Bluesound NODE2 to connect to Phantom

[Johnnydev]

I can't understand this.

I use 2x silver,1x dialog the remote, quobuz, MacBook pro with all my music and my tv connected by optical to one of the phantoms.
only my dialog have an internet cable from my netgear r7000 router and the rest functions complete wireless.

I use only spark from all my devices without any problem.

No any other problem at all and the best sound i ever had

Thanks for everyone's input - you inspired me to keep trying.

Just been playing and gone back to an earlier configuration of Netgear r7000 connecting wirelessly to Bluesound NODE2 which is connected to Devialet Dialog via Optical, Dialog then talking to Phantom and it's now working - not really sure why as it did not first time round. Only difference was my Netgear was on 1.0.3.x and we upgraded it to 1.0.6.x, or was it just user error in set-up???

I did initially get it running with Optical from the NODE2 directly to Phantom but had to control volume from Spark and tracks from NODE2.

With my final configuration (via the Dialog) you still need to set the volume to full (or as near as your dare) on Spark (can then shut it down and don't need it running in the background) when NODE2 audio settings is set to variable, which allows you to control the volume from NODE2, also works on the Android app :-)

Why do you want to use a NODE2 or even a sonos???
It sounds the best on the way devialet has designed it
I have also used a sonos to the dialog in the beginning, but without it sounds better

[MrSinister]

My node is for my wife. She hates the way spark works and just wants to listen to some radio or music for the little one. And control it on the device and not having to look for her ipad, unlock it, start spark etc...i.e. when the phone rings.
This is what the node does for her. When I listen I use spark.

[Gremlin]

The streaming device, whether Sonos or Bluesound, plays back the original lossless stream, in PCM, say 192/24.
If output through TOSLINK optical, the PCM stream will arrive (possibly with some added jitter) as is.
Variable volume control _before_ sending output through TOSLINK optical connection means that the stream is going to be processed in the Sonos/Bluesound device to apply the digital volume control. As the Phantom still expects 24bit signal, there is no other way to reduce volume than to simply send zeroes in some of the bits. So every time you reduce volume by 3dB you lose 1bit of accuracy. At -30dB you will have a signal of 14bit from source 24. If original was 16bit, you will have 6bits of resolution left.
There is no way around this without controlling the volume on the final DAC (inside Phantom).
TLDR: Digital volume control over optical output (on Sonos/Bluesound) is just applying negative gain and reducing bit-resolution, by 1bit per each -3dB.

I'm sorry, but you are simple wrong here. Reduction in bits does not mean reduction in resolution (dither aside). Rather it means reduction of dynamic range; exactly what you want if you wish to reduce the loudness.

Imagine a singer playing a guitar. One (guitar or singer) is going to be louder than the other and will, without compression, require more bits. But both will still be perfectly resolved.

And why do you imagine that a DAC will be any better at this than anything else?

[iliapas]

Hi, Gremlin.

For your example of guitar and singer: this is why many times you record them separately, each with different gain, such that you get maximal dynamic range captured. What's the point in recording in 24bit if you're only going to use 10bits and zeros in the rest?
This of it this way: even if you recorded a singer shouting right into the microphone and play at a much lower volume at home, the same dynamic range can be reproduced, mapped into your new range of sound pressures. If your equipment is good enough, even at very low volume you will be able to distinguish the detail of the original recording.

The difference between Phantom getting a 12bit stream and a 24bit stream is that in the same volume level, you are reconstructing the original audio stream either from a 12bit or 24bit data. The fact that this extra data is lost on the way to Phantom means it is not receiving the extra data that would allow it to reconstruct the sound to a higher accuracy.
Whether that makes a difference for you personally is a different question, and is better suited to some kind of a "Is there a difference between 16bit and 24bit audio" thread.
The Phantom is able to provide you with highly detailed playback even at lower levels.

[Gremlin]

Hi, Gremlin.

For your example of guitar and singer: this is why many times you record them separately, each with different gain, such that you get maximal dynamic range captured. What's the point in recording in 24bit if you're only going to use 10bits and zeros in the rest?
This of it this way: even if you recorded a singer shouting right into the microphone and play at a much lower volume at home, the same dynamic range can be reproduced, mapped into your new range of sound pressures. If your equipment is good enough, even at very low volume you will be able to distinguish the detail of the original recording.

The difference between Phantom getting a 12bit stream and a 24bit stream is that in the same volume level, you are reconstructing the original audio stream either from a 12bit or 24bit data. The fact that this extra data is lost on the way to Phantom means it is not receiving the extra data that would allow it to reconstruct the sound to a higher accuracy.
Whether that makes a difference for you personally is a different question, and is better suited to some kind of a "Is there a difference between 16bit and 24bit audio" thread.
The Phantom is able to provide you with highly detailed playback even at lower levels.

Wrong again, I'm afraid.

It matters not one jot whether it is the Phantom or the provided data stream that reduces or increases the loudness. There is no loss of data, just manipulated data, and at which end the manipulation happens is of no consequence to anyone. And there is certainly no loss of information that any human without special powers would be able to discern.

And however the recording is made, the master will have some sounds louder than others. Do you really think that an orchestra playing a crescendo is better reproduced at the noisy bit rather than the quieter bit? Because, if so, you would be obviously wrong.

And as I said before, the number of bits involved has nothing to do with resolution, only dynamic range. The sampling frequency is what determines resolution.

[Johnnydev]

My node is for my wife. She hates the way spark works and just wants to listen to some radio or music for the little one. And control it on the device and not having to look for her ipad, unlock it, start spark etc...i.e. when the phone rings.
This is what the node does for her. When I listen I use spark.

That was overhere the same with my wife . My advice:
Buy the devialet remote
And listen to the radio via TV ( waiting for the devialet internetradio upgrade )

[Pleyel]

a) every time you reduce volume by 3dB you lose 1bit

At -30dB
b) you will have a signal of 14bit from source 24.
c) If original was 16bit, you will have 6bits of resolution left.

a) I think you rather loose 1 bit every 6 dB attenuation, not 3 dB.

When you send a 16-bit material to a 24-bit DAC, you are supposed to use its whole dynamic, hence add zeros as Least Significant Bits, not Most. And these LSB are precisely those that you drop when applying attenuation.
So dithering aside, at -30 dB into a 24-bit DAC, you miss 30/6 = 5 bits, and there are still 19 bits of resolution sent to the DAC:
b) 5 bits less than a 24-bit material;
c) no lost bit from a 16-bit source (still padded with 3 inactive LSB).

You can play down to -48 dB and loose no information from a 16-bit CD.
Right?

[MrSinister]

My node is for my wife. She hates the way spark works and just wants to listen to some radio or music for the little one. And control it on the device and not having to look for her ipad, unlock it, start spark etc...i.e. when the phone rings.
This is what the node does for her. When I listen I use spark.

That was overhere the same with my wife . My advice:
Buy the devialet remote
And listen to the radio via TV ( waiting for the devialet internetradio upgrade )

Mine are early phantoms without bluetooth i'm afraid.

[Johnnydev]

My node is for my wife. She hates the way spark works and just wants to listen to some radio or music for the little one. And control it on the device and not having to look for her ipad, unlock it, start spark etc...i.e. when the phone rings.
This is what the node does for her. When I listen I use spark.

That was overhere the same with my wife . My advice:
Buy the devialet remote
And listen to the radio via TV ( waiting for the devialet internetradio upgrade )

Mine are early phantoms without bluetooth i'm afraid.

Mine are also without Bluetooth !!!

[MrSinister]

Ok, so it doenst rely on bluetooth. Did not look at it because it is a little to basic with only volume control.