Cayin N6ii, Unlimited Possibilities: a fully modularized smart DAP
Jan 25, 2021 at 5:08 AM Post #5,463 of 8,191
Jan 28, 2021 at 9:53 AM Post #5,464 of 8,191
Is anyone running TIDAL with their N6ii for MQA? I've been trying to help someone else out on a DAC issue and I used my N6ii as an MQA reference. When playing some some TIDAL Masters songs, my N6ii shows sample rates like this:

1B0utB0.jpg


So, in that example it is showing 384kHz as the sample rate (top of screen).

It emerges that this might not be correct. The suggestion is that the N6ii is oversampling the correct MQA unfolded rate (which should be 192kHz for the song shown on the screen in my picture).

Is anyone able to verify what you are seeing when you play a Masters track. I only ever see 352.8kHz for a TIDAL Masters 44.1kHz song, and 384kHz for a TIDAL Masters 48kHz song.

Apparently, some TIDAL Masters are 192kHz when unfolded.

So, the question is this: is the N6ii oversampling TIDAL Masters songs? I tried playing Masters songs in UAPP too and I see the same rates at the top of the screen. If it is oversampling, then is there a way to deliver the true MQA sample rate?
 
Jan 28, 2021 at 2:03 PM Post #5,465 of 8,191
Is anyone running TIDAL with their N6ii for MQA? I've been trying to help someone else out on a DAC issue and I used my N6ii as an MQA reference. When playing some some TIDAL Masters songs, my N6ii shows sample rates like this:

1B0utB0.jpg


So, in that example it is showing 384kHz as the sample rate (top of screen).

It emerges that this might not be correct. The suggestion is that the N6ii is oversampling the correct MQA unfolded rate (which should be 192kHz for the song shown on the screen in my picture).

Is anyone able to verify what you are seeing when you play a Masters track. I only ever see 352.8kHz for a TIDAL Masters 44.1kHz song, and 384kHz for a TIDAL Masters 48kHz song.

Apparently, some TIDAL Masters are 192kHz when unfolded.

So, the question is this: is the N6ii oversampling TIDAL Masters songs? I tried playing Masters songs in UAPP too and I see the same rates at the top of the screen. If it is oversampling, then is there a way to deliver the true MQA sample rate?

A quote from another Internet forum.
"Tidal confirmed that the Windows Tidal software will do 1 x unfolding and offer up to 96kHz to your computer's internal DAC or to external non-MQA DACs. External MQA DACs can do 4 x unfolding for the maximum sampling rate (384kHz)".

The N6ii would appear to be doing 4x decoding/unfolding which is very good.
 
Last edited:
Jan 28, 2021 at 2:34 PM Post #5,466 of 8,191
A quote from another Internet forum.
"Tidal confirmed that the Windows Tidal software will do 1 x unfolding and offer up to 96kHz to your computer's internal DAC or to external non-MQA DACs. External MQA DACs can do 4 x unfolding for the maximum sampling rate (384kHz)".

The N6ii would appear to be doing 4x decoding/unfolding which is very good.

But is it doing proper MQA decoding?

Here's my LG V30+ phone with native MQA decoding:

s0UO1Kl.jpg


As I understand things, not all Masters tracks have four unfolds available. Take a look at the conversation here where I used my N6ii as a reference but I think things are not right:

https://www.head-fi.org/threads/topping-d90.926531/post-16133178

If you follow from there, it looks like things with the N6ii are not quite right. The person I was trying to help has an MQA D90 DAC and it's showing 192kHz for that song. My LG V30+ phone is showing 192kHz. And someone on that thread confirmed that Coldplay Masters songs unfold to 192kHz only.

Hence, I want to check on here if the N6ii is correctly unfolding Masters songs. Here's my N6ii running UAPP, which is showing the correct unfolded sample rate of 88.2kHz, whilst the sample rate shown in the N6ii's info bar at the top is 352.8kHz. This isn't right as this is a very low quality Masters track and doesn't have a 4x unfold. It looks like the N6ii is upsampling decoded tracks. I'm hoping there's an explanation for this! I'd prefer just to hear the correctly unfolded Masters song.
 
Jan 28, 2021 at 2:44 PM Post #5,467 of 8,191
If you follow from there, it looks like things with the N6ii are not quite right. The person I was trying to help has an MQA D90 DAC and it's showing 192kHz for that song. My LG V30+ phone is showing 192kHz. And someone on that thread confirmed that Coldplay Masters songs unfold to 192kHz only.
UAPP can only display a guess when its playing MQA as there is no way for the app to get the true unfolded sample rate from the hardware.

I'm pretty sure the N6ii is unfolding correctly and displaying the correct value on the status bar.

Here is an example of another DAC doing it right (The Brooklyn DAC from Mytek):

mytekbrooklyn_front_black_l.jpg
 
Jan 28, 2021 at 3:33 PM Post #5,469 of 8,191
@MarkParity Okay, so that rules out UAPP as a reliable source of the sample rate. Thanks for that info. Then I suppose there's the MQA D90 DAC that I posted the discussion link. That's showing 192kHz on its display. One of these must be wrong! Either the MQA D90 or the N6ii.
I think the difference comes from the fact that the Topping D90 is a full decoder, so better? Maybe? and the N6ii is a renderer.


Q4. What are advantages of an MQA Renderer?

  • An MQA Renderer can provide a low-cost or lightweight upgrade for playback of music encoded in MQA.
  • Some Class I USB devices can only receive 96 kHz audio. In these cases, a Renderer is able to take in the MQA Core and complete the unfolding up to 384 or even 768 kHz internally.


Its all here, but not necessary self explanatory.

http://bobtalks.co.uk/blog/science-mqa/mqa-playback/
 
Feb 1, 2021 at 4:29 AM Post #5,470 of 8,191
Well, my 2 cents on MQA, FWIW.

First, the observation on my N6ii with the original Tidal app is the same, every MQA track ends up unfolded to 352.8 or 384 kHz, depending on whether it is a folded 44.1 or 48 kHz.

Second, I did a bit of playing around on Windows with Tidal via their own app and integrated within Roon, using the N6ii as an MQA capable USB-DAC and my Chord Hugo 2 as an MQA-incapable (should I better say averse?) USB-DAC. Unfortunately, I couldn‘t get to conclusive results yet regarding how to properly do the settings for the DACs, either in Tidal or in Roon.

Third, my thoughts on MQA are more on the reluctant side than on the enthusiastic.

For example, look at ECM Records, a fully respectable label (although having joined Universal for distribution) who decided to jump the hi-res bandwagon after quite some waiting, and went into releasing MQA versions even later. I was just wondering why on earth their MQAs do all unfold to 384 kHz, independent of whether the hi-res PCM releases are 96 or 192 kHz. So if there were some "true“ audio contents up to 384/2 kHz, that would mean that they must necessarily have done a new remastering from the original analog tapes. The easier and more probable way to produce an MQA file would be to take the pre-existing hi-res PCM and subject it to the - lossy! - MQA encoding.

From that observation - as you said before - it might be possible that the N6ii would upsample after unfolding the MQA. On the other hand, what does "MQA renderer" mean?

So the underlying question is: What is the "true" original sample rate of an MQA track??? I.e., up to what frequency does its real audio content extend??? What one would like to do is: We - at least some of us users and some hi-res download stores - have been analyzing the frequency distributions of PCM tracks using software like Spek or Music Scope. This does quickly distinguish true hi-res PCM masters from upsampled fakes being sold. So one would really like to look at the frequency spectrum of unfolded MQAs.

Another example: During my - admittedly still short - journey with MQA, I came across quite some MQA releases on Tidal that unfold to 300+ kHz on the N6ii; however, no hi-res release at all does exist anywhere, just CD quality on Qobuz / Tidal / download stores. I think that‘s a bit astonishing, but not impossible. I found quite a number of such releases on Rhino / Atlantic.

Finally, I agree with
Its all here, but not necessary self explanatory.
http://bobtalks.co.uk/blog/science-mqa/mqa-playback/
The way MQA is advertised and explained is not even close to what I would expect to be transparent and understandable stuff. Then yes, they need to keep it proprietary. It reminds me a lot of the historic HDCD stuff who wanted to conquer a portion of the Redbook CD financial cake by adding proprietary hardware and licensing. (Anyway, at least they [Pacific Microsonics] gave the audio world the best sounding analog-to-digital converter which stood the test of time long after they folded.)

Notwithstanding the last paragraph, I do enjoy to have so many choices nowadays, and of course to have the N6ii. Just got my E02 module, unfortunately still fighting to get a balanced IEM cable ...
 
Last edited:
Feb 1, 2021 at 4:51 AM Post #5,471 of 8,191
@111MilesToGo Thank you for your post. So there is a possibility of an issue with 'pure' MQA decoding on the N6ii. I'm a 'bit-perfect' kind of person and don't really want any upsampling of any kind. I don't ever use EQ either. I just want to hear the original unfolded Masters track. I guess some official clarification on this point should be made. Does the N6ii upsample all Masters tracks? Nevertheless, I'm still loving the N6ii. It's just a concern in the back of my mind.
 
Feb 1, 2021 at 5:46 AM Post #5,472 of 8,191
@111MilesToGo Thank you for your post. So there is a possibility of an issue with 'pure' MQA decoding on the N6ii. I'm a 'bit-perfect' kind of person and don't really want any upsampling of any kind. I don't ever use EQ either. I just want to hear the original unfolded Masters track. I guess some official clarification on this point should be made. Does the N6ii upsample all Masters tracks? Nevertheless, I'm still loving the N6ii. It's just a concern in the back of my mind.
Thanks, @Taz777. I‘m absolutely the same type of person. And I would hope for an official clarification, too. (Guess that won‘t com from the MQA inventor / owners...) The questions have been asked - upsample after MQA unfolding, what is MQA rendering, and what‘s the ”true“ frequency contents of an audio file.

In general, I would not make the objection to upsampling a strict and harsh point of view. But it needs to be executed very well. Example: The Chord M-Scaler is reported to be overwhelming in its upsampling to e.g. 768 kHz. Never heard it though, and never will be able to own one plus its appropriate DAC...

Just for expanding on such topics, without any watering down the request for MQA explanation: Why are we talking about such high frequencies / sample rates in digital audio? My concept is to think of the human ears and brains as two ”devices“, a frequency analyzer and a time analyzer. The frequency analyzer has that standard <20-20,000 Hz response or so, which degrades with age. The time analyzer - as far as I read - has the ability to differentiate signal arrival times at the two ears down to 2 to 8 microseconds - this is what calls for sampling rates up to several hundred kHz or even a MHz. You know, in those ancient ages one needed to know it was a lion that one heard (frequency analyzer), but one desperately needed to know where it was and how far away (time analyzer).

So hope for understanding MQA better!
 
Feb 1, 2021 at 11:06 PM Post #5,473 of 8,191
Is anyone running TIDAL with their N6ii for MQA? I've been trying to help someone else out on a DAC issue and I used my N6ii as an MQA reference. When playing some some TIDAL Masters songs, my N6ii shows sample rates like this:

1B0utB0.jpg


So, in that example it is showing 384kHz as the sample rate (top of screen).

It emerges that this might not be correct. The suggestion is that the N6ii is oversampling the correct MQA unfolded rate (which should be 192kHz for the song shown on the screen in my picture).

Is anyone able to verify what you are seeing when you play a Masters track. I only ever see 352.8kHz for a TIDAL Masters 44.1kHz song, and 384kHz for a TIDAL Masters 48kHz song.

Apparently, some TIDAL Masters are 192kHz when unfolded.

So, the question is this: is the N6ii oversampling TIDAL Masters songs? I tried playing Masters songs in UAPP too and I see the same rates at the top of the screen. If it is oversampling, then is there a way to deliver the true MQA sample rate?

Well, my 2 cents on MQA, FWIW.

First, the observation on my N6ii with the original Tidal app is the same, every MQA track ends up unfolded to 352.8 or 384 kHz, depending on whether it is a folded 44.1 or 48 kHz.

Second, I did a bit of playing around on Windows with Tidal via their own app and integrated within Roon, using the N6ii as an MQA capable USB-DAC and my Chord Hugo 2 as an MQA-incapable (should I better say averse?) USB-DAC. Unfortunately, I couldn‘t get to conclusive results yet regarding how to properly do the settings for the DACs, either in Tidal or in Roon.

Third, my thoughts on MQA are more on the reluctant side than on the enthusiastic.

For example, look at ECM Records, a fully respectable label (although having joined Universal for distribution) who decided to jump the hi-res bandwagon after quite some waiting, and went into releasing MQA versions even later. I was just wondering why on earth their MQAs do all unfold to 384 kHz, independent of whether the hi-res PCM releases are 96 or 192 kHz. So if there were some "true“ audio contents up to 384/2 kHz, that would mean that they must necessarily have done a new remastering from the original analog tapes. The easier and more probable way to produce an MQA file would be to take the pre-existing hi-res PCM and subject it to the - lossy! - MQA encoding.

From that observation - as you said before - it might be possible that the N6ii would upsample after unfolding the MQA. On the other hand, what does "MQA renderer" mean?

So the underlying question is: What is the "true" original sample rate of an MQA track??? I.e., up to what frequency does its real audio content extend??? What one would like to do is: We - at least some of us users and some hi-res download stores - have been analyzing the frequency distributions of PCM tracks using software like Spek or Music Scope. This does quickly distinguish true hi-res PCM masters from upsampled fakes being sold. So one would really like to look at the frequency spectrum of unfolded MQAs.

Another example: During my - admittedly still short - journey with MQA, I came across quite some MQA releases on Tidal that unfold to 300+ kHz on the N6ii; however, no hi-res release at all does exist anywhere, just CD quality on Qobuz / Tidal / download stores. I think that‘s a bit astonishing, but not impossible. I found quite a number of such releases on Rhino / Atlantic.

Finally, I agree with

The way MQA is advertised and explained is not even close to what I would expect to be transparent and understandable stuff. Then yes, they need to keep it proprietary. It reminds me a lot of the historic HDCD stuff who wanted to conquer a portion of the Redbook CD financial cake by adding proprietary hardware and licensing. (Anyway, at least they [Pacific Microsonics] gave the audio world the best sounding analog-to-digital converter which stood the test of time long after they folded.)

Notwithstanding the last paragraph, I do enjoy to have so many choices nowadays, and of course to have the N6ii. Just got my E02 module, unfortunately still fighting to get a balanced IEM cable ...

I just catch up with the MQA discussion here. I don't recall we have added any upsampling feature after MQA unfold in our MQA implementation, but I'll consult our engineer just to be sure. In fact, Cayin has never implement any upsampling algorithm to N6ii Audio Motherboards in our in-house digital audio processing, some DAC chipset might have their own internal upsampling process (e.g, ASRC in ESS chipset), so we won't described our Audio Motherboard as NOS either.

Just one quick reminder, when you test MQA decoding, you MUST connect N6ii to a headphone. When N6ii is not connected to any headphone, it will fall back to digital output status (to extend battery duration), and MQA is not exactly friendly to digital output.
 
Cayin Stay updated on Cayin at their sponsor profile on Head-Fi.
 
http://en.cayin.cn/
Feb 2, 2021 at 1:30 AM Post #5,474 of 8,191
Thank you for checking, @Andykong. I will also retest with headphones connected. This enquiry only came about when someone with a Topping D90 MQA DAC in another thread was seeing a 192kHz unfolded rate for Coldplay Masters albums in TiDAL and the N6ii was showing 384kHz. I will also check against some of poor quality 16-bit MQA songs that TIDAL recently added to its catalogue. Those are definitely not 384kHz unfolded.
 
Feb 2, 2021 at 2:48 AM Post #5,475 of 8,191
Thanks for your answer, @Andykong.

Yes, I had observed the requirement to have a headphone connected. Without one, the N6ii status bar would display a rate one big step lower. Saw this for both the A01 and T01.

Regarding MQA, it seems I still need to understand much more than I do now - IF MQA lends itself to being understood by the general public at all. Maybe the point is in ”full decoder“ versus ”renderer“.

Now, on a related note, I am left with these questions, dear @Andykong: What Settings are the correct ones for using the N6ii as a USB-DAC with Roon (on Windows in my case)? In particular, what Settings regarding the MQA capabilities? And what am I supposed to see when looking at the Signal Path during MQA playback from Tidal integrated within Roon? And even though I didn‘t use it yet, I would also like to know what Settings need to be dialed in in the original Tidal Windows app? When running the N6ii as a USB-DAC on Windows, it might also be interesting to look at the status display provided by the Cayin (Thesycon) driver; off-hand I‘d guess it would show 88.2 or 96 kHz when Roon runs as a first-unfold MQA Core.
 

Users who are viewing this thread

Back
Top