, giving each instrument more presence and space of its own, everything was more palpable and grabbed me more. The Goldfrapp Felt Mountain was more dramatic and bigger, with greater sweep, “pop” and drama. There was also increased more bass extension.

Similarly, the Haitink LSO HD recording of Beethoven’s Pastorale, arrayed the orchestra more convincingly and precisely across the room, the bass and cello underpinning had greater presence and vibrancy, the orchestra “breathed” more convincingly, had more life. Detail was enhanced by the lower noise floor.

A big test for me is how systems handle the young Schwarzkopf’s voice on the Strauss Four Last Songs, where it is captured beautiful despite the age of the recording, which shows in almost every other area. Even here the PhoenixNet worked enhancements, giving her more space, presence and above all bringing out the breath and body in her younger voice.

So, the values of the PhoenixNet were pretty clear and the bang for buck argument looked very straightforward. However , I listened to each piece first with the PhoenixNet and then with the Statement.

Well… switching the Goldfrapp from ND555+PhoenixNet to ND555+Statement surprised me. Firstly, a lot of the room-filling soundstage was there too - less expansive, with less separation of performers, but still a noticeable difference. Fair enough, I thought. But then listening in, I realised a lot more was happening: detail of tone and texture was richer and more complex, more rounded, more real, more dynamic but also more subtle with greater richness of microcontrast. This was clearly not going to be simple at all!

The differences were equally clear with the Beethoven. Whereas the PhoenixNet made an impressive impact it was in many ways an impact of scale, whereas the Statement brought insight and a sense of organic reality - in hi-fi terms, there was more analogue flow and integrity as well, less digital hash.

The Schwarzkopf confirmed this, with the Statement bringing her very rich and complex but unmannered young voice more convincingly and movingly to life.

If I had to choose to add just one to a bare ND555, it would at this point be the Statement. The PhoenixNet is wonderful but the Statement is transformative.

So, yes, the PhoenixNet wins on bang for the buck, but the bangs simply aren’t comparable…

Sadly, I realise that doesn’t help at all. As always, audition first if you can…

7 Likes

Thanks a lot for this insight, very very useful. Now since I already have an ER with external clock and good power supply, it would make even more sense for me then to go the Statement and skip the phoenix net (for now)…

…but I would have to pay for all the USB part that I do not need since I only use ethernet streaming…can innuos launch a “pure streaming” statement, please?