MQA: A Review of controversies, concerns, and cautions

[Frank Dernie]

You guys are making me feel smug about going back to CDs

 

[Eric Auer]

Personally I cant wait for the gigantic consumer uproar over MQA

*crickets..

Eric

 

[Blumlein 88]

The bigger wrong by far is Blu-ray, not MQA. You want digital audio from that? Not only do you have to deal with its copy protection but you also have to put in new interface in the form of HDMI!

Really, in the world of audio today, MQA is not remotely big enough problem for people to go after. People do because it gives them a platform to shout about online and make a name for themselves.

Look at me. MQA is there. And I am here. Whatever it wants to do, doesn't impact me one bit. It is not like when Tidal supported MQA, it stopped releasing non-MQA versions of the title. It is just one more thing for people want it to consume it. For the other 99%, we can all go about our business.

Now if we went after video and liberated that, we would have something. That is something we all use today. Wouldn't it be nice to not have to have a different player to install and use for each format?

HD video uncompressed is more than 374 megsbytes per second. Not 4k and not HDR. Therefore video is of necessity compressed. Various formats are about better and more efficient compression encoding. If such high data rates ever become trivial we can have plain pure pcm-like video.

 

[Cosmik]

Personally I cant wait for the gigantic consumer uproar over MQA

*crickets..

Eric

The same words could be expressed about net neutrality - that some people around here, somewhat inconsistently, do seem to care about.

 

[amirm]

But if all the streaming services adopt MQA then, as things stand, the only digital stream that we DSP geeks will have access to will be the non-decoded MQA. Correct?

I am not understanding your question. Answering anyway , if currently lossy/256 kbps adopt MQA, that will be a good thing and I would then upgrade my hardware to support it. That quality boost will be worth it to me for sure.

 

[amirm]

I have no idea what your talking about but if you believe the labels would ever dump MQA for a open source and bit perfect solution your being incredibly naive.

Huh? Labels today are providing all manner of high resolution audio not just for streaming, but for downloads. With downloads piracy becomes trivial yet they are OK with it. So I say again, the music labels are way past worrying about copy protection.

And I can absolutely get label support for a new format that is free and is being adopted by content distributors and device makers alike. It will be a hell of a lot simpler than what MQA is doing by requiring royalties.

Remember, I spent a decade working with major labels and movie studios on acquiring content for new formats. This is part and parcel of what I know and did. So please don't be dismissive of what I am saying.

 

[amirm]

Therefore video is of necessity compressed. Various formats are about better and more efficient compression encoding. If such high data rates ever become trivial we can have plain pure pcm-like video.

Video is compressed but then wrapped in copy protection for distribution. By doing so, the level of access we get to video is hugely reduced. I know I love to download compressed video just like I download flac compressed high res music. I can move it around with ease, back it up, etc.

 

[Cosmik]

I am not understanding your question. Answering anyway , if currently lossy/256 kbps adopt MQA, that will be a good thing and I would then upgrade my hardware to support it. That quality boost will be worth it to me for sure.

OK, putting it differently: if a streaming service that currently offers 16/44.1 'upgrades' to MQA , then I, as a listener who uses DSP for crossover filtering will not have access to anything but the undecoded (~13-bit equivalent..?) MQA data for my DSP to work on. Is that correct?

(and I will be using double the previous FLAC'ed bandwidth for the privilege...)

 

[Cosmik]

There's more.

MQA is promoted as an end-to-end solution, starting with analogue capture of sound at the microphone, and ending with an analogue music signal feeding a customer's amplifier and speakers. In another twist that could be understood from more than one viewpoint, what happens between—from the encoder, through distribution channel, to the decoder—represents a black-box system into which we should not delve. This is for our benefit, the company insists, in order that the original quality is not compromised by non-bit-perfect processes. So like Blu-ray and similar rights-managed video technologies, current implementations of MQA also deny user access to the raw high-resolution bitstream.

"In our model," counters Stuart, "there is no single high-res digital signal outside the context of the decoder matching the output of a DAC. That is not the creation of a digital lock, more like avoiding an absurdity."

https://arstechnica.com/gadgets/201...hing-you-need-to-know-about-high-res-audio/5/

I think this is saying that there can be no full quality 'digital stream' for DSP enthusiasts to access, anyway, even if the MQA people were to cobble together some sort of software decoder. The final refinement must happen in MQA-licensed, approved, secure hardware. So 'room correction', DSP crossovers etc. must always be applied to an inferior signal - possibly even via analogue conversion.

And what does an end-to-end solution mean? It means that MQA's dubious algorithms and processes are being applied at the acquisition stage. There is no pure high res PCM master that can optionally be supplied alongside MQA. If MQA is flawed, then the flaws will be written through all of our future music archives.

 

[amirm]

OK, putting it differently: if a streaming service that currently offers 16/44.1 'upgrades' to MQA , then I, as a listener who uses DSP for crossover filtering will not have access to anything but the undecoded (~13-bit equivalent..?) MQA data for my DSP to work on. Is that correct?

I don't see any scenario under which that would happen. Take Tidal. They probably have 20 million tracks in 16/44.1. In contrast, they probably have 20 thousand tracks in MQA. No way, no now MQA can ever get even 10% of the CDs out there encoded let alone 90% or so that may be enough for someone to switch completely to them.

CD equiv. MQA files are also bigger due to the need to include the high-res bits as random data. That randomness increases entropy making flac less efficient. The net result is that the cost of streaming will go up for CD rate delivery, rather than go down (which is the case when MQA replaces high-res).

 

[Cosmik]

I don't see any scenario under which that would happen. Take Tidal. They probably have 20 million tracks in 16/44.1. In contrast, they probably have 20 thousand tracks in MQA. No way, no now MQA can ever get even 10% of the CDs out there encoded let alone 90% or so that may be enough for someone to switch completely to them.

CD equiv. MQA files are also bigger due to the need to include the high-res bits as random data. That randomness increases entropy making flac less efficient. The net result is that the cost of streaming will go up for CD rate delivery, rather than go down (which is the case when MQA replaces high-res).

I read your words, Amir, but I just can't understand them..! Isn't the idea that MQA is the future of streaming with its higher quality and blue light? Surely the streaming services are going to sell it as a feature, and when it takes off properly, it will become the ubiquitous way of delivering music - there'll be no distinction of whether the original is derived from a CD or not; those recordings will be "de-blurred" and piped out with everything else as MQA, surely?

 

[Cosmik]

Huh? Labels today are providing all manner of high resolution audio not just for streaming, but for downloads. With downloads piracy becomes trivial yet they are OK with it. So I say again, the music labels are way past worrying about copy protection.

And I can absolutely get label support for a new format that is free and is being adopted by content distributors and device makers alike. It will be a hell of a lot simpler than what MQA is doing by requiring royalties.

Remember, I spent a decade working with major labels and movie studios on acquiring content for new formats. This is part and parcel of what I know and did. So please don't be dismissive of what I am saying.

And I fail to grasp the main point of this, too. If MQA becomes ubiquitous, the record companies can finally squash piracy even if their attempts have failed until now. They can degrade the undecoded audio to any degree they like, presumably - even if they leave it at an acceptable quality for 99% of people at the moment. The degradation can gradually be ramped up as people begin to forget the 'wild west' days of the internet and free copying of pristine digital audio.

And then MQA hardware can refuse to decode content that is not authenticated with the correct, current credentials. It could even be used to block anything but pukka files, and if the chips are fitted to all player devices in future, this would give the 'suits' total control. Maybe an amateur band will in effect need to pay something to MQA for the privilege of distributing their garage recordings. Why not?

 

[amirm]

And I fail to grasp the main point of this, too. If MQA becomes ubiquitous, the record companies can finally squash piracy even if their attempts have failed until now.

No, they can't. MQA baseband is in the clear and so are every file you download from every digital service to Apple, Google, Amazon, etc. The industry has spoken and that is the situation. MQA does not provide anything to change that dynamics. Pleasing a handful of audiophiles will not amount to anything whatsoever in changing music dynamics.

 

[amirm]

And oh, let's not forget the CD which continues to be released without copy protection.

 

[amirm]

Isn't the idea that MQA is the future of streaming with its higher quality and blue light? Surely the streaming services are going to sell it as a feature, and when it takes off properly, it will become the ubiquitous way of delivering music - there'll be no distinction of whether the original is derived from a CD or not; those recordings will be "de-blurred" and piped out with everything else as MQA, surely?

If it becomes ubiquitous, it means the consumer has spoken and wants it. In that case, that is it and we better not complain.

Fortunately the reality is different. Replacing CD streaming with MQA is simply not in the cards for reasons I mentioned. MQA has put itself in the encoding path which hugely limits how much content can be made available. It would take tens of millions of dollars to encode any significant library of content to compete with CD. And for what? The real music market has no need or care for something called MQA or high-resolution.

Let's also remember that MQA is absent from download market right now. That market continues to be served with digital PCM and DSD downloads.

 

[Cosmik]

If it becomes ubiquitous, it means the consumer has spoken and wants it. In that case, that is it and we better not complain.

I just find that a non-argument...

The consumer apparently wanted - or could be persuaded to accept - diesel cars, radium toothpaste, fluoridation of water, VHS over superior Betamax, etc. etc.

 

[Palladium]

If all the support MQA has is the coffers of RIAA shitheads, I said good luck.

 

[Werner]

That said, I have said elsewhere that I think I can hire the right signal processing experts to build an open-source competitor to MQA for around $100,000.

$100k is nothing, so you do agree that there is not much in MQA.

Everyone who is up in arms on this should create a funding campaign and I will then get the people to build it.

Why that? If MQA is a solution without problem, why would making an open-source solution be any better, if there is still no problem to solve?

As to JJ, have you asked him what he would charge you to build this solution including ...

What has that to do with anything?

 

[Cosmik]

Replacing CD streaming with MQA is simply not in the cards for reasons I mentioned. MQA has put itself in the encoding path which hugely limits how much content can be made available. It would take tens of millions of dollars to encode any significant library of content to compete with CD. And for what? The real music market has no need or care for something called MQA or high-resolution.

But...

The implication being that MQA can potentially correct the temporal smearing of almost every recording in existence. Success here largely depends on whether the original A/D converter’s make/model is known. If it is known, precise settings can be applied, otherwise a best guess filter is applied.

“Does MQA encoded music played back on a non-MQA DAC sound better, worse, or the same as playing back the file in its native resolution? While it depended on the recording, un-decoded MQA file did not sound worse than the original native file and in some cases, the un-decoded MQA version sounded better.”

https://darko.audio/2016/06/an-inconvenient-truth-mqa-sounds-better/

It seems to me that having put that story out there, streaming could just go MQA, promoted as improved sound quality with or without the decoding hardware. As regards the tens of millions of dollars to process existing files, a simple default "best guess filter" would suffice (if I were them I would make its characteristics a bit random for each file so troublesome people couldn't identify it a simple default 'filter'). Set a few PCs running for a few weeks, or do the "de-blurring" of existing material 'live' on demand.

Once the MQA foot is in the door, it gradually takes over until there is nothing but MQA. I can't see a problem with that strategy for the music business.

 

[RayDunzl]

Did we define 'blur' yet?