MQA is Vaporware

[Thuaveta]

39 minutes ago, Lee Scoggins said:

 

One thing you clearly are is a useful idiot.  You are being manipulated by bad mastering engineers like Brian Lucey who stand to lose money from the MQA approach.  By attacking honest journalists, you are creating large amounts of bad information that obscure the good news of more hirez music.  As a result many readers of the CA community will be uninformed about MQA and may miss the advantage of large catalogs of music.

 

Perhaps even worse, your constant attacks on those with a different opinion is scaring away good people from even wanting to participate on the board thereby lowering the quality of good discussion and exploration.

 

When a community has a widespread reputation for having biased discussion and personal attacks, very few people with experience want to participate. And that is what has happened here.

 

Quote

You have 5 posts.  Have you even followed the discussion and understand the basics of the topic?

 

Please don't appeal to authority, it's below both of us. And please don't call what you're doing journalism, it's an insult to the profession.

 

As for being a useful idiot, and if you want to insist on appeal to authority, I'll proudly be Bruno Putzeys and John Siau's useful idiot. I also tend to think that there's more competence, integrity, and knowledge of the matter at hand in either Anton Schlesinger's or Christoph Engemann's pinky nails than in your entire body, but that's just an opinion based on my reading of your many posts.

 

I think your problem isn't that the CA community is uninformed about MQA. It's that the CA community is very well informed about MQA, in a way that you happen to disagree with. Since you fancy yourself a journalist now, I'd wager the word you were looking for was "misinformed", something I'd therefore be led to believe you're trying to "correct". Now I can only wonder why.

 

[firedog]

9 hours ago, Lee Scoggins said:

And speaking of hardware support, Astell & Kern has jumped in.

 

IRIVER To Add MQA Support For Line Of Audio

Streaming Devices

MQA Support Will Be Added To Streaming Devices Starting With ACTIVO, Then Added To Other Streaming And Audio Devices 

 

Las Vegas, NV – IRIVER, the parent company of Astell&Kern, the global leader in high-resolution portable audio players, announces support for MQA decoding and playback will be added to the ACTIVO portable high-resolution audio player in January 2018.

 

MQA’s award-winning technology captures and reproduces the sound of the original studio master in a file that’s small enough to stream and download easily.  With the TIDAL Masters integration, users can instantly stream thousands of MQA tracks on their players.

 

James Lee, CEO of IRIVER said, “IRIVER is committed to bringing the latest innovation and technology to consumers.  We are happy to add MQA support to the new ACTIVO audio player, powered by Astell&Kern’s new TERATON sound solution.  We will continue to add MQA support to other streaming and audio devices in the future.”

 

Mike Jbara, CEO of MQA, commented, “It’s exciting that MQA’s technology will be integrated into IRIVER devices.  IRIVER’s addition to the list of MQA partners is a milestone in the creation of an environment where all consumers can easily enjoy high quality audio.”

 

The ACTIVO CT10 is the first high-resolution audio player from groovers Japan and features the new TERATON sound module by Astell&Kern, allowing high resolution audio playback up to 24bit, 192 kHz high resolution PCM audio and up to double-rate DSD (converted to PCM).  The ACTIVO CT10 also supports music streaming services including TIDAL and

groovers Japan.

 

The ACTIVO CT10 will be displayed in the Astell&ASPR booth during CES2018, Central Hall, booth # 18218.

 Reposting verbatim a  pro MQA press release doesn't help convince your critics that you are objective or critical.....

[firedog]

9 hours ago, Lee Scoggins said:

 

Keep in mind this is a series of articles to explore the format.  I have not commented on the technical elements yet.

 

The elegance of the MQA business model is that it does not look at hirez availability from the audiophile perspective but instead looks at what the record labels recognize is important and that is streaming.  We may disagree but this idea around enhancing existing files for better quality in streaming is sort of, in my view, a way to piggyback hirez quality onto a way into non-audiophile's homes.  Best of all, it seems to have worked as the three big labels and the independent community have signed up for the whole catalog.  

 

Also, there are legitimate bandwidth concerns at scale so I would differ with you there.  

 

And keep in mind that many mastering engineers are unhappy as the MQA approach limits the amount of times they can get paid to release in different formats.

 

This first article was positive because I am genuinely excited about getting millions more tracks in hirez.

 

As more music becomes available, we will see more hardware providers jump in.  As more hardware can do MQA, we will see more consumer interest and/or exploration.

 

With all of the major labels on board, it seems likely that MQA is becoming a standard whether we have different technical opinions or not.   In my view, getting MQA done is a simpler path to getting hirez released than the audiophile-centric way of limited releases on download sites and formats like SACD/DVD-Audio that were great but lacked traction and title availability.

And you still haven't given any kind of substantial answer to your claim that MQA is needed to get more hi-res released to streaming. 

MQA isn't necessary for streaming hi-res. All those files being released for streaming as MQA could be released not in MQA. The only reason it is happening this way is that the labels are using  MQA as  the  excuse for the releases. 

Their real goal is to keep us from ever having access to the actual master files, and to be able to control our use of the files through the DRM built into MQA. Just wait till MQA becomes ubiquitous. Other hi-res formats will disappear, and suddenly prices for "hi-res" streaming will go up. 

[firedog]

5 hours ago, Don Hills said:

 

Is your view the same as the labels' view? Think very carefully about what value proposition MQA offered the labels. Hint: It's not to bring higher quality to non-audiophile consumers. The consumers don't care and won't pay for it.

 

Edited to add:

Look carefully at how the labels expect to recoup their investment. They've paid the license fees and paid for the work of encoding their back catalogues. They wouldn't do that if  they didn't expect to profit in some way.

 

Yes, he seems incapable of thinking about these aspects and has ignored all attempts to get him to engage on this level. 

As always, "follow the money". When you do it is obvious that his rose colored view can't be the correct one. 

 

The idea that the availabilty of all  those "hi-res" MQA downloads will get the masses to pay premium prices for streaming them is ridiculous. If that were true, Tidal would have seen millions of users migrate to it's "hi-fi" tier from other services over the past couple of years, and especially since they began streaming MQA.

It hasn't happened, Lee. Sort of punches a hole in you whole theory about what MQA is and why the labels are promoting it.

[mcgillroy]

CA Forum MQA-threads: where audio “journalists” come to die.

 

They play chess, we play Go. This is going to be fun.

[Bob Stern]

1 hour ago, Lee Scoggins said:

From a business or technical standpoint, DRM doesn't make any sense.

 

Then why did Bob Stuart and Peter Craven file multiple patent applications on DRM for streaming audio?

 

https://www.google.com/patents/WO2014125285A1

[Thuaveta]

51 minutes ago, Lee Scoggins said:

 

You have 5 posts and joined just a couple of weeks ago.  Have you even followed the whole discussion and understand the basics of the topic?

 

You need to put a good faith effort into asking yourself why you don't have more experience and why others have a different opinion.

 

Oh, and since you believe in learning by experience (I do too, it's something we completely agree on), it might be instructive to the CA community to compare "bad recording engineer" Brian Lucey's experience to yours, don't you think ?

 

So, with no further ado, and for everyone's convenience, here's Brian Lucey on AllMusic, and here's Lee Scoggins on AllMusic. Both, of course, have more experience than yours truly, and of course, experience isn't just quantitative, but it does take time to build, as Mr Scoggins so rightly pointed out.

 

For others, if you give value to Mr Scoggins' opinion and advice, and since experience is an important part of whose opinion you should listen to in a debate according to him, I think we can agree that, based on his recommendation and criteria, you should trust Brian Lucey's opinion rather than his.

[Norton]

1 hour ago, firedog said:

And you still haven't given any kind of substantial answer to your claim that MQA is needed to get more hi-res released to streaming. 

MQA isn't necessary for streaming hi-res. All those files being released for streaming as MQA could be released not in MQA. 

 

That may be the case, but yesterday I was streaming the Barenboim Elgar symphonies, glitch free wirelessly over my sub-par network,  via Tidal/MQA at 96 kHz ( the same resolution  as the non MQA downloads)  to my non MQA DAC.    These are recordings I am very familiar with and they sounded impressive.  

 

Theoretically your argument may be correct, but I'm not aware, in practice, of  many other services (Qobuz possibly?) that stream reliably at 96 kHz. In 2018, streaming  RBCD still seems a big deal (and a cost option) for most.

 

I'm sure MQA would argue they actually support streaming up to 384kHz, but I am aware that subsequent unfolds are consider by many to be simply proprietary upsampling and I don't have a MQA DAC by which to judge comparative SQ.

 

 

[Norton]

Deleted

[asdf1000]

15 minutes ago, Norton said:

I'm sure MQA would argue they actually support streaming up to 384kHz

 

There are MQA 353kHz titles available in Tidal already (see attached below). This title has a DXD copy that you can purchase, so I assume this was the master used.

 

Also, the streaming itself is only ever at 24/44k or 24/48k

 

I don't have an MQA DAC btw but I do use Audirvana sometimes if I want to listen to MQA (first unfold) to try and compare.

 

I don't see/hear the fuss personally, in terms of better SQ. CD quality can still sound stellar to me.

 

I'm more concerned about the DRM potential.

 

[Fokus]

28 minutes ago, Norton said:

I'm sure MQA would argue they actually support streaming up to 384kHz, but I am aware that subsequent unfolds are consider by many to be simply proprietary upsampling

 

Not 'considered'. 'Considered' still has the possibility that it is not more than an opinion, whereas it is a hard, objective fact that after the (misnamed) 'first unfold' only upsampling happens. And this upsampling remains under the MQA aegis until the original master's sampling rate is reached.

 

This is the process for a 384k master:

 

-downsample to 192k with leaky MQA filter

-downsample to 96k with leaky MQA filter

-fold into 48k space using Quadrature Mirror Filter pair 1 (*)

=============================================

-unfold 48k to 96k using Quadrature Mirror Filter pair 2

-upsample to 192k using leaky filter

-upsample to 384k using leaky filter

-light the blue LED

 

 

(* As an aside: the output of QMF1 is what people without MQA decoding have to listen to.

QMF1 has to be optimised to allow a lossless split-join in the origami folding step. This is an

extremely limiting constraint. This means that QMF1 cannot likely be optimised, at the same time,

for optimal sound quality for non-MQA listening. This is mathematics.)

 

[Thuaveta]

53 minutes ago, Norton said:

 

That may be the case, but yesterday I was streaming the Barenboim Elgar symphonies, glitch free wirelessly over my sub-par network,  via Tidal/MQA at 96 kHz ( the same resolution  as the non MQA downloads)  to my non MQA DAC.    These are recordings I am very familiar with and they sounded impressive.  

 

Theoretically your argument may be correct, but I'm not aware, in practice, of  many other services (Qobuz possibly?) that stream reliably at 96 kHz. In 2018, streaming  RBCD still seems a big deal (and a cost option) for most.

 

I'm sure MQA would argue they actually support streaming up to 384kHz, but I am aware that subsequent unfolds are consider by many to be simply proprietary upsampling and I don't have a MQA DAC by which to judge comparative SQ.

 

 

 

What you might be witnessing might not be the genius of MQA, but rather the hard work of many network engineers. You also happen to, maybe inadvertently, be making the exact same point that a number of others have been making, which is that bandwidth is abundant enough that MQA-as-compression is a solution looking for a problem, and not a solution to a problem.

 

Let's break it down.

 

@Miska posted a filesize comparison between MQA and FLAC right here.

 

TL;DR: FLAC, at identical resolution, is, give or take, 30% smaller. A 44.1/24 MQA file, which if I'm not mistaken unfolds to 96Khz, is only about 6% smaller than 176.4/18 FLAC. Since you've determined that your playback chain can deal with either, which would you rather have: MQA, or 176.4/18 ?

 

You can argue that Miska's post is based on a single file, and I agree with you that it'd be nice to have a wider pool to compare to, if only because the bitrate of FLAC compression fluctuates based on a number of factors, including content, so let's have a quick look at what it generally takes to stream FLAC, with a simple, comparative criteria to see if your "sub-par" network (I'm sure it isn't ) is fast enough to reliably do that.

 

Uncompressed CD-DA (or redbook) is 1,411.2 kbps.

96/24 FLAC is around a 1500 kbps (let's say 2000 Kbps to be comfortable).

192/24 FLAC is around twice that, let's make it 4000 Kbps.

 

You know what else is around those numbers ? That HD button on YouTube. According to Google, 720p (not 1080p, not 4k, but lowly 720p) is between 1,500 and 4,000 Kbps. Does that ballpark remind you of anything ?

 

In practical terms, if you have bandwidth enough to stream 720p YouTube videos, and assuming your streaming service has a good CDN (which they should, that's part of what you're paying them for, after all), you could just as easily stream HiRes FLAC rather than MQA. Put in wider terms, the pool of consumers that could comfortably stream FLAC is smaller than those who can comfortably stream Netflix in HD. And to keep with your anecdote, if you can reliably stream MQA, you could reliably stream uncompressed redbook. Doesn't that make you want to go out and hug one of those hardworking networking engineers at the manufacturer for you networking gear, at Tidal, and at your ISP, that made it possible ?

[Norton]

25 minutes ago, Fokus said:

 

Not 'considered'. 'Considered' still has the possibility that it is not more than an opinion, whereas it is a hard, objective fact that after the (misnamed) 'first unfold' only upsampling happens. And this upsampling remains under the MQA aegis until the original master's sampling rate is reached.

 

This is the process for a 384k master:

 

-downsample to 192k with leaky MQA filter

-downsample to 96k with leaky MQA filter

-fold into 48k space using Quadrature Mirror Filter pair 1 (*)

=============================================

-unfold 48k to 96k using Quadrature Mirror Filter pair 2

-upsample to 192k using leaky filter

-upsample to 384k using leaky filter

-light the blue LED

 

 

(* As an aside: the output of QMF1 is what people without MQA decoding have to listen to.

QMF1 has to be optimised to allow a lossless split-join in the origami folding step. This is an

extremely limiting constraint. This means that QMF1 cannot likely be optimised, at the same time,

for optimal sound quality for non-MQA listening. This is mathematics.)

 

 

I suspect the fact that we are currently discussing MQA is incidental, could just as well be about vinyl.

 

Fundamentally this is "objective, v. "subjective".  You raise technical objections about MQA, I come  back with the fact that my DAC tells me it's receiving a MQA stream at 96kHz which in turn sounds to me every bit as good, if not better,  than my 24/96 non-MQA download copy.

 

Bearing in mind that it is most likely to be used in streaming services, concerns about DRM specific to MQA seem at best to be "deckchair rearrangement"

[Fokus]

11 minutes ago, Norton said:

 I come  back with the fact that my DAC tells me it's receiving a MQA stream at 96kHz which in turn sounds to me every bit as good, if not better,  than my 24/96 non-MQA download copy.

 

Even if MQA's sound was utterly and consistently sublime, MQA itself, and especially its current implementation, would still be a very bad idea for the consumer.

[Norton]

41 minutes ago, Thuaveta said:

 

What you might be witnessing might not be the genius of MQA, but rather the hard work of many network engineers. You also happen to, maybe inadvertently, be making the exact same point that a number of others have been making, which is that bandwidth is abundant enough that MQA-as-compression is a solution looking for a problem, and not a solution to a problem.

 

Let's break it down.

 

@Miska posted a filesize comparison between MQA and FLAC right here.

 

TL;DR: FLAC, at identical resolution, is, give or take, 30% smaller. A 44.1/24 MQA file, which if I'm not mistaken unfolds to 96Khz, is only about 6% smaller than 176.4/18 FLAC. Since you've determined that your playback chain can deal with either, which would you rather have: MQA, or 176.4/18 ?

 

You can argue that Miska's post is based on a single file, and I agree with you that it'd be nice to have a wider pool to compare to, if only because the bitrate of FLAC compression fluctuates based on a number of factors, including content, so let's have a quick look at what it generally takes to stream FLAC, with a simple, comparative criteria to see if your "sub-par" network (I'm sure it isn't ) is fast enough to reliably do that.

 

Uncompressed CD-DA (or redbook) is 1,411.2 kbps.

96/24 FLAC is around a 1500 kbps (let's say 2000 Kbps to be comfortable).

192/24 FLAC is around twice that, let's make it 4000 Kbps.

 

You know what else is around those numbers ? That HD button on YouTube. According to Google, 720p (not 1080p, not 4k, but lowly 720p) is between 1,500 and 4,000 Kbps. Does that ballpark remind you of anything ?

 

In practical terms, if you have bandwidth enough to stream 720p YouTube videos, and assuming your streaming service has a good CDN (which they should, that's part of what you're paying them for, after all), you could just as easily stream HiRes FLAC rather than MQA. Put in wider terms, the pool of consumers that could comfortably stream FLAC is smaller than those who can comfortably stream Netflix in HD. And to keep with your anecdote, if you can reliably stream MQA, you could reliably stream uncompressed redbook. Doesn't that make you want to go out and hug one of those hardworking networking engineers at the manufacturer for you networking gear, at Tidal, and at your ISP, that made it possible ?

 

Again, this is theory vs practice.  My  point simply as a civilian audiophile is that Tidal/MQA is offering me what I consider to be, in certain cases at least, true  96kHz streaming today.  I'm not aware that anyone else offers that, other than maybe Qobuz.

 

According to the 2l site, a MQA file is c. half the size of the 24/96 equivalent, I'm sure that has something to do with it.

[Norton]

23 minutes ago, Fokus said:

 

Even if MQA's sound was utterly and consistently sublime, MQA itself and especially its current implementation, would still be a very bad idea for the consumer.

 

Presumably if I listen to Wagner conducted by Gergiev via MQA, in some people's eyes I'm going straight to hell? My point being once you go down that argument path, where do you stop?

[fung0]

2 hours ago, Lee Scoggins said:

Oh shut up.  We are plenty experienced to not be at the whim of a first impression.  I've heard valid MQA demos on several occasions.  And you don't even understand how process standards like authentication differ from DRM. 

 

The real problem here are armchair idiots like yourself who can't respect a differing opinion.  Go back to school and take a Debating 101 course and learn how to present some points backed up by some evidence before slamming the journalists who are working hard to explore audio.

 

"Oh shut up?" Really? I know that several posters have been, shall we say 'forthright' in their criticism of your writing, but I am not one of them. I am merely suggesting, politely, that you are mistaken about MQA - not because you're stupid, or because you're corrupt, but because, like any of us, you can be led astray by your preconceptions, by peer pressure and by cunning marketing tactics.

 

As it happens, I'm anything but an "armchair idiot." I'm an engineer, an author and a journalist who's been writing about things like DRM probably since before you figured out how to plug in an amp. And I've written enough reviews to know how that process works too. Pack journalism has wrought havoc in the markets I've written about over the years, and allowed all sorts of technological monstrosities to thrive. That's why it's important to use wisely whatever leverage we've got.

 

I also know how to do research, and I'm afraid yours is the opinion that lacks evidence. So far, the weight of both theory and experiment is overwhelming: if MQA has any sonic advantage over high-res PCM, it's subtle and subjective at best. In fact, it's pretty certain that MQA actually reduces fidelity, but I'm willing to give MQA the benefit of the doubt on this - because 'close' is simply not good enough, when an entire industry is being herded backward into proprietary formats and DRM.

 

All the arguments you've put forward here and in your column have been thoroughly debunked. But, most importantly, the idea that MQA is the only way we can have a huge library of high-res music is absolute rubbish... provided that we dig our heels in and refuse the ugly compromise that we're currently being offered.

 

My previous post was sincere: I think if you read this thread with an open mind, if you can truly accept the possibility that MQA is nothing more than an industry Trojan horse, you could do a lot of good for audio consumers.

[Fokus]

21 minutes ago, Norton said:

 Tidal/MQA is offering me what I consider to be, in certain cases at least, true  96kHz streaming today.  I'm not aware that anyone else offers that, other than maybe Qobuz.

 

There is nothing wrong with a streaming provider using proprietary technology to transport data between their servers and their proprietary app as installed on your PC, phone, whatever.

 

It gets totally wrong when, without any valid justification, you are required to purchase a new DAC (presumably ditching the old one), and in the case of a fully digital system (i.e. a system where the DAC cannot be replaced), to purchase a large part of that system anew (presumably ditching the old bits).

 

Now from a manufacturer's point of view ...

[Norton]

16 minutes ago, Fokus said:

 

There is nothing wrong with a streaming provider using proprietary technology to transport data between their servers and their proprietary app as installed on your PC, phone, whatever.

 

It gets totally wrong when, without any valid justification, you are required to purchase a new DAC (presumably ditching the old one), and in the case of a fully digital system (i.e. a system where the DAC cannot be replaced), to purchase a large part of that system anew (presumably ditching the old bits).

 

Now from a manufacturer's point of view ...

 

Maybe, but I'm just enthusing about the initial decode to 96kHz, in my case via XXHighEnd.  Based on my 20 years' listening experience across virtually all formats, no one can tell me I'm not getting a great sound.  And new DAC not required.

[Fokus]

4 minutes ago, Norton said:

Maybe ... And new DAC not required.

 

Enjoy.

 

But know that in the original plans MQA had for us this scenario was explicitly not allowed. You had to buy a new DAC.

 

I'd like to think that after the initial wave of critique on various forums in 2015-2016 (but not in the audio press!)  MQA compromised and opened the door for partial decoding in software. But so far this decoding is only available from Tidal and from the Node2 streamer.

[asdf1000]

26 minutes ago, Fokus said:

partial decoding in software. But so far this decoding is only available from Tidal and from the Node2 streamer.

 

And currently Audirvana also. Roon and Amarra confirmed to be coming too.

[firedog]

2 hours ago, Norton said:

Theoretically your argument may be correct, but I'm not aware, in practice, of  many other services (Qobuz possibly?) that stream reliably at 96 kHz. In 2018, streaming  RBCD still seems a big deal (and a cost option) for most.

 Qobuz, some classical sites, some other smaller download sites. 

The point is that the reason it isn't being done isn't file size. It's not being done b/c potential providers apparently don't see it as a moneymaker.  Why pay to provide something that isn't in general demand and that you can't get many customers to pay for?

 

Streaming - even mp3 - still isn't a moneymaker. The CD level of streaming at Tidal got it a chunk of audiophile customers, but made ZERO impact on the larger market - they didn't grow tremendously and/or snatch large amounts of customers from the other streaming services -  and they are mostly streaming mp3, not even CD. Tidal apparently hasn't even kept up with the growth of the big boys over the last year or two. 

 

Which is exactly why you should be suspicious of MQA and why the labels are supporting it. They are only doing it for what they see as a future payoff. And I'm willing to bet the payoff doesn't have the best interests of the audiophile at heart - either sonically or financially. 

[Shadders]

5 hours ago, Lee Scoggins said:

I don't just accept any prevailing opinions of this board, regardless of how many audiophiles are on it.  I've been doing professional recordings since 1989 and was on the cutting edge of hirez before it even launched.  And my opinions will be informed by my recording work, research, interviews with industry experts, and listening test.  If my thoughts on the topic don't please members of CA then I really don't care.  

 

I have read all 249 pages of this discussion over the past year and it's quite clear that many here have an axe to grind.  If the evidence against MQA was so obvious, there wouldn't be literally dozens of members here responding with personal attacks.

Hi Lee,

Since you have read all 249 pages, then why have you not understood and accepted the evidence provided on CA forum, that MQA have lied ?

 

Opinion is one thing, but facts are facts, and for those statements regarding MQA technical aspects, it has been proved that MQA has lied.

 

So, i am unsure as to what your purpose is here, and on your website.

1. Are you commenting on MQA as an opinion only ?.
2. Do you repeat only the MQA supplied facts ?
3. Are you discounting the evidence presented on this forum because of arguments between posters ?
4. Are you a journalist ?

Thanks and regards,

Shadders.

[FredericV]

1 hour ago, Fokus said:

 

Enjoy.

 

But know that in the original plans MQA had for us this scenario was explicitly not allowed. You had to buy a new DAC.

 

I'd like to think that after the initial wave of critique on various forums in 2015-2016 (but not in the audio press!)  MQA compromised and opened the door for partial decoding in software. But so far this decoding is only available from Tidal and from the Node2 streamer.

So the node2 can do the first unfold to the digital output? So the samplerate of MQA via the digital output is up to 96 Khz? Or do they also do the second unfold and expose this on the digital out (so in case of first unfold 96 K, this would mean 192K on the digital output).
 

[asdf1000]

7 minutes ago, FredericV said:

So the node2 can do the first unfold to the digital output? So the samplerate of MQA via the digital output is up to 96 Khz? Or do they also do the second unfold and expose this on the digital out (so in case of first unfold 96 K, this would mean 192K on the digital output).
 

 

The Node 2 can only do the first unfold (up to 96k) via it's digital outputs.

 

The 2nd unfold is only possible via it's analogue outputs.