MQA: "Final" thoughts... On Politics & Paradigms.

Cartoon looks about right and reflective of MQA... From article here.

In the last few weeks as I was preparing the series of posts discussing the MQA blind test, I noticed The Absolute Sound (TAS) ramping up the hype again in September with two articles in defense of the value of MQA. Let's use those articles to close off discussions about MQA among other ideas at least for a bit here at The Musings...

Sure, I'll be around to add my two bits and answer questions where I can, but to be honest, I'm a little tired of MQA by this stage and how much of a big deal it isn't, IMO.

First, consider the TAS article "The Politics of MQA" by Andrew Quint. I found the title appropriate but not so much the opening sentence: "The codec known as MQA (Master Quality Authenticated) is clearly in its ascendancy." Are we sure about that?

You see, I agree very much about the title because really it's about politics. It's about influence, control, and money. MQA is a business, it needs to generate revenue, and to do so it must gain adoption of course. In contrast, the average audiophile when presented with another "new format" cares about the utilitarian aspects of what is presented (actual sound quality potential) and value from the purchase if adopted. The frustration I think this new file type brings is a result of this dissociation. The company and mouthpieces insist that they have produced a "technology" of value because it brings the "studio sound" home, that it's "revolutionary", that it embodies a new "paradigm" (more below), and that we will find benefit. It does this with ostensibly something significant - temporal "de-blurring". But for audiophile consumers, it's hard to imagine what the company, advertisers, and promoters are talking about... Where is the "blurring" exactly to begin with? On what scientific basis (for those who wonder about these things)? Even if true that the technique improves temporal accuracy in a certain way, what evidence is there that this really is appreciable to music lovers (ie. where are the comparisons even in audio shows right from the start to show an openness and confidence in what's being sold)? From my perspective, looking at what has transpired as a whole, the sales job, whether from Bob Stuart or the audiophile press, clearly hasn't been engaging in a way I think things need to be in this day of vastly improved communications - especially when dealing with passionate and knowledgeable members of such a small hobby!

Remember that doubts about the claims made in the audiophile world are nothing new! Consider the context of audiophilia as a whole. After years of obviously questionable claims from "experts" in the audiophile media promoting mindless tweaks, crazy cables, and preposterous theories, is it any wonder that there would be "blowback" when consumers yet again are subjected to the aroma of more politics (and the attendant profit motive) rather than what we truly care about? Just some no-nonsense technological advancement which actually makes sense and can be presented without hyperbole.

"Experts in general are viewed with suspicion nowadays." Maybe this is more true these days overall, I don't know, but specifically in the "high end audio" world, why would anyone not question such "authority"? It's hilarious to see the next sentence advocating that readers have a look at Tom Nichols' book The Death of Expertise but conveniently not showing the full title - The Death of Expertise: The Campaign Against Established Knowledge and Why It Matters. Are MQA's claims "established knowledge", Mr. Quint? Are those who view MQA critically (presumably including my posts here) trying to dissuade anyone from scientific inquiry, exploration of empirical truths, or in any way undermining decades of engineering and scientific principles; you know... established knowledge?! Which group truly has the weaker critical thinking skills?

Here's another gem: "... the MQA critiques I've seen are conspicuously silent on the extensive psychoacoustic research that underlies the development of the codec..." Here too is another claim that MQA wants us to be impressed by. But it's simply not impressive because even MQA says little about the research they claim to be so influential! (For reference, I've discussed some of this research here, here, here and here over the years. Also, I see that MQA likes to reference a 2013 Oppenheim paper on how "Human Time-Frequency Acuity Beats the Fourier Uncertainty Principle" and yet there's this comment refuting claims worth considering thanks to the National Research Council of Canada.)

Yes, research can refine our understanding of the human hearing and perception organs, but let's not be naïve thinking that any of this research somehow has opened a doorway into remarkable levels of audible resolution. There are no major theories or psychoacoustic breakthroughs that high-fidelity equipment or file formats can suddenly exploit! [Consider over the years all the claims in the bizarro world that companies like Synergistic "Research", Machina Dynamica, Shun Mook, cottage industry cables, questionable software, etc... operate in with their claims of sonic improvement based on incomprehensible science.]

The final comment I want to make about this article is this statement in the last paragraph: "Let’s face it: There are some audiophile elitists that are going to resent this challenge to their special status." Huh? Who are the elites in this fairy-tale opposing MQA? Are not those who promote "high end" products, those who have direct access to the manufacturers, those who believe that they are endowed with the gift of "golden ears" and can value the benefits of $10,000+ audio cables (like these "beauties" in your magazine) at least mouthpieces for these elites? In any event, I fail to see how this little codec which appears to make little empirical difference pose any challenge to one's pride in owning a nice, expensive sound system with all its non-utilitarian benefits. As if the megabuck elites would have any trouble coughing up a few thousand to upgrade a DAC should MQA actually take off. Specifically then, about whom are you referring to?!

As I said, I agree that MQA is political. With little to no evidence that it sonically makes a difference whether technically or empirically in listening tests, the only people that would gain from its adoption would be companies with a stake in obtaining influence and financial gain (primarily MQA Ltd.). Music labels would probably also prefer something like this so they don't need to release original "studio master" hi-res files (the "crown jewels" as discussed here) - saved for yet another remaster down the road.

As usual, lots of claims, opinions and allegations, no demonstration of facts from the politicians. Obviously embodied in the emptiness of this article.

Then there's "Let the Revolution Begin" by Robert Harley.

Hmmm, where shall I begin? Let's see, in the last 20 years, I have been an "early adopter" of many new technologies and media formats. I bought the first Creative Labs PC-DVD drive and decoder card on the market in the late 90's, purchased some of the first DVD's when they were still single layer flip disks, got into SACD with a Sony SCD-1 for awhile, bought a few DVD-A's but wasn't enthralled by the experience. I enjoyed both Blu-Ray and HD-DVD and still have an old Xbox 360 HD-DVD drive in storage. By the early 2010's, I started purchasing hi-res audio downloads for select albums that seemed to be of high resolution provenance.

In the last year I've been enjoying my 4K/HDR TV, bought my first UHD Blu-Rays, upgraded to an octa-core CPU, and in the last few weeks bought a new electric vehicle to replace my gas-guzzling car. In each of these recent purchases, I and the family experience obvious benefits... My kids clearly know which version of Planet Earth II they want to watch when given a choice between the 4K/HDR and 1080P, I know what CPU I'd want to use for encoding long videos to HEVC/H.265. It's great to know that I've driven more than a thousand kilometers already without burning any gasoline. Clearly my wife appreciates the new vehicle because we have to fight for the keys to the EV in the morning :-).

Would any of these tangible benefits be evident when it comes to choosing between MQA and another digital audio format at home (hi-res or otherwise, including high bitrate MP3)? Would my kids or wife make a deliberate decision to select an MQA-encoded album over a standard PCM? I can certainly say without reservation that this simply would not happen.

As audiophiles, in the last few decades, I think we've seen some genuine "revolutions". From a sound quality perspective, the switch from LP/vinyl/cassette to CD was a big one in the 1980's that got better over time as digital matured (at least with the albums that survived the loudness war). From the perspective of convenience, the CD allowed for high quality car playback, MP3 allowing for thousands of songs in the pocket, computer audio and storage provided high quality home access, and streaming with access to millions of tracks over the cloud are all worthy contenders for at least the adjective "significant", and "evolutionary" if not quite "revolutionary". Since the CD and 16/44 PCM encoding, we have already had our shot with a new encoding "format" paradigm. It is called SACD and DSD, and it's still with us though on life support with few new releases. But we know it was not "revolutionary" in how it impacted sound quality (nor did it add anything to convenience although multi-channel was nice). How could it when 2-channel CD sound quality was already excellent (yes, the Sony and Philips engineers did their homework more than 30 years ago).

DSD with multi-megahertz sampling rate already provided very high frequency response and excellent time domain qualities as per nice looking impulse-response graphs touted by the MQA folks to be important. Yet clearly these were not enough to overcome the barriers to mass adoption in the marketplace. In fact, DSD64 has even better dynamic range and lower noise floor than MQA in the audible spectrum. Sure, we can put some blame on the marketing approach and possibly other poor business decisions made by Sony and partners, but this doesn't excuse the core issue of limited differentiable audibility for human consumption.

MQA as typically found on TIDAL is a 24/44 or 24/48 PCM stream with a lossy-encoded piece embedded in the least significant bits in much the same way as HDCD embedded instructions under the expected noise floor with 16/44 PCM. The other part to this scheme is of course the 16 customized upsampling filters as demonstrated by the AudioQuest Dragonfly Black's measured impulse responses.

In comparison to a jump from 16/44 PCM CD to multi-megahertz sampling rate 1-bit DSD (ie. SACD), how much of a "paradigm shift" is the jump from 24/96 to MQA as described in the previous paragraph? Mr. Harley speaks of the "'crisis' in which a 'battle' (Kuhn’s terms) breaks out between followers of the old and new paradigms" implying that it's happening right here and now. But the "crisis" (let's not be so dramatic - this is more a "disagreement") we're engaged in is not about science and demonstrable sonic enhancement but of politics as alluded to by TAS's own article above!

For Mr. Harley to frame this debate on a grand scale of Kuhn's scientific revolution is laughable and obviously hyperbolic. It damages further a magazine which IMO has little credibility when it comes to being able to show critical thinking and understanding of the technology (remember extremely strange articles like this?). It also exemplifies to the world the madness of audiophilia and the ineptitude of some of the "journalists" in this hobby among other technology-based enthusiast pursuits. These are the esteemed journalists that audiophiles are supposed to respect? This is the level of intellectual discourse we engage in to divine whether a new and "revolutionary" product honestly deserves attention (much less financial support)?

Notice how he spent many paragraphs creating the impression that MQA has embodied amazing breakthroughs in "the new psychoacoustic paradigm". Snapping twigs, cracking leaves, wind, rain, and running water have "no frequencies"! The hearing mechanism has bidirectional neural afferent/efferent pathways! Supposedly MP3 has "failed" in practice!

You see, these are all meaningless at best or erroneous at worst arguments. Yes, there are frequencies to be found when we record a twig snap :-). Nobody denies that humans are remarkably complex and as we understand more about the biological architecture, new neural network pathways will be found. No, MP3 has not "failed". In what universe is this man living in? If I had a role in creating MP3, I'd certainly be proud of the success over these decades! Considering that the vast majority of Internet streaming is still lossy, satellite Sirius XM radio is equivalent to 128-160kbps MP3, the local DJ "spins" MP3's at all the recent dance parties I've been to, and lossy audio continues to advance with codecs like Opus, there's nothing to be ashamed of. I'm obviously not saying that audiophiles should not embrace lossless audio or hi-res, rather, in the vast majority of situations, an MP3 of reasonable bitrate would be more than adequate to successfully encode the sound at a high enough resolution.

None of these arguments can be linked to anything MQA actually does. I need to point out that he is creating "straw men" arguments. Just because nerve fibers go this and that direction doesn't mean MQA makes anything better. Likewise, so what about MP3? The issue is whether MQA actually is better than lossless PCM and hi-res PCM. Ironically it is actually MQA itself that contains lossy elements in the compression algorithm, throwing out parts of the ultrasonic frequency felt to have no useful benefit. (Not that this is unreasonable BTW, just as it's not unreasonable for MP3 to toss out some data also for the sake of compression.)

Here's an interesting quote (bold emphasis mine):

So, here we are in 2017, with our digital-audio systems designed around first-generation paradigms of information theory (Nyquist-Shannon) and psychoacoustics (frequency-based, the ear as a linear and static device). MQA comes along and forges a new path, building on the advances in other fields and developing from first principles an entirely new way of looking at the question of how best to encode, distribute, and decode digitally represented music.

Wow, doesn't that sound just grand?! Let's rewrite that paragraph and create a realistic version of what MQA is:

So, here we are in 2017, with our digital-audio systems designed around tried-and-true paradigms of information theory and psychoacoustics. MQA is forging ahead with its proprietary, partially lossy compressed 24/44 or 24/48 PCM stream with a little bit extra to tell the DAC how it should dither and with which of 16 weak, poorly anti-imaging, upsampling reconstruction filters it should use embedded in the lowest few bits of data. The PCM stream is then typically losslessly compressed using an open-source FLAC encoder for delivery. It can then be streamed by TIDAL or downloaded as a file through typical Internet mechanisms. 

On playback, the audio data is software decoded either on your computer or in your DAC with MQA-compatible firmware back into a reconstituted PCM hi-res stream at 24/88 or 24/96. From there, dither and/or noise shaping can be applied, then the embedded upsampling filter choice is used for final conversion to analogue.

Doesn't sound that grand any more when we describe what it actually does while taking out phrases like "first-generation paradigms", a "new path", "first principles", etc. Calling a spade a spade isn't nearly as romantic nor perpetuates the grandeur.

It's just odd to think that some in the media are so apparently taken with what amounts to faith in a DSP algorithm. And when others come along and try to demonstrate why it may be deserving of criticism, a grand conflict threatening the very foundations of scientific thinking gets invoked! We might as well drag Heaven and Hell, or virtue and sin into this earth-shaking dialogue. Is it any wonder that audiophiles sense this gross dissociation? Is it also not fair to ask why is it that folks who could benefit from industry incentives (not just financial incentives) seem to be so supportive of this "technology"? To not question these so-called "experts" who provide mere opinion would be obviously foolish!

To end off, I think it's important to remember what's happening here with MQA. In an unregulated free enterprise system, the arguments, tests, debates are necessary. The consumer is trying to figure out whether what is being sold to us has merit. In an age of free speech with online forums and blogs, the consumer has a powerful platform to express itself; much different from the landscape of years ago when magazines can print whatever they wanted with consumer discontentment expressed in the short "Letters to the Editor" section. It really doesn't help when the press - especially a publication like this one - appears so grossly one-sided and out of touch. As I have said before, I believe that the press should really be independent and aligned with consumer interests in mind. If in this day and age the audiophile press is nothing more than the advertising arm of an industry, then let's be transparent about that as well.

We expect that the MQA/Meridian company will spend money on partnerships, buying advertising space, and whatever forms of influence it feels it needs (politics). If there is truly merit, MQA could reap rewards... If not, they've wasted a lot of money and time on a potentially risky direction which could have been engaged productively in many other ways to build the company and its products. Through all of this, the pressure remains on them to deliver.

I suspect we'll know how this goes by next year. Is MQA truly "clearly in its ascendancy", TAS?

-------------------------------------------------

For those who want to hear it directly from Bob Stuart - here's the man himself in a recent interview:



BTW: An interesting figure he mentioned is that MQA audio is "typically 15.85 bits" and "up to 17 bits" of resolution (31:05). He still claims that undecoded MQA is better sounding than CD because "de-blur" has been done to a certain extent already (30:18).

Here's the "MQA talk" given by Danny Kaey of Positive Feedback at last week's Rocky Mountain Audio Fest 2017 streamed by Chris Connaker (Computer Audiophile):



Overall not exactly a satisfying discussion about MQA at all. MQA is barely mentioned after the first few minutes. He's instead meandering into a rather low-level and unfocused rambling about all kinds of stuff ("let's pivot the conversation") of varying significance to hi-fi. Ultimately, the guy is saying "Technology moves on... Let's get behind MQA cuz it's the only train in town!" without any good reasons provided whatsoever. As usual, he only sees good in the sound of MQA (19:35); no apparent ability to show balance nor acknowledge criticisms.

He also claims that a guy from the 1950's showing up in a modern audio show would not be surprised by what he experiences (31:00), implying little change in the hobby over >50 years. What's this dude talking about?! What about ubiquitous stereo material and even multi-channel, all that digital tech from CD onward, computer audio hardware & software, room correction DSP, speaker designs of overall better quality than the 50's, thousands of albums on a NAS, streaming audio, mobile players, generations of headphone refinement, wireless connectivity? Honestly, a guy from the 1950's would be amazed by the upgrade in fidelity, convenience, and often reasonable price tags (as well as audacious price tags) he would find. And that's just on the equipment side - imagine all the different genres of music, unique sounds, studio effects he would have to catch up on! Increase in audio fidelity might not be as impressive as the jump from B&W TV to a 4K/HDR massive flat-panel, but let's not be so devaluing of audio technology's evolution over the decades!

It's hilarious seeing that back-and-forth "yes / no" exchange on whether music producers/musicians/conductors could insist on how music is heard like Christopher Nolan and Quentin Tarantino insisting on the cinematic experience (32:30). It's one thing for a movie director to say that theaters have to meet certain criteria of video resolution and audio capability (there are all kinds of standards in place from THX to IMAX for decades), but who can insist, much less enforce, how one must listen to music a certain way at home or what equipment you use!? What a joke... The audience is right - NO. I'd be happy if a favored artist even had the power to request that their album not be overly dynamically compressed when being mastered.

There's more I can point out, but at the end of the day, clearly the audience members "get it". Have a look near the end of the talk at 1:04:30. The audience member is correct to imply that failure to expand hi-fi isn't just because of marketing, or just because the hi-fi industry "doesn't know how to say it well" in their advertising. You must have something of substance to sell when pushing the technology. Yes, Blu-Ray sells, and 4K sells. This is simply because these technologies have resulted in perceptible change compared to what came before in many situations. You cannot extract money from the consumer when they cannot differentiate a substantial improvement between CD and hi-res under most circumstances especially with yet another remaster of decades old material or new albums of poorly recorded/mixed/mastered quality; that's really all there is to it. I've said this since the beginning when discussing my "expectations" of Hi-Res Audio (and related discussion on value of Hi-Res Audio) which of course applies to MQA.

Considering that in recent years, we have already witnessed the demise of Neil Young's Pono, how many more Hi-Res Audio-based schemes must meet their wasteful end before the hi-fi industry moves on to something more worthwhile? If we were to write a book about 2-channel hi-res audio "formats", perhaps the first few chapters could be devoted to hardware playback of SACD and DVD-Audio. Somewhere in the middle would be the rise of computer audio, file-based media, and services like HDtracks and Qobuz. The last chapters of this 2-channel Hi-Res story might end up starting with the passionate madness of Neil Young, and perhaps mercifully ending in the calculated genius of Bob Stuart.

------------------------------------

Well folks, I think I'm "good" with talking about MQA for a bit. That was certainly longer than I had initially intended! Unless there are truly new developments with this codec, I'll just linger. :-)

For the pop/rock lovers, I've been enjoying The Lemon Twigs' latest EP Brothers of Destruction. Also, Cécile McLorin Salvant's latest Dreams and Daggers [Live At The Village Vanguard] 2 CD set (DR11) sounds great if you like live jazz vocals. The new Avishai Cohen album 1970 was also a fun listen last night with some friends; essentially pop with a Middle Eastern world music flair (DR9 - a bit volume compressed, don't bother with the 24-bit version).

A few nights ago, I decided to spend an evening just listening to multi-channel music and was reminded of how amazing the 2015 multi-channel Blu-Ray version of Roger Waters' Amused To Death is (a very natural evolution for an album that helped demonstrate the abilities of Q-Sound back in the early 90's). Excellent sound engineering, deep ideas, and the spatial canvas of surround sound... Wish more albums were like this!

I'll be out of the country again for a bit. Will see if I have some material to post over the next few weeks.

Until next time, enjoy the music everyone...