Coax RCA output better than USB?

S

skbe

New member
Joined
Nov 14, 2019
Messages
177
Location
France
Hey

Need to make sure of my configuration and audio skills
I have my lumin u1 mini connected to my Marantz nd8006 DAC then to my Sony z1r headphones
Using Roon / Meshell Ndegeocello - The spirit of music jamia (CD from Tidal)

I tried :
1- Coax output from lumin to my marantz with QED performance coax cable
I felt the sound is nice, clear, and a very nice crispy feeling with drums, bass and cymbals

2- USB output using Audioquest carbon usb cable
Very clear, clean and precise but struggled to find that crispy feelings with instruments. Felt like the sound is too smooth, too polite, no noise at all. Something I cannot describe but making music subtly less explosive
But the thing is that with usb I can upsample music to Dsd and than I must admit the scene is larger, more open and spatial with Dsd. I like how Roon upsamples the music

Is this normal? Is it logical to prefer coax rather than usb?
I am asking myself these questions because coax limits bit depth and sample rate. So is there a way to make music more crispy with usb? Is it a limitation of hardware stuff?

thanks for help
 
It always depends on the DAC and streamer as to which you may prefer especially if cables are a different brand and/or model.
 
There is no golden rule. It all depends on implementation, both on the transport and DAC end.
 
There are a couple of big technical differences between coaxial and USB that will influence the sound. With S/PDIF coaxial, the clock information is embedded in the data stream itself. So the source makes a big difference in terms of how clean the clock information is. The ability to extract and use this clock information is also very dependent on your DAC. Some are better than others at cleaning up the incoming data stream. In this case both the source and the DAC make a difference.

USB audio is “asynchronous isochronous”, meaning that the source sends its data to the DAC and the DAC receives and temporarily stores the data. The DAC then retrieves the data and processes it independent of the sending device. The DAC and it’s internal clock totally determine the quality of the end result.

So you will see different end results using S/PDIF and USB. One is not better than the other and a little experimentation will lead to the result you like the best using your current set of gear.
 
There are a couple of big technical differences between coaxial and USB that will influence the sound. With S/PDIF coaxial, the clock information is embedded in the data stream itself. So the source makes a big difference in terms of how clean the clock information is. The ability to extract and use this clock information is also very dependent on your DAC. Some are better than others at cleaning up the incoming data stream. In this case both the source and the DAC make a difference.

USB audio is “asynchronous isochronous”, meaning that the source sends its data to the DAC and the DAC receives and temporarily stores the data. The DAC then retrieves the data and processes it independent of the sending device. The DAC and it’s internal clock totally determine the quality of the end result.

So you will see different end results using S/PDIF and USB. One is not better than the other and a little experimentation will lead to the result you like the best using your current set of gear.
 
With my mini into my bryston BDA3 which has a very good USB input AES was dramatically better. AES probably much like spdif will sound more analog IMO. I would think this has to do with the outputs on the mini and where the engineers put the effort and parts in.
 
Tom has pretty much nailed the answer this question.

Just a couple things I'll add: the USB receiver in the DAC will play a role, in particular the galvanic isolation of the USB receiver will influence the sound quality.

The design, construction, and implementation of the USB cable matters quite a bit, too.
 
Tom has pretty much nailed the answer to this question.

Just a couple things I'll add: the USB receiver in the DAC will play a role, in particular the galvanic isolation of the USB receiver will influence the sound quality.

The design, construction, and implementation of the USB cable matters quite a bit, too.
 
W9TR said:
There are a couple of big technical differences between coaxial and USB that will influence the sound. With S/PDIF coaxial, the clock information is embedded in the data stream itself. So the source makes a big difference in terms of how clean the clock information is. The ability to extract and use this clock information is also very dependent on your DAC. Some are better than others at cleaning up the incoming data stream. In this case both the source and the DAC make a difference.

USB audio is “asynchronous isochronous”, meaning that the source sends its data to the DAC and the DAC receives and temporarily stores the data. The DAC then retrieves the data and processes it independent of the sending device. The DAC and it’s internal clock totally determine the quality of the end result.

So you will see different end results using S/PDIF and USB. One is not better than the other and a little experimentation will lead to the result you like the best using your current set of gear.
Click to expand...

Thank you very much
Indeed I started to read many threads about this

My understanding so far is that USB works on a pull mode: the DAC pulls the signal from the streamer and the streamer pulls it from internet on-demand => so the DAC handles the info as you explained

With SPDIF including COAX, my understanding is not 100% clear yet: you say that the streamer pulls the data from internet on-demand and then pushes it into the DAC using its own clock and then the DAC handles the info using its own clock? So both the streamer and the DAC have great roles to clock the signal, right? And then, good compatibility between the 2 clocks is crucial for the audio signal
 
A standard DAC implementation would be slave to the embedded clock from COAX SPDIF and therefore is usually heavily dependent on quality of the SPDIF source clock and coaxial cable (due to interference).
 
Thank you
I also read this thread which is a very important source of info
https://community.roonlabs.com/t/raat-and-clock-ownership/6915/27

How can I know how my DAC handles spdif audio in? - is it a slave to the clocked spdif signal or is it handling it?
I have 2 DACs Marantz ND8006 and Sony TAZ1HES

And then after these explanations, should I conclude that my feeling about more crispy, realistic music coming from coax compared to USB comes from: the ND8006 USB clock is less efficient/worse than the clock pushing the signal from the Lumin?

Sorry for the lack of expertise in the words I use
 
Outside of marketing materials, typically we cannot know how it is handled. With USB it's even more complex as it's no longer just a clock thing. Ultimately, it's your preference that's most important.
 
Thanks Peter
In the Lumin u1 mini, supposing listening via usb only, no other digital connection open, the other USB port unused, is there a hardware advice for using usb 1 rather than usb 2 or the contrary? Are the 2 ports exactly the same?
 
skbe said:
How can I know how my DAC handles spdif audio in? - is it a slave to the clocked spdif signal or is it handling it?
Click to expand...
Unless the manufacturer has made it public, you would need to ask them or open up the device to try and figure it out yourself. All S/PDIF signals have to be clocked by the receiver, so to speak, as that's part of the S/PDIF protocol. It's what happens afterwards that can differ.

Old DACs, or DACs that don't care as much, will forward the S/PDIF "clock" through to the DAC chip clock. The DAC I designed uses the Wolfson WM8804 to sort-of buffer and correct any incoming clock drift within a window of tolerance. A more complicated receiver could buffer longer for greater tolerance, and correct even greater amounts of clock drift.
 
Hmmm sorry too complicated for me
I just understood in the Roon community thread I posted above that Spdif is clocked by the streamer, not by the DAC, the DAC only enables buffer in order to adjust to the clock of the steamer... but I may be mistaken as this begins to be abstract to me
Anyways here is my DAC open - I don’t have the confidence to do it my self

https://www.qobuz.com/ie-en/info/hi-fi/bancs-d-essai/marantz-nd8006-the-exclusive180332

They say this:
All the digital inputs are gathered on the same card (under it the power supply is located, which we couldn’t access), starting with the optical and coaxial S/PDIF inputs handled by a Burr-Brown PCM9211 (24-Bit/192 kHz) Digital Audio Interface Transceiver, which controls their commutation. It also manages the signals coming from the Bluetooth receiver integrated to the Wi-Fi card standing on the circuits thanks to some small columns. Just on top of the Wi-Fi card a Toshiba TC94A92FG CD servo controller is located, which also decodes compressed audio (Bluetooth). On the left of the picture, you can see the CD playback control mechanism, a Rohm BA5810FM.

The USB interface calls upon a XMOS 8U7C10 processor while the decoding of the signals coming from the network is handled by an Altera Max Ⅴ Complex Programmable Logic Device (CPLD). All those things are supervised by a R5F5631 Renesas Electronics 32-Bit microcontroller.

With reading the names of the electronics involved, can anyone says which audio digital input should be better on the ND8006?
 
I believe the PCM9211 carries forward the S/PDIF clock reconstructed using the phase-locked loop. So no buffering or re-clocking. Assuming I didn't miss anything in the datasheet. I do remember that when I selected the WM8804 for my DAC several years ago, Texas Instruments didn't seem to have a S/PDIF receiver with the same features.
 
skbe said:
With reading the names of the electronics involved, can anyone says which audio digital input should be better on the ND8006?
Click to expand...

Only you can decide, for your setup, your cables, and your preferences.
 
You must log in or register to reply here.
Back
Top