Is MQA Bad For Music?

[1_sufferin_mind]

In my never-ending quest to raise awareness of all that is audio, I've discovered yet another "interesting" blog, MQA Is Bad For Music, Here's Why.. This one sets MQA on an accusatory stage illuminated in a dismal light.

What is MQA? According to Wikipedia, MQA is: "an audio codec using lossy compression and a form of file fingerprinting, intended for high-fidelity digital audio file download and internet streaming." But the blog doesn't delve into the technical aspects of the format. Instead, it discusses how MQA:

• Appears to be a major label venture between Warner and Meridian
• Is a supply chain monopoly
  
• Uses fingerprinting that will check if MQA has been paid at each stage
• Stifles the artists' creativity

I might be able to stomach all this bleakness, to use the author's own term, were it not for the concluding paragraph which seems to offer contradictory and dismissive tones. It makes it seem as if the author compiled the evidence as an exercise in it's own right, rather than to enlighten and reward. Yet on second read, I really feel on-board with the idea, and realize not everything's in a title.

What's your opinion? Do you think the infrastructure behind MQA will support or hinder the format?

 

[Sonnie]

Do we need another audio codec? I can't learn enough about it really to make a justified decision, but I'm still not sure why we need another audio codec.

 

[1_sufferin_mind]

Ahhh, I'm glad you asked, Sonnie! It seems the answer you get depends on who you turn to.

According to this source, MQA is not just another audio codec, it's a philosophy of sorts: a hierarchical method and set of specifications for recording, archiving, archive recovery and distribution of high quality audio. So from that perspective it's a welcome newcomer. MQA Lossless deals with data in the digital domain. The biggest problem is getting it from analogue and back to analogue with the least audible damage. Outside of this perspective, MQA looks strange.

According to this source, MQA is a new way to deliver digital audio, to pack an incredibly data-hungry analogue or, more likely, digital source into a form we average phone-holding folk can consume: a stream or download. It’s a solution to a very ‘now’ problem: in this era of hi-res audio, how can you deliver top-quality lossless sound without zapping everyone’s data allowance or having to buffer constantly on anything but the best connection?

So why do we need another codec? I'm under the impression there are none available today which stream hi-res material in true high-fidelity format.
Maybe someone with deep knowledge can chime in?

EDIT: Added bold highlight

 

[Jack]

Ahhh, I'm glad you asked, Sonnie! It seems the answer you get depends on who you turn to.

So why do we need another codec? I'm under the impression there are none available today which stream hi-res material in true high-fidelity format.
Maybe someone with deep knowledge can chime in?

I admittedly do not have more knowledge on the subject then the experts referred to in your missive, however, I have to say I don’t really mind having a new codec with which to move high fidelity music through our systems. In my most humble opinion, although many would not consider me humble, as the capture of recorded music does not necessarily seem to improve with time, the presentation of those recordings through our systems and codecs seem to improve with time. Maybe I should actually say can improve with time if given the correct substructure to which the new Hi-Rez codecs can be used.


In most cases the codec such as MP3 or other low-level compression and tends to rid the music of some if not much of its fidelity while other codecs such as ALAC, FLAC and

WAV tend to preserve most of the fidelity of a recording. Personally now that space has become much cheaper high use WAV format for all of my music retained on my music server hard drive.


Now comes a codec, MQA, that allows a moderate reduction in size due to compression that can tend to make most recording sound a bit better while a good recording well sound a great deal better. Further, it would appear that the MQA format is easily downloaded or transmitted over the Internet in particular via Tidal, that combined with licensing from large recording companies can provide the listener with much better sound on a limited bandwidth. I have heard MQA in practice and can say without question that what I was hearing was very very good and if MQA can use to make inroads into the music scene I might just go in that direction.


Is there a downside? Absolutely there is a downside in that you now have to buy into MQA hardware that ranges from about $200 to well over $20,000 for the grand poohbah. I can now say with some conviction that I will not be buying the $20,000 model but may find a way to invest in the $200 model once I am sure MQA will stay with us. Until that time, WAV will rule my airwaves as it will play on pretty much anything and sound good on systems ranging from petite to gargantuan.

 

[1_sufferin_mind]

Now comes a codec, MQA, that allows a moderate reduction in size due to compression that can tend to make most recording sound a bit better while a good recording well sound a great deal better. Further, it would appear that the MQA format is easily downloaded or transmitted over the Internet in particular via Tidal, that combined with licensing from large recording companies can provide the listener with much better sound on a limited bandwidth. I have heard MQA in practice and can say without question that what I was hearing was very very good and if MQA can use to make inroads into the music scene I might just go in that direction.

The available literature I've perused so far mentions downloads alongside streaming. But if downloads were MQA's only forte, I imagine it would have a difficult time treading water as there are plenty of lossless codecs currently available. MQA isn't a brand-new way of encoding audio data the same way that DSD is; rather, it's a novel way of packing up familiar music files so they sound better in less space. To help clarify to the audience what MQA is and is not, and to add a bit of meat to the skeletal definition we've established so far, consider Bob Stuart's comment in this article: "The strength of MQA is that one file can be played back in a wide variety of situations by the customer."

An MQA-encoded file can be played back in four ways; with no decoding, software decoding, hardware decoding, and a combined software/hardware decode.

If you play back a 24-bit/192kHz MQA-encoded file using iTunes through a regular DAC (i.e. a non-MQA DAC), you will get a 24/48 file.

If you play back a 24-bit/192kHz MQA-encoded file through an MQA software decoder like Tidal HiFi, Audirvana, or (soon) Roon, and you are using a regular DAC (i.e. a non-MQA DAC), you will get a 24/96 file. A software decoder does not offer the ability to 'unfold' the original file to resolutions higher than 24/96 (or 24/88.2).

If you play back a 24-bit/192kHz MQA-encoded file through an MQA-enabled DAC, you will get a 24-bit/192kHz file. If you are also using a software decoder like Tidal HiFi, Audirvana, or (soon) Roon, you can have the software decoder perform the first 'unfold'.

One thing to note: if the original MQA file is 24/48, 24/96, or 24/88.2, it will pass through the software decoder and be 'unfolded' to its original resolution. This brings up the interesting fact that people who love their DAC and its proprietary digital filter may very well be able to have their cake and eat it too, especially when streaming MQA content from Tidal HiFi.

 

[Sonnie]

Well with your new description, it sounds like it might be a good offering for some. I am not much on streaming, and certainly not from my phone. I prefer to outright buy and own in my hands what I plan to listen to. I might get a sampling of it from YouTube, but it has nothing to do with sampling the quality since I am sampling with headphones and not in my listening room. I still can't get over folks paying 20 bucks a month for streaming music.

 

[AJ Soundfield]

MQA is a sham. It's a way to route all music through Bob Stuarts bank account.

The "reasoning" behind it is largely nonsense. The presumption that all music is "time smeared" unless encoded/decoded with MQA is ridiculous. Bob Stuarts own AES papers show no such thing. The comments section there (AES) is interesting, as they are called out on it.
No one can hear the difference between 16/44 and "Hi Rez" with their ears http://www.aes.org/e-lib/browse.cfm?elib=14195 (unless you crank the volume to ear damage levels during silence so the quantization noise floor of 16bits becomes audible).
Of course audiophiles believe they can hear all kinds of things when the imagination runs wild minus any controls. Hence we have MQA, which will "fix" AD filter time domain errors (which Bob Stuart cannot demonstrate as audible). The problem is by doing so, you discard basic sampling theory and introduce (usually) unwanted aliasing distortion artifacts. A very nice layman explanation by Dr Lesurf: http://www.audiomisc.co.uk/MQA/origami/ThereAndBack.html

Yes, I've heard it ;-). Via Tidal and a Mytek Brooklyn with "full" unfold.
The gent who owned the Mytek is returning it after a month ;-).
$100k system and couldn't hear any significant difference, much less improvement. He even had both his daughters who play in orchestras listen and try to "hear" MQA.
Like HDCD, DVDA and SACD, yet another audiophile tempest in a teapot.

cheers,

AJ

 

[Sonnie]

I was wondering when you'd get around to speaking your mind. Good info AJ... interesting to say the least.

 

[AJ Soundfield]

You're welcome sir

Too many "solutions" in search of problems in this hobby of ours

 

[Todd Anderson]

I've heard MQA in action... along with various Hi-Res files.

I'd have to agree with AJ: I can't really identify a difference between it and CD-quality sound. Perhaps on a $200K system? Maybe. I don't know... and at the end of the day it doesn't really matter, because that kind of system is prohibitively expensive (though, hearing them at a show is treat).

I do like to error on playing Hi-Res if I can, just as an assurance that it's of some relatively valid quality, but I can't say I'm reaching for any one file type because of known characteristics or quality...

 

[Sonnie]

I have that mentality that SACD sounds better... it's got me fooled if I'm fooled.

 

[Todd Anderson]

Perhaps if the SACD features a particularly notable remaster of the material? It's possible that you'd hear a notable difference in sound quality.

 

[Sonnie]

Yeah... that would make sense, especially if they processed the sound to improve it over the original.

 

[Jack]

MQA isn't a brand-new way of encoding audio data the same way that DSD is; rather, it's a novel way of packing up familiar music files so they sound better in less space.

I believe that DSD is the original way of encoding digital as offered by Sony and others, it is not new by any means.

 

[Jack]

I have a good number of SACD's and some sound very good, I believe better than the plain CD's but others are not worth any extra dough, Tapestry and Toys in the Attic to name two.

 

[Sonnie]

I think some of them are enhanced in some way to make them sound a bit different... and perhaps better in some instances.

 

[Jack]

I dont doubt for a second that some "Magic" may be involved in some cases.

 

[Todd Anderson]

My understanding is that remastering can make a presentation go one of several ways: better, the same, or worse... I don't have depths of expertise in this, but I'm fairly sure that #1 (better) isn't always the outcome.

 

[AudiocRaver]

Todd nailed it, in my opinion. Remastering can make a track sound better or worse or virtually the same. One or more human beings are involved in making sonic decisions, and there is no guarantee in the outcome. Corporate agendas are often involved, usually not for the better.

MQA is a means to get standards working for the benefit of the listener. That is going to cost money and someone will make a profit. OK. Will it sound better? It depends. A little TLC on the part of a few engineers and producers can accomplish the same thing. Standardizing every piece of the chain, hardware and software, is a lot of expensive trouble to be wasted when a listener plays it back in a poorly set up room. Or will the standardized chain extend to room treatment and speaker setup? And who gets to say that a room is worthy of MQA?

Then there is the fact that a well-recorded CD at 16-44.1 can sound phenomenal. And a 24-96 track can sound lousy if one step in the chain is poorly executed.

Standards are a necessary part of progress, they are the static side of static and dynamic quality. as described by PIrsig in Lyla (he also wrote Zen and the Art of Motorcycle Maintenance). Every step of change in technology needs a firm foundation to reach out from. Standards give that static foundation. MQA is a set of standards, like HDMI.

Is MQA needed? Useful? Silly? Only time will tell. Will it affect every piece of the audio chain from recording mic to speaker? Probably not. Will people make money from it? Of course. Will the music sound better? Sometimes. MQA goes beyond the question of high resolution vs 16-44.1.

 

[1_sufferin_mind]

MQA isn't a brand-new way of encoding audio data the same way that DSD is...

I believe that DSD is the original way of encoding digital as offered by Sony and others, it is not new by any means.

That's what I get for abbreviating my thoughts. I meant that MQA is NOT a paradigm shift the same way that DSD was.

I think some of them are enhanced in some way to make them sound a bit different... and perhaps better in some instances.

Or even the opposite - that the CD layer is "dumbed down" to portray the SACD performance in a better light.

 

[AJ Soundfield]

I have that mentality that SACD sounds better... it's got me fooled if I'm fooled.

You're not imagining things, this is exactly what the AES study found! The care taken with original DSD and/or transfers often resulted in "audiophile" sonics, since that was the target audience.
This is exactly why they did the ADA "loop", to eliminate this as a variable. They took those stellar masterings and ran them through a lowly 16/44 ADC-DAC "loop" and then compare the output to the "direct" feed from the SACD.
Now you are comparing apples to apples, not mastering skills. The results were funny ;-).
MQA is a bit different in that if you read the Dr Lesurf kink, he clearly explains how the MQA "process" is entirely capable of "changing" the sound (anharmonic HF aliasing distortion and reEQ). The problem is that even this change can be very difficult to detect...unless you've swallowed the whole jug of Koolaid.
This would be all swell if it were just another round of harmless audiophile nonsense, but as expected, there is now talk of MQA "encoding" of new music. What does that mean for options when you don't want it?
Since it is proprietary, the bitstream cannot be passed through devices that result in very very real audio benefits like MinDSPs, etc. One would have to take the MQA analog out and reconvert via the MiniDSPs ADC.
There should be a much greater backlash against this scam, but alas...

 

[Sonnie]

So this ain't similar to THX... it is a codec.

 

[AJ Soundfield]

It's a encode/decode scheme where all music, past, present and future, is to be "processed" through Bob Stuarts bank account, like laundry.
They don't include it in the patent submission block diagrams, but it's there.

 

[Leonard Caillouet]

There will be a backlash against it once people start to sort it out.

 

[Leonard Caillouet]

Schiit is not buying into it...

http://schiit.com/news/news/why-we-wont-be-supporting-mqa