MQA

[gints]

Interesting. I came to US in 1977 as a 13 year old.

That year I went to school.
Thinking back about tapes, how freely we copied and transferred to other media, well, only other media was cassette. With LPCM we can do the same, we can reencode it to MP3 or DSD, if we wish. Some thinks that DSD is better for D/A process. With MQA we cannot do anything like that, it does not exist in digital format decoded, only analogue output is allowed. Same as with DSD some time ago. Isn't sort of DRM? SACD gave up, I understand, no problems get digital data from it, yes? Why MQA wants to step on the same raker? If they want a part of market, it would much easier to get it by allowing decoded digital data. Instead of new dac you can get add-on decoder box and everybody will be happy. If many people actually liked MQA sound as you say, why MQA do not want to distribute it as easy as possible.
I understand that in creating MQA file no analog processes involved, it is just repacking already existing digital data and all miracles happening in actuall D/A process. So, why do not drop this questionable folding, unfolding thing and present unaltered data to their own proprietary filter, like it was said before. They still can charge for it, people pay money for all sorts strange things to get good sound from their systems. Maybe it will be less money, but then it definitely will be about sound and not suspections about something else. And is it actually need to be hi-res data to make that specific MQA sound? 16/44 is not enough for streaming purposes? Maybe anything between it and 24/192 will do. For downloads should be fine any size what they want, I guess.

 

[gints]

But they just might slip you an unlabeled MQA CD...

Scary stuff.

Yes, actually it is, it is already happened with HDCD, I understand.

 

[DimitryZ]

That year I went to school.
Thinking back about tapes, how freely we copied and transferred to other media, well, only other media was cassette. With LPCM we can do the same, we can reencode it to MP3 or DSD, if we wish. Some thinks that DSD is better for D/A process. With MQA we cannot do anything like that, it does not exist in digital format decoded, only analogue output is allowed. Same as with DSD some time ago. Isn't sort of DRM? SACD gave up, I understand, no problems get digital data from it, yes? Why MQA wants to step on the same raker? If the want a part of market, it would much easier to get it by allowing decoded digital data. Instead of new dac you can get add-on decoder box and everybody will be happy. If many people actually liked MQA sound as you say, why MQA do not want to distribute it as easy as possible.
I understand that in creating MQA file no analog processes involved, it is just repacking already existing digital data and all miracles happening in actuall D/A process. So, why do not drop this questionable folding, unfolding thing and present unaltered data to their own proprietary filter, like it was said before. They still can charge for it, people pay money for much stranger things to get good sound from their systems. Maybe it will be less money, but then it definitley will be about sound and suspections about something else. And is it actually need to be hi-res data to make that specific MQA sound? 16/44 is not enough for streaming purposes? Maybe anything between it and 24/192 will do. For downloads should be fine any size what they want, I guess.

You can copy MQA physical files and you can decode the LPCM portion to other formats.

MQA is primarily a streaming format existing mostly inside Tidal's universe. Which means you can't capture anything from Tidal anyway.

You can now do DSP on MQA as well.

 

[DimitryZ]

Yes, actually it is, it is already happened with HDCD, I understand.

The horror of it! Can things get any worse?

 

[gints]

You can copy MQA physical files and you can decide the LPCM portion to other formats.

Wasn't it already showed earlier that it is not the same LPCM data what went into encoder?

 

[gints]

The horror of it! Can things get any worse?

So, you pay for milk and get a kefir. Not big problem for some, but try to add it to coffe.

 

[gints]

You can now do DSP on MQA as well.

Inside Tidal or independently?

 

[DimitryZ]

Wasn't it already showed earlier that it is not the same LPCM data what went into encoder?

It's the standard resolution portion with a bit of high frequency noise, where the MQA stuff is hidden.

If you have hearing normal for middle age, it's unlikely that you will be bothered by increase in the noise floor above 16KHz. I do hear it sometimes if I push an MQA stream into non-MQA DAC without letting Tidal do a soft unfold.

 

[DimitryZ]

Inside Tidal or independently?

Independently, in Roon, or other software that preserves MQA messanging channel and applies it to the stream after DSP. This tells the DAC which digital reconstruction filter to apply and how much to upsample.

 

[DimitryZ]

So, you pay for milk and get a kefir. Not big problem for some, but try to add it to coffe.

More like you pay for milk and get milk with a bit of honey.

 

[eevo1969]

The outcome of that survey was inconclusive. It proves nothing.

That reply is just a little misleading If it’s the survey in the link above, 3 out of the 4 tests favoured MQA in listening tests, nothing inconclusive about their preferences according to the graphs and the test subjects weren’t dragged in off the streets either.
For the sake of clarity wasn’t it music industry experts in said tests I.e. engineers/musicians and other “experienced” insiders?
I thought that you took issue with other aspects of MQA anyway, not the actual quality of sound that it was capable of delivering? What is your personal impression of the sound quality that it’s capable of delivering?

So MQA sounds good enough for some people. Fair enough.
So make MQA available as an optional filter choice that people can choose to select. Don't drench MQA onto all music, please.

A filter already exists, you.
Don’t buy a MQA DAC or use the services that provide it. Just walk away from it. I’ve had no trouble avoiding it over the last 5 years or so.
Personally I’d prefer a DAC with no option for MQA as I just won’t use that function as it stands. I have no source of MQA files and I don’t see that changing anytime soon. With Tidal considering exploring the option of lossless or hi-res as another tier MQA’s days seem a little more numbered.

 

[Jim Audiomisc]

That reply is just a little misleading If it’s the survey in the link above, 3 out of the 4 tests favoured MQA in listening tests, nothing inconclusive about their preferences according to the graphs and the test subjects weren’t dragged in off the streets either.
For the sake of clarity wasn’t it music industry experts in said tests I.e. engineers/musicians and other “experienced” insiders?

The above seems to me to raise a different issue. We've had decades of 'industry' people telling me that buyers *like* LOUDNESS and large amounts of sustained clipping on CDs. - even on 'highly regarded' classical CDs. I've also documented case after case of mess-ups like HDCDs with no flagging and poor conversions with needless error patterns that degrade the sound. So TBH I tend not to regard those who 'produce' CDs as 100% reliable. They have skills and experience, but it doesn't always mean they are correct. So their views are worth considering, but not automatically accepting. Unless you also think that distortion is a bonus which improves the sound.

 

[Jim Audiomisc]

It's the standard resolution portion with a bit of high frequency noise, where the MQA stuff is hidden.

If you have hearing normal for middle age, it's unlikely that you will be bothered by increase in the noise floor above 16KHz.

Incomplete description. Determinstic additions like anharmonic components are not 'noise' and thus may be more audible, particularly because of their coherent relationship with the original components that they alias. And we get back to the conundrum: If you can't hear this it becomes difficult to argue that you *can* hear the 'unfolded' stuff at *higher* frequencies which is the selling point in high rez terms.

Change can be of different types:

1) Noise

2) Linear 1:1 mapped distortion (nominally correctable)

3) Many:1 distortion (not always correctable

4) Memory affected distortion (can be a nightmare and impossible to undo)

The differences may matter.

 

[EPear]

In polls, about 25% of audiophiles hate MQA, 25% like it, and the rest either haven't decided or haven't heard of it.

I think it's quite early to declare that our community has come to a concensus on the matter.

Certainly, I would not attempt to speak for the community.

Can you expand on your actual experience with MQA?

Another week, another loop ... but to reply and address a couple of points here:

with regards to my actual experience, I guess suffice to say it's years long and long enough. And another thing - in my experience MQA does not sound better in a solid state system, absolutely not, but understand it may be your preference on yours. Qobuz hires sounds clearly better to me. I cannot comment on tube based systems.

Otherwise, while I do not attempt to speak for the community or think it's a consensus (clearly, we still have you!), reading on this topic recently only seems to show that there "seems to be" a community view here and elsewhere - in fact looking at the number of folks leaving Tidal, it looks more like a landslide! But don't take my word for it .... why don't you do a poll here and see what everyone thinks? How about that? Just one or two simple questions:

- Do you prefer Tidal MQA or Qobuz hires?
- Would you like an MQA technology (filter) hard engineered into your DAC as an integral part of the signal processing?

 

[Jim Audiomisc]

To give people a brief update relevant to #1853:

I'm still focussing on analysing various files. This has shown up some interesting aspects that I'm exploring. But it means I have to keep modifying or writing new analysis progs to get answers. So remains slow. Once I'm ready I plan to present the details. However I'll make one comment that I used to call "Jim's Law" back in the days when I used to give undergrad lectures on Information and Measurement.(1)

"Data only becomes Information when you know exactly how it was produced."

This has implications which people often don't notice until pointed out. That'll make more sense when I get to finish and present what I've found.

(1) if curious, the old lecture notes are here

https://www.st-andrews.ac.uk/~www_pa/Scots_Guide/iandm/intro.html

 

[Yomanze]

After reading through these docs it’s hard to argue against the fact that it is a lossy format, with distortion and DRM issues, an archaic solution to a non-issue.

It is easy to do some great MQA masters, compare against a poorer redbook master, and say it sounds better. On an even playing field it appears that WAV / FLAC is the technically superior format, and trivial to stream given today’s broadband speeds.

 

[Jim Audiomisc]

BTW I'm still happily buying and playing CDs. In some ways this is a golden age for them. Loads of 2nd hand ones which - unlike old LPs - are often in mint condition. e.g. set of 10 Ellington's "Private recordings" CDs that I got for a fiver, each. Lots of new ones at good prices like the Barbirolli set remastered last year. The sad thing is the lack of local music shops. So I have to buy from people I know over the phone, email, etc.

 

[Wolfmancatsup]

I thought this was interesting. A review of MQA-CD by the not-very-techy Techmoan. Interesting that in Japan some MQA-CDs were released with regular CDs in the box as well, to give the listener a way of comparing the two. Also worth reading Techmoan’s later notes beneath the post, to put his comments in the video into perspective.
Please note - I have no axe to grind in either direction. I’ve never listened to MQA-anything, and so long as it doesn’t take over and remove access to non-MQA stuff I have no real interest; I just thought the video was interesting from a techy point of view:-

Mick

 

[EPear]

More like you pay for milk and get milk with a bit of honey.

Doesn't sound like a good idea to me, but let's say it's honey - you just forgot to mention it's some sort of synthetic honey

Now - where it gets really exciting is that you go to your local supermarket and all the milk, all of it, is honey flavoured, no other milk on the shelves! Even better, isn't it?

And the manufacturers say that this is how even the cows and milkmen (or are they the milking men?) like their milk and this is how it should be at source, from the Master! And, of course, you will never ever again be able to take the honey out of the milk. And you will not know what milk without honey is, but hey by then milk without this synthetic honey would make no sense anyway, it's how it should be, right?

And, finally to make this even more real, here is something from one of the most real movies you will ever watch (it's effectively a documentary):

"Joe: For the last time, I'm pretty sure what's killing the crops is this Brawndo stuff.
Secretary of State: But Brawndo's got what plants crave. It's got electrolytes.
Attorney General: So wait a minute. What you're saying is that you want us to put water on the crops.
Joe: Yes.
Attorney General: Water. Like out the toilet?
Joe: Well, I mean, it doesn't have to be out of the toilet, but, yeah, that's the idea.
Secretary of State: But Brawndo's got what plants crave.
Attorney General: It's got electrolytes."

 

[mansr]

That reply is just a little misleading If it’s the survey in the link above, 3 out of the 4 tests favoured MQA in listening tests, nothing inconclusive about their preferences according to the graphs and the test subjects weren’t dragged in off the streets either.
For the sake of clarity wasn’t it music industry experts in said tests I.e. engineers/musicians and other “experienced” insiders?

With the small sample size, poorly controlled conditions (including self-selected participants), unknown confounding variables, etc, the reported preference would have to be much stronger in order to mean anything one way or the other.

I thought that you took issue with other aspects of MQA anyway, not the actual quality of sound that it was capable of delivering? What is your personal impression of the sound quality that it’s capable of delivering?

The MQA core process is probably audibly transparent for most inputs. However, since true lossless encoding is guaranteed transparent for all inputs, MQA serves no useful purpose. When you buy MQA, you are, at best, giving Bob Stuart money for nothing. If you're feeling generous, give to a local charity. Bob doesn't need a handout.