• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Tidal, MQA, D90 and sampling rates

snakearmpit

Member
Joined
Mar 9, 2021
Messages
11
Likes
2
Location
Ireland
(noob here, be kind)
Hi, can someone please explain or point me to an article that explains why different mqa tracks in tidal result in different sampling rates being reports by the dac? I'm running tidal pc->usb->topping d90. I have exclusive mode, force volume and mqa passthrough all enabled and topping drivers installed. Some tracks show up as "MQA<dot> 192KHz" which is what I'd expect to see most of the time, but many show up as "MQA <dot> 48KHz" or "MQA <nodot> 44.1KHz". Are not all mqa tracks created equal?
Bonus points if you can tell me how to make the D90 report current bit depth of whatever input is selected.
cheers,
 
Last edited:

Jimbob54

Grand Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
11,096
Likes
14,753
(noob here, be kind)
Hi, can someone please explain or point me to an article that explains why different mqa tracks in tidal result in different sampling rates being by the dac? I'm running tidal pc->usb->topping d90. I have exclusive mode, force volume and mqa passthrough all enabled and topping drivers installed. Some tracks show up as "MQA<dot> 192KHz" which is what I'd expect to see most of the time, but many show up as "MQA <dot> 48KHz" or "MQA <nodot> 44.1KHz". Are not all mqa tracks created equal?
Bonus points if you can tell me how to make the D90 report current bit depth of whatever input is selected.
cheers,
Everything working as it should. Those are indeed the "unfolded" sample rates of some tidal mqa music.
 
OP
S

snakearmpit

Member
Joined
Mar 9, 2021
Messages
11
Likes
2
Location
Ireland
thank you, great to know the setup is fine.
I'm still puzzled by why that is though. What's the alleged benefit of mqa at 44.1KHz in this case? Are there good resources out there on what different unfolding steps actually do and why some tracks have different unfolding levels?
(and before someone eventually screams snake oil and all, I'm trying to learn to make up my own mind, thanks)
 

Jimbob54

Grand Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
11,096
Likes
14,753
thank you, great to know the setup is fine.
I'm still puzzled by why that is though. What's the alleged benefit of mqa at 44.1KHz in this case? Are there good resources out there on what different unfolding steps actually do and why some tracks have different unfolding levels?
(and before someone eventually screams snake oil and all, I'm trying to learn to make up my own mind, thanks)
You are going to hit the wall of mqa scepticism on this site.

Have a look at the mqa site for their version of the background and the process.

Why are they different rates? Because that's what the studio sent them out as very simply (or what it was encoded at possibly). What is the point of an mqa file that fully unfolds to CD sample rate? No idea!
 

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,371
Likes
18,280
Location
Netherlands
It’s actually in the D90 manual:
321C5A76-F026-4FA9-825B-68DFE72D73DD.jpeg


See https://bobtalks.co.uk/blog/mqa-philosophy/mqa-authentication-and-quality/ for some background.
 
OP
S

snakearmpit

Member
Joined
Mar 9, 2021
Messages
11
Likes
2
Location
Ireland
thanks, yeah I'm familiar with the manual and the mqa modes. My question was more around the sampling rate and my [incorrect] expectation that every mqa file would be 192KHz or higher.
cheers,
 

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,371
Likes
18,280
Location
Netherlands
I’m guessing here, but I tend to be quite good at that: I think the 44.1 and 48 KHz versions are probably just MQA authenticated. If the original master was not available as a higher sample late version, this is the best you can get. There will be no unfolding in that case, just authentication. Basically meaning next to nothing..

By any chance are these low sample rate versions older recordings?
 

Jimbob54

Grand Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
11,096
Likes
14,753
I’m guessing here, but I tend to be quite good at that: I think the 44.1 and 48 KHz versions are probably just MQA authenticated. If the original master was not available as a higher sample late version, this is the best you can get. There will be no unfolding in that case, just authentication. Basically meaning next to nothing..

By any chance are these low sample rate versions older recordings?

A lot of the new releases top out at 44 or 48 oddly. Which, like you say, makes it all about the coloured light more than any (perceived) benefit of higher res.
 

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,371
Likes
18,280
Location
Netherlands
Should be easy to compare to Qobuz’s versions sample rates of the same songs/albums.
 
OP
S

snakearmpit

Member
Joined
Mar 9, 2021
Messages
11
Likes
2
Location
Ireland
'old' is relative :) one example is Natalie Merchant, Tigerlily from 95 is MQA. 96KHz, Ophelia from 98 is MQA 44.1.
 

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,371
Likes
18,280
Location
Netherlands
Ophelia is also 44.1 kHz on Qobuz, Tigerlily is 96 kHz.. so as @Jimbob54 said: It’s just to make the fancy MQA light go on :facepalm:
 

nimar

Active Member
Forum Donor
Joined
Jan 25, 2021
Messages
213
Likes
216
Location
Ontario, Canada
I suspect its more likely due to the fact that a record company / streaming service agreed to reprocess everything in MQA and the only thing they have is 16/44.1. From a technical perspective, as to the USP of MQA, it does seem rather pointless, I suppose they would argue at the very least they've chosen the "best" filter for that album based on their analysis of the recording.
 

sal

Active Member
Forum Donor
Joined
Mar 23, 2021
Messages
126
Likes
169
My DAC also shows the bit rate. All MQA files I have played off of Tidal unfold to 24 bit. So you’re getting 24/44.1khz worst case as opposed to 16/44.1. Is there an audible difference? Sometimes. YMMV
 

Jim Matthews

Major Contributor
Forum Donor
Joined
Mar 25, 2021
Messages
1,051
Likes
1,286
Location
Taxachusetts
An important consideration with Tidal is the cost for access to MQA is $19.99 per month.
 

bhobba

Member
Joined
Oct 29, 2021
Messages
20
Likes
6
My question was more around the sampling rate and my [incorrect] expectation that every mqa file would be 192KHz or higher.
cheers,

MQA claims they give you what the recording engineer intended. If it was 96k, then that is what you get. If it was 44.1k, then that is what you get. If it was 768k, that is what you get. The blue light is meant to indicate what you are getting what was intended. Now, does it sound better, worse, or the same as the original? Only you can decide that in comparison. Although MQA claims it is lossless, it is as lossy as hell. Have a look a figure 6 in the following:

MQA believes all relevant audio information lies in that triangle and has designed a system to capture it. It doesn't really - there is a slow roll-off above 20khz, which they at least think is inaudible. But basically, that's is what MQA is designed to do - capture what is in the triangle. It is easy to break MQA by creating a file with information outside the triangle, but the claim is they are artificial and do not occur in practice.

The 0-20kz part of what you are getting is lossless except for dithering. But dithering has been used for years, and most would consider it good engineering practice, even though it is strictly speaking lossy. That's the confounding thing about MQA - is it just good engineering practice, or is it a lossy load of rubbish? I can't answer that except to say I like it. The marketing hype promulgated about it is, how to put it, 'nauseating'.

IMHO they should forget this - exactly as the recording engineer intended rubbish - and concentrate what MQA does best - capturing that triangle. Let the decoder decide on what the DAC sounds best with - 48k, 96k, 384k, or even a whopping 1536k. Just my view.

Thanks
Bill
 

Jimbob54

Grand Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
11,096
Likes
14,753
MQA claims they give you what the recording engineer intended. If it was 96k, then that is what you get. If it was 44.1k, then that is what you get. If it was 768k, that is what you get. The blue light is meant to indicate what you are getting what was intended. Now, does it sound better, worse, or the same as the original? Only you can decide that in comparison. Although MQA claims it is lossless, it is as lossy as hell. Have a look a figure 6 in the following:

MQA believes all relevant audio information lies in that triangle and has designed a system to capture it. It doesn't really - there is a slow roll-off above 20khz, which they at least think is inaudible. But basically, that's is what MQA is designed to do - capture what is in the triangle. It is easy to break MQA by creating a file with information outside the triangle, but the claim is they are artificial and do not occur in practice.

The 0-20kz part of what you are getting is lossless except for dithering. But dithering has been used for years, and most would consider it good engineering practice, even though it is strictly speaking lossy. That's the confounding thing about MQA - is it just good engineering practice, or is it a lossy load of rubbish? I can't answer that except to say I like it. The marketing hype promulgated about it is, how to put it, 'nauseating'.

IMHO they should forget this - exactly as the recording engineer intended rubbish - and concentrate what MQA does best - capturing that triangle. Let the decoder decide on what the DAC sounds best with - 48k, 96k, 384k, or even a whopping 1536k. Just my view.

Thanks
Bill
I'm not sure the blue light ties so the sample rate output.

Blue light is an indicator (allegedly) of provenance. I'm not sure the creator /rights owner of something recorded to tape in the 70s can say something should play back at 44.1 /48/96/192 etc etc

 

nickt1

Member
Joined
Oct 12, 2021
Messages
12
Likes
6
(noob here, be kind)
Hi, can someone please explain or point me to an article that explains why different mqa tracks in tidal result in different sampling rates being reports by the dac? I'm running tidal pc->usb->topping d90. I have exclusive mode, force volume and mqa passthrough all enabled and topping drivers installed. Some tracks show up as "MQA<dot> 192KHz" which is what I'd expect to see most of the time, but many show up as "MQA <dot> 48KHz" or "MQA <nodot> 44.1KHz". Are not all mqa tracks created equal?
Bonus points if you can tell me how to make the D90 report current bit depth of whatever input is selected.
cheers,
I also have a D90 and use Tidal and think it sounds better than qobuz to my ears in back to back listening with great resolution and to me appears to have some noise canceling abilities. Is it a cash grab-yes. Is it better than lossless hires probably not.It does however offer the ability to stream hi res. files over a wide range of devices using much less bandwidth.
It is a new format and most new formats are proprietry and are comercialised (eg.CD,LP etc).
Will there be better formats available in the future?probably but for now it is the best solution I can see.
 
Top Bottom