MQA is Vaporware

[mcgillroy]

3 minutes ago, Nikhil said:

 

+1  

 

I also don't know how they missed seeing the much bigger potential of the cell phone market.    For all the effort they have poured into pitching MQA to the audio industry, the cell phone / mobile industry would have dropped a couple of milllion dollars without thinking much of it.   This is completely their environment.  Another badge on the feature list to show something new - right next to the MP3 logo.     

 

 

 

Actually I believe that MQA was at least in part aimed exactly at the cell-phone market. Namely at Apple and Google which dominate this market - everybody else is also-runs loosing money.

 

Apple & Google not only sell phones and their operating systems, they provide platforms where music-streaming is part of the consumption-options. As such they basically control access to more than 90% of the listeners and certainly a sizeable chunk of the profits. The coming wave of smart-speakers will only extend this reach.

 

The labels are at their mercy. 

 

So what do you do if you have the catalogs but not the access to the customers?

 

You invent a licensing regime that promises to meter the endpoints where music is consumed. These being the DACs which are in every Smartphone, Smartspeaker etc. You call it "authentication" cause you actually cryptographically authenticate that the vendor has payed MQA a licensing fee for every DAC-instance deployed. Add some bla-bla about quality and mastering for the consumers and your got a nice marketing angle to your little license game taking aim at Apple, Google and co.

 

Given that you are betting on getting into eventually billions of Smartphones and a good number of millions of smart speakers even a very small licensing fee would mean $$$.

 

Deals in tech typically involve license and/or patent negotiations and exchanges. Your licence and patent portfolio is the most important asset you have.

 

MQA is a gamble to bolster the labels license and patent portfolio.

 

It could have worked if they didn't had it botched so badly with their "let's punk the audiophiles-first" marketing strategy.

[Nikhil]

9 minutes ago, mcgillroy said:

 

Deals in tech typically involve license and/or patent negotiations and exchanges. Your licence and patent portfolio is the most important asset you have.

 

 

Exactly.  MQA has done all the legwork on the patents etc   But where did they go with all that?   The audio industry companies are quite frankly not making a whole of money in comparison.  How much value do you think they created with that move?    In comparison the cell phone industry - with likes of a Samsung who would genuinely pay for something that would give them an edge over Apple IMO - would have been an incredible move.  

 

Apart from the cell phone industry, the automobile industry is a very good target.  Just ask Bose.

 

 

[mansr]

8 hours ago, Don Hills said:

Do you know if it's a decoder or just a renderer - will require a core decoder in the app? (I suspect renderer only.)

There's no such thing as an MQA decoder or renderer inside a DAC chip. It's all done in software on the application processor. The Oreo requirement probably has something to do with how the Android audio subsystem handles various sample rates, volume control, etc.

[mansr]

On 10/26/2017 at 9:37 PM, mansr said:

I'll publish them somewhere tomorrow.

https://code.videolan.org/mansr/mqa/blob/master/render-filters.txt

[Shadders]

26 minutes ago, mansr said:

https://code.videolan.org/mansr/mqa/blob/master/render-filters.txt

Hi mansr,

Thanks a lot - much appreciated.

Regards,

Shadders.

[daverich4]

On 10/27/2017 at 3:44 PM, Charles Hansen said:

. The article I saw said that iOS 11 was ready for all phone with 64-bit processors, but not yet for iPads with the same processor. Keep 'em confused... 

 

Perhaps I’m misunderstanding your post but for what it’s worth, my iPad Air 2 has a 64 bit processor and is running iOS 11.0.3. 

[rickca]

This quote is from a recent John Darko article:

For DAC manufacturers operating in 2017, the call to add full MQA rendering to their products is beginning to transcend the controversy surrounding the format. The high end audio world is littered with Format-First Audiophiles who now more than ever shop with a “No MQA, No Deal” approach, especially when it comes to big ticket items like those offered by dCS.

 

Wow!  Where does he get this crap?  MQA is just making stuff up,  it's alternative facts.  This is what sales guys call a presumptive close.

[Tony Lauck]

On 10/25/2017 at 1:23 AM, esldude said:

Count me as another of those who found Nakamichi to work well with Dolby.  Did all my needledrops (not what they were called then) with Dolby C and good tapes.  Of course I played them back on Nak tape decks.  I even had Nak head units in my cars in those years.  A TD700 and TD500.  Those head units were excellent all the way around.  Not only great tape sections, the FM tuners were exemplary as well.  Still have one head unit, one home tape deck and some tapes. 

 

 

Cut me out when it has anything to do with Dolby.  In the late 1980's  I got a Nak C7-A cassette deck and used it to make live vs. recorded A-B tests of my wife's Steinway B.  I also used a Tandberg 7.5 2T reel to reel deck.  The Dolby destroyed all of the dynamics of the sound.  The only way to get acceptable sound out of the Nak was to turn off Dolby, turn the recording volume way down and live with the tape hiss.  If you recorded without Dolby at a high enough level that the hiss wasn't audible then the tape saturation destroyed the dynamics.  By contrast the Tandburg would produce a recording that was essentially equivalent to the live performance, unless you got real close to the speakers, at which point you could barely hear the tape hiss.

 

IMO, the end of classic hi-fi came when professional tape recording introduced Dolby - A which had the same deleterious effect on sound quality (albeit muted) as did Dolby B and C.  Worse, it made practical multi-channel recordings and mixed down, after which there were few record labels making purist recordings.

 

As far as I am concerned, Dolby was an evil force on audio, but made a lot of money.  And my take on MQA is that it is an ill-advised attempt to create negative technology in another "get rich quick" scam.

 

[labjr]

On 10/27/2017 at 1:20 PM, mansr said:

ROTFL. Show me one, just one, "technical" explanation from Shunyata that is even scientifically meaningful, never mind accurate.

 

Try listening to Ted Denney of Synergistic Research sometime. I demoed some cables about 20 years ago and spoke with him on the phone. I swear he was making up  some Feng Shui voodoo  crap while he was telling it to me.

[Charles Hansen]

9 hours ago, mansr said:

https://code.videolan.org/mansr/mqa/blob/master/render-filters.txt

 

Hello Mansr,

 

Thank you very much for making these available. This is extremely useful for many. For those who are unfamiliar with the nitty-gritty of digital filters, there are several points of interest:

 

1) MQA has lied through their teeth when they say that the MQA filter is "customized to correct for the deficiencies of the DAC chip itself. At least four different DAC chips from at least two different manufacturers have been examined (at least visual plots thereof, if not to the detail of the actual 12-decimal point accuracy of the coefficients themselves) and there are no visible differences whatsoever.

 

2) The sum of the coefficients in any particular digital filter is (in general) = 1.

 

3) Any time there is a negative coefficient, it means that the impulse response of the filter will drop below zero and there will be undershoot and "ringing". The "ringing" will be at the corner frequency of the filter and is presumably inaudible, yet I've conducted many tests where by I can hear this effect - I think it is safe to say that there is more to be learned about the functioning of the ear/brain hearing mechanism than our current understanding. For example bone conduction responds up to two or three time the frequency response as conduction through the air, eardrum, and middle ear. If people don't think that bone conduction is real, just check out the music of profoundly deaf master musician Evelyn Glennie. (See for example: https://www.ted.com/talks/evelyn_glennie_shows_how_to_listen)

 

There have also been specific tests whereby transducers were attached to the bone in the jaw or skull of a participant and things like transposing the pitch of speech up to ultrasonic frequencies (while maintaining the original tempo) allowed the subject to "hear" identifiable speech. It is clear that that is at least part of the "mystery" of the importance of ultrasonic sound response.

 

One place that I agree with the "objectivists" is that with a properly limited bandwidth signal these artifacts ("ringing") of the filter should never be excited. Where I disagree is that something else is going on, specifically:

 

a) It is possible that the fact that no PCM A/D converter extant uses a filter that properly limits the bandwidth of the signal is the source of audible problems. A "proper" filter is the sinc function - a physical impossibility as it would require a filter with an infinite number of taps - while the vast majority of them use a linear-phase "half-band" filter that is only down -6dB at the Nyquist frequency - an old habit left over from the days that silicon memory and computational power was super-expensive, and a linear-phase half-band filter only needs half as many taps and coefficient memory storage than a minimum-phase filter, while a half-band filter only needs half the number of taps and coefficient memory storage as any other windowed digital filter. (A "window" is a means to truncate the infinite extent of a true "theoretically perfect" sinc filter.

 

In my experience even changing the windowing function used to limit the number of taps will change the subjective sound quality of the filter. Clearly something is going on which we don't yet understand.

 

I don't believe it has ever been demonstrated that simply inserting an ultra-sharp cutoff band-limiting filter of and by itself does not change the subjective quality of the audio. Doing so with an analog filter would introduce phase shift, at least in the top octave - which may or may not be audible. The only way to do so without introducing phase shift would be to implement the filter digitally - which merely continues the "digital vs analog" debate.

 

b) It is possible that there are events that occur during the processing or mixing of the digital signal that create signals on a digital audio file that (just like the artificially-generated single full-value impulse test signal) excite the misbehavior of one of the many digital filters in the record/playback chain.

 

In other words, I am following the example of the late Richard C. Heyser. I don't start with the theory and let it tell me what I can or cannot hear. Instead I observe (in this case listening, in the case of Newton looking at a falling apple) events in the real world and then try to understand what is the underlying mechanism that creates those events.

 

[Charles Hansen]

5 hours ago, daverich4 said:

 

Perhaps I’m misunderstanding your post but for what it’s worth, my iPad Air 2 has a 64 bit processor and is running iOS 11.0.3. 

 

My understanding is that you should be able to import a FLAC file and play it merely by double-clicking (or whatever non-intuitive "gesture" Apple now uses). Can you play FLAC files now without the assistance of a 3rd-part app?

 

Thanks!

[labjr]

On 10/26/2017 at 6:36 AM, Charles Hansen said:

 

Hi Rick,

 

Prezactly. For more info on this, please refer to:

https://www.audioasylum.com/forums/critics/messages/8/88367.html

 

Wow! Been away for while and hadn't seen this thread. Glad to read an opinion like this from a manufacturer.

[Charles Hansen]

2 hours ago, rickca said:

This quote is from a recent John Darko article:

“No MQA, No Deal”

 

Wow!  Where does he get this crap?  MQA is just making stuff up,  it's alternative facts.  This is what sales guys call a presumptive close.

 

Great question! With TAS we are almost certain it was pure bribery. With Stereophile it is becoming clearer and clearer that JA considers Bob Stuart to be one of his "heroes", so when given a deliberately deceptive demo fell for it hook, line, and sinker.

 

But with Darko I don't know. I've pushed back on him many times via private e-mail correspondence. It is likely one of the above (or a combination). but I don't know which.

[Charles Hansen]

1 hour ago, Tony Lauck said:

Cut me out when it has anything to do with Dolby.

As far as I am concerned, Dolby was an evil force on audio, but made a lot of money.  And my take on MQA is that it is an ill-advised attempt to create negative technology in another "get rich quick" scam.

 

Hi Tony,

 

Agree 100%. The problem with Dolby is that unlike the "purist" approach that John Curl started with his JC-2 preamp for Mark Levinson - the first commercial preamp in the world to drop tone controls (and "loudness buttons"), Dolby noise reduction schemes sent the analog signal through a horribly complex maze of additional circuitry and filters. All else being equal, a simpler signal path is going to sound better than a long convoluted one - especially when Dolby B was implemented on consumer-grade tape decks with ICs and terrible-sounding capacitors.

 

Even in the pro world where Dolby A was used (with four separate bands of compansion rather than the single band of Dolby B used for consumer products), I know of at least one recording that was made with a pro deck using Dolby A. When the artist heard it played back he was horrified by the poor sound quality.  The only thing that could be done at that point was to play back those tracks (a complete symphony orchestra!) without the Dolby engaged on the playback side. Even though this caused measurable errors in the frequency response versus level and strange frequency-related compression artifacts, the artist felt that it was better to leave it like that than add another layer of horrible sounding Dolby circuitry to those tracks (which subsequently were part of the overall mix of many other instruments and vocals).

[labjr]

7 minutes ago, Charles Hansen said:

 

Great question! With TAS we are almost certain it was pure bribery. With Stereophile it is becoming clearer and clearer that JA considers Bob Stuart to be one of his "heroes", so when given a deliberately deceptive demo fell for it hook, line, and sinker.

 

But with Darko I don't know. I've pushed back on him many times via private e-mail correspondence. It is likely one of the above (or a combination). but I don't know which.

 

Yeah,  Darko suddenly flipped one day like someone got to him. Maybe he took an immunity deal.  I don't read his blog anymore for various reasons. I noticed he was censoring everything and deleting parts of my posts. I'm Done.

[Charles Hansen]

12 minutes ago, labjr said:

 

Yeah,  Darko suddenly flipped one day like someone got to him. Maybe he took an immunity deal.  I don't read his blog anymore for various reasons. I noticed he was censoring everything and deleting parts of my posts. I'm Done.

 

Interesting that he would censor your posts. I've no idea what they were about. But it is equally interesting that any time there was any mention of MQA, (paid?) fanboy Peter Veth would make dozens of comments on the article, which Darko had no problem publishing.

 

Which of your viewpoints was Darko trying to suppress?

[Charles Hansen]

16 hours ago, mcgillroy said:

Actually I believe that MQA was at least in part aimed exactly at the cell-phone market. Namely at Apple and Google which dominate this market - everybody else is also-runs loosing money.

 

Apple & Google not only sell phones and their operating systems, they provide platforms where music-streaming is part of the consumption-options. As such they basically control access to more than 90% of the listeners and certainly a sizeable chunk of the profits. The coming wave of smart-speakers will only extend this reach.

 

The labels are at their mercy. 

 

It could have worked if they didn't had it botched so badly with their "let's punk the audiophiles-first" marketing strategy.

 

Actually Apple and Google also lose money by streaming if you look at that business in isolation. But there are other offsetting compensations that allow Apple and Google to make money in other ways.

 

As an example, even though the music industry completely screwed themselves by giving 30% of the profit to Apple, and then allowing Apple to sell the individual songs on every album at 1/12 the album price (the old model in the days of vinyl was to sell the best song and a "throw-away" together for 1/4" of the album price, and if you wanted more songs than that you had to buy the entire album), Apple never made any money directly from iTunes.

 

Instead, iTunes was just a way to sell highly profitable iPods, which in turn pushed people towards buying highly profitable Mac computers, and highly profitable iPads. It's the exact same business mode where updates to the operating system are "free". Since the Mac OS only runs on Apple products,  they use it as a loss-leader to boost sales of their other highly profitable products.

 

~~~~~~~~~~

 

Yes, at the present times the labels are at the mercy of the tech companies - first through the sales of downloads, and even more so through streaming. Currently over half the revenue of the record labels comes from royalties through streaming. YouTube (aka Google aka Alphabet) currently pays billions to the labels for all of the music that is streamed at 126kb/s (not 128 - I don't know why) on YouTube - which is pretty much everything out there except for a handful of artists (not record labels) who have demanded YouTube remove their copyrighted material.

 

The details of the real story are incredibly complex, but the basis of it is very simple -  money and greed.

 

By the way, I know for a fact that Google (at least) is completely and totally anti-MQA. So if either Bob Stuart - or even the record labels - think they can push MQA through, when Google is opposed, then Bob Stuart (and the labels) are deluding themselves. Unless the record labels think that they can live without the billions in revenue they currently receive from Google. I would assume that Apple's position is similar to Google's.

 

In a way it would be great for the record label to be independent of the tech giants. But if it comes at the cost of adding MQA to the mix, it's not worth it to me, at least. Google and Apple sell crap to people who don't care about quality. MQA removes quality and there is a strong potential that the record labels would stop making non-DRM'ed high-res (or even Redbook quality) files available if MQA ever becomes successful.

 

But at this point I don't think we have to worry. Google has been monitoring the situation fairly closely. Their assessment (particularly after the latest AES convention) is that MQA has already failed. Dead. Done. Not worth even worrying about. Interesting, eh?

[labjr]

31 minutes ago, Charles Hansen said:

 

Interesting that he would censor your posts. I've no idea what they were about. But it is equally interesting that any time there was any mention of MQA, (paid?) fanboy Peter Veth would make dozens of comments on the article, which Darko had no problem publishing.

 

Which of your viewpoints was Darko trying to suppress?

 

It's probably been a year since I've been on the site. But most of my posts were similar to stuff I post here. He would delete sarcasm, argumentative language and anything that seemed remotely negative. Looked to me like he was trying to fine tune every little thing so it looked pretty for his advertisers. 

 

 

 

Edit: As it was, it was hard to have a conversation because it took sometimes an entire day for him to moderate posts. 

[rickca]

1 hour ago, Charles Hansen said:

the music industry completely screwed themselves by giving 30% of the profit to Apple

What if MQA did the same thing? Of course, at the moment there is no profit.

[Bob Stern]

On 10/22/2017 at 5:35 AM, mcgillroy said:

The main news here is that nobody less than Bruno Putzeys comes out as a vocal MQA sceptic.

 

He asked „scathing questions” & classifies MQAs bla, bla as “speculations about neuroscience and physics.”

 

Putzeys posted the following on Facebook, which Siegfried Linkwitz quoted on his website at  http://linkwitzlab.com/links.htm#Putzeys:

 

Yesterday there was an AES session on mastering for high resolution (whatever that is) whose highlight was a talk about the state of the loudness war, why we're still fighting it and what the final arrival of on-by-default loudness normalisation on streaming services means for mastering. It also contained a two-pronged campaign piece for MQA. During it, every classical misconception and canard about digital audio was trotted out in an amazingly short time. Interaural timing resolution, check. Pictures showing staircase waveforms, check. That old chestnut about the ear beating the Fourier uncertainty (the acoustical equivalent of saying that human observers are able to beat Heisenberg's uncertainty principle), right there.

At the end of the talk I got up to ask a scathing question and spectacularly fumbled my attack*. So for those who were wondering what I was on about, here goes. A filtering operation is a convolution of two waveforms. One is the impulse response of the filter (aka the "kernel"), the other is the signal.

A word that high res proponents of any stripe love is "blurring". The convolution point of view shows that as the "kernel" blurs the signal, so the signal blurs the kernel. As Stuart's spectral plots showed, an audio signal is a much smoother waveform than the kernel so in reality guess who's really blurring whom. And if there's no spectral energy left above the noise floor at the frequency where the filter has ring tails, the ring tails are below the noise floor too.

A second question, which I didn't even get to ask, was about the impulse response of MQA's decimation and upsampling chain as it is shown in the slide presentation. MQA's take on those filters famously allows for aliasing, so how does one even define "the" impulse response of that signal chain when its actual shape depends on when exactly it happens relative to the sampling clock (it's not time invariant). I mentioned this to my friend Bob Katz who countered "but what if there isn't any aliasing" (meaning what if no signal is present in the region that folds down). Well yes, that's the saving grace. The signal filters the kernel rather than vice versa and the shape of the transition band doesn't matter if it is in a region where there is no signal.

These folk are trying to have their cake and eat it. Either aliasing doesn't matter because there is no signal in the transition band and then the precise shape of the transition band doesn't matter either (ie the ring tails have no conceivable manifestation) or the absence of ring tails is critical because there is signal in that region and then the aliasing will result in audible components that fly in the face of MQA's transparency claims.

Doesn't that just sound like the arguments DSD folks used to make? The requirement for 100kHz bandwidth was made based on the assumption that content above 20k had an audible impact whereas the supersonic noise was excused on the grounds that it wasn't audible. What gives?

Meanwhile I'm happy to do speakers. You wouldn't believe how much impact speakers have on replay fidelity.
________
* Oh hang on, actually I started by asking if besides speculations about neuroscience and physics they had actual controlled listening trials to back their story up. Bob Stuart replied that all listening tests so far were working experiences with engineers in their studios but that no scientific listening tests have been done so far. That doesn't surprise any of us cynics but it is an astonishing admission from the man himself. Mhm, I can just see the headlines. "No Scientific Tests Were Done, Says MQA Founder".

 

*****

 

After quoting Putzneys, Linkwitz added his own comment:

 

My thoughts: Human hearing is a non-linear process of sound perception as can be deduced, for example, from the equal loudness contours. Hearing evolved for survival. High frequency ticks and clicks are instrumental in determining the direction to the location of a potentially threatening source. I wonder if we have hearing acuity for such type of signals that goes beyond the frequency range for steady-state stimulus perception. I doubt that hearing can be fully described in Fourier analyzer terms. If Bob Stuart truly has discovered a new perceptual phenomenon, then he needs to demonstrate it scientifically. Otherwise MQA is just a marketing ploy to resell previously recorded material in a proprietary file format and they are Phishing for Phools. - SL 

[daverich4]

14 hours ago, Charles Hansen said:

 

My understanding is that you should be able to import a FLAC file and play it merely by double-clicking (or whatever non-intuitive "gesture" Apple now uses). Can you play FLAC files now without the assistance of a 3rd-part app?

 

Thanks!

 

I’ve read that if you have the FLAC files in iCloud or Dropbox or something similar you can access and play them from there with an iPad. I didn’t try that but using the new Files App in iOS 11 I tried clicking on a FLAC file but nothing happens so in my case, I guess the answer to your question is no. 

[The Computer Audiophile]

On 10/28/2017 at 5:10 AM, mcgillroy said:

Actually I believe that MQA was at least in part aimed exactly at the cell-phone market.

 

Meridian has a pile of intellectual property that could use a home. Said another way, the company has many solutions in search of problems. 

 

MQA was at first going to be used as the solution for Pono. Not sure what problem this solution addressed, but the team at Pono (when real businessmen ran the company, not Neil Young or his industry chronies) decided MQA didn't make sense. 

 

Then MQA had to find a new problem to solve or market at which to take aim. 

[mansr]

20 minutes ago, The Computer Audiophile said:

MQA was at first going to be used as the solution for Pono. Not sure what problem this solution addressed, but the team at Pono (when real businessmen ran the company, not Neil Young or his industry chronies) decided MQA didn't make sense. 

And talk at the time included some rather heavy-handed DRM features.

[Charles Hansen]

2 hours ago, The Computer Audiophile said:

MQA was at first going to be used as the solution for Pono. Not sure what problem this solution addressed, but the team at Pono (when real businessmen ran the company, not Neil Young or his industry chronies) decided MQA didn't make sense.

 

Sources for this story, please?

 

Thanks!

[The Computer Audiophile]

2 hours ago, Charles Hansen said:

 

Sources for this story, please?

 

Thanks!

My own conversations with people involved.