MQA: A Review of controversies, concerns, and cautions

[Julf]

Can't see FLAC there.

The plain "Stereo" columns are FLAC, but I am surprised about how much larger the 24/96 and 24/192 files are compared to the 16/44.1 ones. In terms of raw data you would expect 24/96 to be 3.3 times as large, but in the 2l ones the 24/96 files are larger than that.

 

[PierreV]

Can't see FLAC in the table.

De-blur is put forward. Has it been defined, yet heard in listening tests?

No idea. For the record, if I was asked if I was "for" or "against" MQA, I would answer "against" based on my perceived lack of benefit and the drawbacks/issues that others have addressed. But I am not militantly against it, as I feel it is (IMHO) a non-issue long term anyway.

 

[Eirikur]

The plain "Stereo" columns are FLAC, but I am surprised about how much larger the 24/96 and 24/192 files are compared to the 16/44.1 ones. In terms of raw data you would expect 24/96 to be 3.3 times as large, but in the 2l ones the 24/96 files are larger than that.

Bottom bits are all random noise (sometimes purposely created dither) and that doesn't compress well.

When I was young I experimented a lot with audio compression (think 1992) and found that the best compression is usually simple deltas based on word size (i.e. per 16 or 24 bits), either per channel or based on averaged channels. Next is to choose the best encoder for the entropy optimization (arithmetic is best but slow, basically one long floating point number) and you're done.

FLAC uses Coulomb/Rice coding and indeed samples delta's - and this coding scheme doesn't like noise very much... and rightly so, as the input data is not supposed to be noise anyway! Adaptive Huffman entropy coding would give similar results but may use more processing memory.

To give some idea, this is the block scheme for adaptive Coulomb/Rice (all integer arithmetic)

while (floating point) arithmetic coding would give rise to this

Some reading material: Comparison of Entropy Coding mechanism on IEEE1857.2 Lossless Audio Compression Standard

 

[KozmoNaut]

No idea. For the record, if I was asked if I was "for" or "against" MQA, I would answer "against" based on my perceived lack of benefit and the drawbacks/issues that others have addressed. But I am not militantly against it, as I feel it is (IMHO) a non-issue long term anyway.

I'll come out as staunchly against MQA. It's a solution looking for a problem, sound/bandwidth-wise it "solves" something that really isn't an issue. MQA comes out to roughly equivalent to 20-bit 96kHz PCM in available dynamic range and frequency range, yet the files are larger than a FLAC file at that quality. And the ultrasonic parts are even lossily endoed, making it completely pointless. AFAIK the claimed de-blurring produces no audible benefit and if it did, you could just as well encode that into normal FLAC files, rather than a proprietary format.

Business-wise, MQA seems like a ploy for MQA Ltd. to latch onto the production pipeline and monopolize part of the process. No one else is allowed to have an MQA encoder or even know the specifics of which processing is applied. It's a black box, completely under the control of one company that closely guards its secrets.

As a strong supporter of open standards and formats, and someone who is vehemently against DRM, closed formats and monopolies on every level, MQA goes against everything I believe in. It is utterly abhorrent to me.

I will never support MQA, be it as part of streaming services, as downloads or as physical media.

 

[SIY]

...Stuart's greed and desire to milk the music listening world...

I was going to take umbrage at that before I realized it wasn't me.

 

[BDWoody]

In one objective sentence, please.

What is the problem that MQA is solving?

The one they just told me I had. Whatever that is.

 

[Sal1950]

In one objective sentence, please.

What is the problem that MQA is solving?

#1. I solves the record companies paranoia over having full lossless copies of their prized vault masters in the wild, they've been pissed since they lost the DRM issue with CD's.

It solves the paranoid audiophlile worries about the SQ of his music, the little blue Authenticated light comes on. (what bullshit).

It may be solving any lack of income loss due to maturing video patents.

 

[mkawa]

i have to agree with amir here. MQA may or may not sound slightly better than 16/44 flac or redbook, but at worst it's indistinguishable, and i get it for no additional cost over a tidal subscription. if it dies, i don't care: i have amazon HD for the next year to see how they fill out their HD and ultra HD catalog. if it doesn't die, my DAC of choice happens to be MQA enabled (M500) and was cheaper than the near-equivalent without MQA (D5). sure, SMSL may have sacrificed 2 OPA1612s in the design for MQA, but the difference between the two stage output and the single stage output is almost certainly inaudible in my system.

in short, as a consumer who cares, but also has stuff to do outside of worrying about the audio industry, MQA is something that barely affects my life. i can see how recording techs will have to make some choices, but as was said near the beginning of this thread, the big 3 labels have already licensed and pledged to MQA all their masters until they're either done or MQA goes out of business. so this whole thing seems like much ado about nothing.

 

[KozmoNaut]

i have to agree with amir here. MQA may or may not sound slightly better than 16/44 flac or redbook, but at worst it's indistinguishable

Since MQA uses the lower bits to encode their "audio origami", they're effectively raising the noise floor for anyone listening without an MQA-capable DAC.

They can deliberately make this degradation as severe as they would like, further degrading sound quality for anyone not using a MQA-approved signal chain.

 

[mkawa]

99.9% of tidal's hifi users don't have MQA enabled hardware, and maybe 1% of its catalog is MQA encoded. if MQA were to decide to be an evil empire and degrade everyone's experience, tidal could just drop them, and they would immediately go out of business. (amir mentioned this, but it's worth saying again). further, if tidal did not drop them, amazon ultra HD is probably on the cusp of either buying them or crushing them. either option is going to cut off their revenue source and then they go kaput.

even worse, hardware MQA is really difficult to enable in the tidal app. tidal is relying on people liking the software decoded MQA over amazon ultra HD, frankly, so MQA would be shooting themselves in the foot and/or head by doing the above.

the consumers will decide either way, and maybe 98% of those hifi customers are people measuring the DACs in their phone, where they do 70-80% of their listening.

 

[KozmoNaut]

Obviously they wouldn't do it at this point, but they obviously want to embed themselves in the entire ecosystem, and become mandatory. They've already gotten in with the biggest labels, where there is a possibility of locking up the master vaults with an "MQA or nothing" clause.

These labels are sitting on so much of our shared cultural history and guarding it fiercely with crazy copyright extension schemes. MQA is yet another tool for them to lock down access to classic recordings.

 

[mansr]

Again - what is the real bandwidth saving?

In reality, somewhat worse than zero. With a high-res source, MQA delivers a quality comparable to 18-bit 96 kHz using roughly the same bandwidth as plain FLAC needs for the same resolution. Looking at what's actually on Tidal, about 40% of the MQA tracks are made from 44.1/48 kHz masters. Here the MQA version is substantially larger than a straight 24-bit FLAC.

Overall, MQA increases the required bandwidth. Claims of savings are entirely without substance.

 

[Eirikur]

You are confusing me with someone who is purely technical.

Well pardon me

I represent both technical and business aspects of these topics. And the business aspect says that when someone buys an expensive DAC, he doesn't want it to be feature deficient such as lacking MQA. Such a customer doesn't care what Archimago has to say about it either. Or the chatter about its technical aspects. He gets Tidal, listens subjective in uncontrolled environment to MQA. If he likes it, he wants it in any DAC he buys.

This I can understand up to a point: you don't want to discourage or disappoint potential customers based on a relatively minor item - unless that item actually interferes with the proper operation of your expensive product, as you have already proven!
At the same time I would expect this same manufacturer to give the buyer proper usage guidance, and inform them to choose lossless hi-res over MQA at all times. There can be a use-case for MQA if nothing else is available, but that should be the exception because of its inherent flaws. This is no different from using a crackly record as source if no better material exists.

The world doesn't just turn on some technical argument....

That's for sure - I still have a working Philips V2000 VCR in the attic, in its day far superior to VHS but still a flop!

PS: it is great we (ASR residents) can strongly disagree on matters and still have an adult exchange!

 

[beefkabob]

DRM is never a win for the consumer.

 

[Rusty Shackleford]

What data? You presented none. All you said was raise an objection and I responded with it being out of order: Bob Stuart is one of our top luminaries and has huge standing in our industry both on research and engineering. His previous work, MLP, became mandatory compression technology in DVD-A and later became Dolby TrueHD. These are facts. It means the he knows more than average blogger how to develop formats that are performant and get industry adoption. You think it is an accident that he continues to get design wins?


The article linked to was not "deconstruction of MQA." It was a response to something RH had written which I found fallacious. I explained why. I gave data on myself having content he says people don't have. I gave examples of others. Yet you come in defending him saying he has evidence and I don't?

You want to be partisan, fine. Just don't do it in such an obvious manner. Between the two of us, I am the only one who has led the development of technologies that have been adopted by the industry such as in Blu-ray. Lecture me with one-liner academic arguments at your own peril.

Your first response is another appeal to authority fallacy. Once again, in terms of MQA, Stuart’s previous work is irrelevant. We shouldn’t trust MQA because he’s done other stuff previously.

Secondly, my post mentioned Archimago’s “deconstructions” (plural) of MQA. These have been published and discussed at his blog, Computer Audiophile, and RMAF. They were also discussed earlier in this thread. I presumed you knew them.

Thirdly, your reference to your resume is, once again, irrelevant. Moreover, you’ve been involved in several high-profile failures (WMA, Zune, DVD-HD). Does that mean we should trust you less?

Finally, your entire argument against this particular Archimago post is a bait and switch. You’re focusing on some pedantic debate about whether he’s in the “audiophile community” (a weird argument for someone who runs a site that says $99 DACs are better than $5000 ones). Whether Archimago is in that community (by your definition or others’) is irrelevant to his point. His point was that, after backing down from claiming MQA sounded better, its supporters have resorted to the file size argument. Archimago’s (demonstrably correct) point is that with storage becoming ever cheaper and bandwidth becoming ever faster, there’s no longer a need to worry about compressing hi-res material.

The only partisan in this thread when it comes to MQA is you, Amir. It would’ve been smart for you to keep this thread locked. You were pilloried the first go round for your weird anti-science defenses of MQA, and it’s happenign again.

 

[amirm]

Your first response is another appeal to authority fallacy. Once again, in terms of MQA, Stuart’s previous work is irrelevant. We shouldn’t trust MQA because he’s done other stuff previously.

You don't seem to have any use for facts. In format making, prior experience in the field is everything. To the extent people want to make market/business statements as Archimago made, and folks in this thread keep making, that is super relevant. You can't dismiss it and equal it to lay arguments you are making. This is the real world, not a made up monopoly game.

Bob has shown his skills in getting all these design wins from labels to music distributor to many devices. That is a fact and is based on the factors I mentioned.

Some of you hate to hear that Bob is exceptionally qualified and respected signal processing expert. You rather him be a no-name guy so that they can throw rocks at him easily. Watch Paul's video I post earlier in how people in the industry don't dare doing that:

"Bob stuart ... some of extremely bright people. One of the most knowledgeable bright people I know in our industry. Super nice guy. I really like Bob a lot. He has done a lot of terrific things. We as a company support MQA because our customers asked for it."

His prior work in standardization of formats he developed opens doors for him that would not be open to others. His deep knowledge of signal processing and psychoacoustics overcomes internal objections from people who know less than him.

So don't waste your breath on me with these debating tactics. You are not being told to trust MQA. You are being told to learn who you are dealing with and the powers in play in the industry. He got Paul to support MQA even though he says it is no good.

 

[amirm]

Secondly, my post mentioned Archimago’s “deconstructions” (plural) of MQA. These have been published and discussed at his blog, Computer Audiophile, and RMAF. They were also discussed earlier in this thread. I presumed you knew them.

I knew them, but not what we were discussing. I expect you to follow the discussion when you comment on my posts.

 

[amirm]

Thirdly, your reference to your resume is, once again, irrelevant. Moreover, you’ve been involved in several high-profile failures (WMA, Zune, DVD-HD). Does that mean we should trust you less?

You really are not a fan of getting your facts straight, are you?

Zune was a hardware device. My group had nothing to do with it whatsoever. Zune was developed in the Xbox group by a team unqualified to do so. I fought it tooth and nail but eventually was told I better stop or else. So I did and left the company about a year later.

WMA has been in massive success. Billions and billions of devices have shipped with it from every manufacturer in the world. A day does not go by that I don't see it listed in the specs of products I review.

As for HD-DVD, that was not our product either. I resigned from the company about a year before the conclusion of format war. What came out of that effort was what I explained: that we got HD-DVD to adopt more advanced codecs and that led to Blu-ray to adopt them. Otherwise, you would have had an MPEG-2 only format in Blu-ray. Why? Because companies that created that format made the most royalties from MPEG-2.

Next, maybe you tell me your qualifications in this field.

 

[Rusty Shackleford]

You don't seem to have any use for facts. In format making, prior experience in the field is everything. To the extent people want to make market/business statements as Archimago made, and folks in this thread keep making, that is super relevant. You can't dismiss it and equal it to lay arguments you are making. This is the real world, not a made up monopoly game.

Bob has shown his skills in getting all these design wins from labels to music distributor to many devices. That is a fact and is based on the factors I mentioned.

Some of you hate to hear that Bob is exceptionally qualified and respected signal processing expert. You rather him be a no-name guy so that they can throw rocks at him easily. Watch Paul's video I post earlier in how people in the industry don't dare doing that:

"Bob stuart ... some of extremely bright people. One of the most knowledgeable bright people I know in our industry. Super nice guy. I really like Bob a lot. He has done a lot of terrific things. We as a company support MQA because our customers asked for it."

His prior work in standardization of formats he developed opens doors for him that would not be open to others. His deep knowledge of signal processing and psychoacoustics overcomes internal objections from people who know less than him.

So don't waste your breath on me with these debating tactics. You are not being told to trust MQA. You are being told to learn who you are dealing with and the powers in play in the industry. He got Paul to support MQA even though he says it is no good.

Your argument is, apparently, that MQA is okay because Stuart and other proponents falsely touted its superiority, thereby misleading customers into , and now further expansion of MQA is justified because “consumers demand it.” QED, Amir.

 

[amirm]

Finally, your entire argument against this particular Archimago post is a bait and switch. You’re focusing on some pedantic debate about whether he’s in the “audiophile community” (a weird argument for someone who runs a site that says $99 DACs are better than $5000 ones). Whether Archimago is in that community (by your definition or others) is irrelevant to his point. His point was that, after backing down from claiming MQA sounded better, it’s supoorters have resorted to the file size argument. Archimago’s (demonstrably correct) point is that with storage becoming ever cheaper and bandwidth becoming ever faster, there’s no longer a need to worry about compressing hi-res material.

You seem to be very bothered by your authority being criticized. I see that all the time in this discussion. Why? I saw mitcho saying he doesn't like to see that happen here. Why? You are feeling good criticizing your host but criticizing him is out of line?

You need cool off and not be so bothered by someone pointing out what he says as being incorrect. I showed specifically what he said and what was wrong with it. You have provided no counter to that. Just protest after protests. Attack after attack. Tell us something useful. I am not there to keep cleaning up after your claims.

You are also severely misunderstood about my position regarding "high-end" audiophiles. I am not saying anyone should be like them. But if they are going to comment about who they are, they better know who they are. Archimago doesn't know who they are based on statements he made. He can talk about MQA technical points but his business/market/industry comments should be left out to people who know that better than he does.