MQA Deep Dive - I published music on tidal to test MQA

[raistlin65]

from the 'papers' i have read on AES, they're borderline a predatory journal.

the papers' transcripts follow no real academic structure, their language is intentionally difficult to understand, and their abstracts have no real meaningful information and is there to look 'sciency'

At least the name of the journal represents that, they're a bunch of engineers who wanted to do 'science'.

He asked if anyone did any research and there are two papers on AES about it, I'm a biologist so i can't really critique the audio science in the paper.

And a lot of the AES papers that are referenced in ASR discussions are convention/convention papers. In academia, they don't carry the same weight as scholarship as full journal articles. Convention paper/conference proceedings are generally less full write-ups of research. And they tend to not go throught the more rigorous peer review and editing that a full journal article will.

So in other disciplines, a researcher might share what they're doing with a convention paper or conference proceeding publication. But serious important research would eventually get written up in a full journal article. And a lot of what goes on at convention and conferences can't make that cut.

As to the predatory nature, note that AES convention paper guidelines specifically talk about company logos and company information in their submission guidelines

https://www.aes.org/journal/con_infoauth.cfm

You just don't see that in academic discipline submission guidelines. So it does seem like this is a pseudo academic organization. It would be interesting to see what percentage of its members work in the academy or research institute versus how many are corporate employees.

 

[Mountain Goat]

Tidal app, Roon etc can all do the core decode (the first one) to 24/96 or 24/88. The render to 192 etc needs to be done on hardware (the dac basically) or, you can set the software to leave the file alone and pass to a "full" mqa dac which will do the core then render.

From what I have gleaned here, that rendering stage is little more than upsampling.

I think I have that the right way round but it's a bit of a minefield

EDIT- its worth pointing out what when I last used Tidal MQA (a year or so ago) - the majority of MQA files topped out at 24/96 so would be "fully decoded" within the core Tidal app/ Roon and wouldnt need an MQA stickered DAC at all. This is a factor that seems to be overlooked when talking about a hardware "tax" . One is pointed towards such devices when in reality, the amount of content that can be rendered to 24/192 isnt (wasnt?) huge.

Thanks for that. Because I only used the app, not Roon, I never knew the details on the resolution of the "Master" tracks.

As for these awful MQA threads, they should instantly be locked with a boilerplate "What We Know About MQA" post. Just give the facts, lossy on high-res where you can't hear it, not DRM, etc. And be done with it.

 

[Raindog123]

from the 'papers' i have read on AES, they're borderline a predatory journal.

the papers' transcripts follow no real academic structure, their language is intentionally difficult to understand, and their abstracts have no real meaningful information and is there to look 'sciency'

At least the name of the journal represents that, they're a bunch of engineers who wanted to do 'science'.

In defense of the Journal of AES, if it needs defending…. (Though I have to admit only sampled a few of their publications, so am speaking ‘in general terms’ here.)

1) Any decent scientific journal performs its reviews “double blindly”. Meaning, reviewers do not know who the authors are (or where from), and the authors (and anyone) do not know who the reviewers are. This is to decouple credentials (personalities, scientific politics, etc) from the content, and only review he content on merit/quality. The journal editorial staff is responsible for managing the review process. [Personally, I both have published in and reviewed top physics journals - Physical Review B, Physical Review Letters, Applied Physics Letters, Physica B/C - and they all (!) followed this double-blind protocol.]

2) In my [past] field - of applied superconductivity physics - the IEEE Transactions on Applied Superconductivity was probably the most ‘go-to’ journal. It combined quality, broad reach, and expedience of its publications - highly desired for dynamic exchange of ideas in exploding/growing fields between scientific groups around the world... And this was an IEEE (electrical and electronics engineers) publication and society. Still, deeply respected by outmost academicians!

 

[raistlin65]

In defense of the Journal of AES, if it need defending…. (Though I have to admit only sampled a few of their publications, so am speaking ‘in general terms’ here.)

1) Any decent scientific journal performs its reviews “double blindly”. Meaning, reviewers do not know who the authors are (or where from), and the authors (and anyone) do not know who the reviewers are. This is to decouple credentials (personalities, scientific politics, etc) from the content, and only review he content on merit/quality. The journal editorial staff is responsible for managing the review process. [Personally, I both have published in and reviewed top physics journals - Physics Review Letters, Applied Physics Letters, Physica B/C, and they all (!) followed this double-blind protocol.

In practice, double blind review doesn't always work. If you have good reviewers who know the discipline, then sometimes they are typically aware of who is doing research on particular topics. And this is even true for corporate research, such as AES.

For example, look at "A Statistical Model that Predicts Listeners' Preference Ratings of Around-Ear and On-Ear Headphones" where the authors say

"In a previous paper [4] we reported the results from controlled listening tests where listeners rated the sound quality of 30 different models of IE headphones according to preference."

So anyone qualified to be a reviewer immediately knows who the authors are of this conference paper.

 

[ebslo]

As for these awful MQA threads, they should instantly be locked with a boilerplate "What We Know About MQA" post. Just give the facts, lossy on high-res where you can't hear it, not DRM, etc. And be done with it.

Ok, but those are poor examples. The first makes an unnecessary qualification; MQA is not lossless under any conditions for the standard definition of "lossless". The second hinges solely on the definition of DRM which is not a technical term; the argument around it's use was based on semantics and even emotion, not facts, and discussion of it ended because progress could not be made, not because consensus was reached.

 

[Hai-Fri. Audio]

Mathematically lossless codecs are quite boring actually. The best and the worse are pretty close to each other. They don't have any kind of psychoacoustic model in them which is another reason they are boring. They also become less efficient the nosier the content even though we don't care much about the noise.

Prof, please correct me if I'm wrong here, I'm trying to understand this in simplest terms for my own sake:
- mathematical lossless-ness appoach attempts to preserve all information that was recorded regardless of whether it has any affect on the listener, even the noise
- perceptual lossless-ness attempts to preserve only what is needed to make the information indistinguishable (for humans) from mathematical lossless-ness (minus noise?)
- this is done by applying psychoacoustic model to the information and for the purpose of reducing the amount of data that needs to be stored/transmitted

If my understanding is correct, while interesting, MQA doesn't seem to solve any problem in our 4k streaming/cloud storage age. (Yes, I'm slow.. thank for being patient with us smooth brained folks ) Would it not be more effective to address noise issues at the recording and mastering level? Rather than applying a universal model to all different noise situations? Or do we have (/will soon have) adequate models for distinguishing between what is noise and what is music?

Here's a really "out-there", follow-up thought: Is the endeavor of psychoacoustic modeling even relevant when we are steadily marching towards readily available human modifications? e.g. a cochlear implant that extends your frequency perception - Would our mushy processors even be able to compute that? Or will it completely ruin our perception of music? From what I've seen of Amir's videos on high-res files and their ultrasonic information, my intuition tells me that as long as the ultrasonics behave in the same predictable way as music in the hearing range our brains should enjoy it. monkey brain likey patterns

I naturally wonder if Amir could use his 'tricks' (not really tricks, just very not-normal listening regimes) and training to detect differences between MQA encoded and PCM source. For example, for 16 bits vs 24, a 'trick' is to select an extremely quiet part of the music, and listen to it at a very high playback level. A level that would be deafening for most of the track. To detect 320kbps (as well as, I presume, very high VBR) mp3 encoding by a high quality encoder , a trick is to use a 'killer' musical selection, difficult to encode, and/or to zero in on a micro-instant that 'tells' and play that over and over until you learn to hear the difference.

I wonder how MQA would fare in such tests that employ very non normal listening....and would it still qualify as 'perceptually lossless' if it failed them?

yes, please!

Also, I'm really not enjoying the engineer/scientist/economist bashing subplot going on here... Just completely irrelevant and not at all in the spirit of this community. Lots of random hot takes on people's identities rather than the substance of their work. We are here to come to a scientifically grounded consensus on the merits of MQA. If you are not an expert in this field and are jumping to conclusions based on who the researchers are, please refrain so the people who have the technical knowledge can take the time to apply due diligence in their scrutiny.

We need to be patient, y'all. Look, I've had my MQA pitchfork sharpened long before this food fight - same as most here... ASR is not built so we can sit on top and stroke our egos, congratulating each other for being right. This is the largest well of audio science knowledge on the anglophone web and Professor M didn't build its reputation on jumping to conclusions.

 

[Raindog123]

The big joke about "perceptually lossless" in the context of MQA…

What I find the biggest joke is that it’s self-proclaimed ‘perceptually lossless’. Without any evidence offered.

 

[Hai-Fri. Audio]

What I find the biggest joke is that it’s self-proclaimed ‘perceptually lossless’. Without any evidence offered.

Me too. And this is a sentiment I'm quite sure everyone shares here. But it proves nothing, and due diligence must be done before I bring out my MQA pitchfork.

 

[abdo123]

And a lot of the AES papers that are referenced in ASR discussions are convention/convention papers. In academia, they don't carry the same weight as scholarship as full journal articles. Convention paper/conference proceedings are generally less full write-ups of research. And they tend to not go throught the more rigorous peer review and editing that a full journal article will.

So in other disciplines, a researcher might share what they're doing with a convention paper or conference proceeding publication. But serious important research would eventually get written up in a full journal article. And a lot of what goes on at convention and conferences can't make that cut.

As to the predatory nature, note that AES convention paper guidelines specifically talk about company logos and company information in their submission guidelines

https://www.aes.org/journal/con_infoauth.cfm

You just don't see that in academic discipline submission guidelines. So it does seem like this is a pseudo academic organization. It would be interesting to see what percentage of its members work in the academy or research institute versus how many are corporate employees.

this is a very new concept to me.

in my field what researchers are allowed to publish before a conference/convention (while keeping their self-respect) is posters, usually an 'abstract' of what they're doing and some graphs.

it is important to understand that you can not (in a respectable journal) publish data that was published before, so whatever was published in the 'convention paper' can not be published in the future.

it seems shoddy all together tbh.

 

[Hai-Fri. Audio]

MQA is not lossless under any conditions for the standard definition of "lossless"/QUOTE]

I think this is a point worth discussing because I am seeing a clear disconnect between the average consumer and experts on both the understanding of, and placement of value in mathematical vs. perceptual lossless

 

[Thomas savage]

Except former moderators...they will start twitching uncontrollably...

Personally I get the urge to clear the freezer and hide inside it but we all have different coping mechanisms ha ha

 

[Eskamobob1]

this is a very new concept to me.

in my field what researchers are allowed to publish before a conference/convention (while keeping their self-respect) is posters, usually an 'abstract' of what they're doing and some graphs.

it is important to understand that you can not (in a respectable journal) publish data that was published before, so whatever was published in the 'convention paper' can not be published in the future.

it seems shoddy all together tbh.

Wait, what? Meta data studies are common in every field i have ever interacted with in academic research (several fields of physics, but most deeply CFD/CHT, and process engineering). Unless I'm missing what you mean?

 

[abdo123]

Wait, what? Meta data studies are common in every field i have ever interacted with in academic research (several fields of physics, but most deeply CFD/CHT, and process engineering). Unless I'm missing what you mean?

you can reference and analyze the data in a future paper but you can't publish the same data.

 

[Eskamobob1]

I think this is a point worth discussing because I am seeing a clear disconnect between the average consumer and experts on both the understanding of, and placement of value in mathematical vs. perceptual lossless

Not consumers and experts. Just consumers and amir.

 

[BDWoody]

Personally I get the urge to clear the freezer and hide inside it but we all have different coping mechanisms ha ha

I should try that, but then I'd have to disturb Mother...

 

[Eskamobob1]

you can reference and analyze the data in a future paper but you can't publish the same data.

Ah, do you just mean reuse data without credit? Tbh I think we just use the term "publish" slightly differently

 

[Thomas savage]

I should try that, but then I'd have to disturb Mother...

Thats ok she won't defrost and start to smell for about 4 hours ime..

 

[Thomas savage]

Thats ok she won't defrost and start to smell for about 4 hours ime..

Wait! that's ambient temperature and humidity dependant, obviously I'm in the UK so its normally 8c and raining ..

 

[Mountain Goat]

Ok, but those are poor examples. The first makes an unnecessary qualification; MQA is not lossless under any conditions for the standard definition of "lossless". The second hinges solely on the definition of DRM which is not a technical term; the argument around it's use was based on semantics and even emotion, not facts, and discussion of it ended because progress could not be made, not because consensus was reached.

It was a poor decision of mine to include those examples. Obviously someone other than me would write the FAQ.

 

[Hai-Fri. Audio]

Not consumers and experts. Just consumers and amir.

A quick search in any academic database would have saved you the embarrassment of this ignorant comment. Perceptually lossless compression is not at all an exclusive concept to audio.

This is what I mean, people... Zero background knowledge, not even a quick google search, but full of vigor when accusing Amir of having no expertise.