Thread: MQA Discussion

Go Barry.

He clearly is saying that Reed/Solomon ECC must be defeated.

Originally Posted by wisnon

Go Barry.

He clearly is saying that Reed/Solomon ECC must be defeated.

That or you can hear the real master PCM just fine, without added unauthentic aliasing distortion and reEQ the MQA slurpers lust.

cheers,

AJ

http://www.gatewaymastering.com/wp-c...er-4Nov17.pptx


Sent from my iPhone using Tapatalk

https://youtu.be/9NHuwOgWYOo

A must watch!

Sent from my iPhone using Tapatalk

Originally Posted by Mike

https://youtu.be/9NHuwOgWYOo

A must watch!

Sent from my iPhone using Tapatalk

Good post Mike!!

An analysis of MQA on CA
https://www.computeraudiophile.com/c...cautions-r701/

Originally Posted by Dr Tone

I wouldn't call it a must watch. He actually said: "in his opinion, MQA has less loss than PCM & DSD". Must be magic pixies.

I thought it was a must watch and funnier than anything on Comedy Central. MQA has "less loss" than PCM or DSD because...Hans said so. Never mind the bits lost and the Patent itself spelling it out. Maybe in his next video he'll explain how it actually has less aliasing distortion...because he said so and "he and his colleagues heard it".
You can't make that stuff up LOL.
Is there any reason why folks can't just like MQA because of the remastering/aliasing distortion/EQ enhancement, rather than a whole bunch of malarkey? Do the SET guys make as many ridiculous excuses as to why they like the sound?
It really is ok to like something just because, rather than for made up nonsense. Seriously, it's ok.

https://m.youtube.com/watch?v=xl3LDwQ1pVs

https://www.stereophile.com/content/...gravity-page-1

Bad link Mike?

https://www.stereophile.com/content/...enters-gravity

Originally Posted by AJ Soundfield

Bad link Mike?

https://www.stereophile.com/content/...enters-gravity

Works for me


Sent from my iPhone using Tapatalk

Originally Posted by AJ Soundfield

https://www.stereophile.com/content/...enters-gravity

Redundant post deleted

https://darko.audio/2018/05/playback...ob-stuart-mqa/


Sent from my iPhone using Tapatalk

https://stereo-magazine.com/index.ph...d045&download=

Interesting comment from High End (Munich) 2018, bottom left page 16

Well, the McGill U study (with BStuart involvement, providing files) is out http://www.aes.org/e-lib/browse.cfm?elib=19396
Surprise, surprise. Exactly what I found

Originally Posted by AJ Soundfield

Well, the McGill U study (with BStuart involvement, providing files) is out http://www.aes.org/e-lib/browse.cfm?elib=19396
Surprise, surprise. Exactly what I found

IMHO that should also put in perspective the effectiveness of double blind tests.

Originally Posted by Music Fiend

IMHO that should also put in perspective the effectiveness of double blind tests.

Totally agree with you, since we both read the entire published paper and saw they had absolutely no problem telling the difference with certain tracks, but that whereas as one track was preferred PCM, the other was preferred MQA.
i.e., the added aliasing distortion/EQ "spicing" of MQA can be better...or worse, depending on original mastering. Yes, blind tests in state of the art controlled listening facilities as shown, with highly trained verified listing abilities testers are indeed the ultimate in effectiveness. Which is why it is the de facto gold standard of science, rather than just casual easy chair all variables uncontrolled sighted "listening". Thanks for concurring. My experience with MQA was exactly the same, hit and miss.

p.s. btw, it was also nice to confirm at a recent private audition with the Florida Orchestra, that they too, like most major orchestras, use blind auditions to eliminate bias when selecting members. How cool is that?

AJ Sound field,

I haven't read the entire paper nor do I concur with the conclusions that you've drawn.

Just think about violins double blind tests outcomes...

Nor do I think that MQA is hit and miss.

In my non expert and certainly not scientific approach I almost always prefer the MQA version to the Redbook alternative whilst using Tidal (Masters).

Cheers!

Re your PS: about that I am convinced that that is a way to mitigate possible race, gender etc discrimination.

Originally Posted by Music Fiend

AJ Sound field,

I haven't read the entire paper nor do I concur with the conclusions that you've drawn.

Ah, those might be related. Sorry, I thought you would have wanted to before casting aspersions or drawing conclusions.

Originally Posted by Music Fiend

Just think about violins double blind tests outcomes...

Exactly, even more proof they work and why they are the gold standard of science, including fields of perception like audio, testing the sound>ears of violins, orchestra members playing violins, etc, etc.

Originally Posted by Music Fiend

Nor do I think that MQA is hit and miss.
In my non expert and certainly not scientific approach I almost always prefer the MQA version to the Redbook alternative whilst using Tidal Masters.

Cheers!

Yes, of course, that all makes sense from an approach vs results standpoint. Enjoy Tidal (while you still can).

cheers,

AJ

Originally Posted by AJ Soundfield

Exactly, even more proof they work and why they are the gold standard of science, including fields of perception like audio, testing the sound>ears of violins, orchestra members playing violins, etc, etc.

Regarding double blind tests and measurements, what you call science gold standard, I am far from convinced that they are truly effective and that you can get all the valid conclusions simply by considering the published results.

The McNamara Fallacy comes to mind.

Specifically about the violin results and double blind tests I also like to take into account what some true experts have to say:


https://www.thestrad.com/blind-testi...t/6944.article

Cheers!

Originally Posted by Music Fiend

Regarding double blind tests and measurements, what you call science gold standard

That's not my descriptor, it's a fact. There is a difference.

Originally Posted by Music Fiend

I am far from convinced that they are truly effective and that you can get all the valid conclusions simply by considering the published results.

Then I sure hope you don't take pharmaceutical drugs, use a intelligible cel phone, attend major orchestras, rely on particle physics results, etc, etc, etc.
What is you viable, proven better alternative to the defacto standard used by science?

Originally Posted by Music Fiend

The McNamara Fallacy comes to mind.

Weak. That has nothing to do with violin/orchestra members tests whatsoever. No statistical data is mined, sorry. My guess is you are totally unfamiliar with musician auditions.

Originally Posted by Music Fiend

Specifically about the violin results and double blind tests I also like to take into account what some true experts have to say:
https://www.thestrad.com/blind-testi...t/6944.article

A true expert in excuses.
He presented zero scientific rebuttal in his pure opinion excuses piece, sorry.

cheers,

AJ

Btw, getting back to MQA, Bob Stuart, creator of MQA, agreed to this test and supplied the files!! Fact.

Originally Posted by AJ Soundfield

That's not my descriptor, it's a fact. There is a difference.


Then I sure hope you don't take pharmaceutical drugs, use a intelligible cel phone, attend major orchestras, rely on particle physics results, etc, etc, etc.
What is you viable, proven better alternative to the defacto standard used by science?


Weak. That has nothing to do with violin/orchestra members tests whatsoever. No statistical data is mined, sorry. My guess is you are totally unfamiliar with musician auditions.


A true expert in excuses.
He presented zero scientific rebuttal in his pure opinion excuses piece, sorry.

cheers,

AJ

In terms of music reproduction enjoyment / preferences I tend to use my own hearing ability. Alas, I can't use measurement devices or double blind tests outcomes to listen to music.
True experts in HiFi like Nelson Pass understand the limitations of measurements.

And I'm under the impression that major orchestras existence preceded measurements, published papers and double blind tests...

Is it weak? The McNamara Fallacy is not just about data mining. From Wiki:
The McNamara fallacy (also known as quantitative fallacy[1]), named for Robert McNamara, the United States Secretary of Defense from 1961 to 1968, involves making a decision based solely on quantitative observations (or metrics) and ignoring all others. The reason given is often that these other observations cannot be proven.

I'd say being a professional violinist not only makes him an expert it probably allows him not to fall into the McNamara Fallacy or suffer from the Dunning-Kruger effect.

Cheers!

Originally Posted by AJ Soundfield

Btw, getting back to MQA, Bob Stuart, creator of MQA, agreed to this test and supplied the files!! Fact.

Does that mean that he concurs with the conclusions that you've reached?

Originally Posted by Music Fiend

In terms of music reproduction enjoyment / preferences I tend to use my own hearing ability. Alas, I can't use measurement devices or double blind tests outcomes to listen to music.

Correct, for mere enjoyment of math, running and especially music etc, one needs only listen and prefer, there is no need to measure (unless one wanted a way better sounding system free of terrible bass peaks, out of phase, reversed channels, etc etc), or blind test test ability there. One can just prefer whatever one prefers and enjoy.
It is only if wanted to actually test those abilities, that measurements and tests would come into play. That's why one doesn't grade ones own math test, why Olympic runs are exactly timed and why all the testers in Bobs MQA test referenced have assessed, verified abilities, just like with math or running. Self assessed, believed abilities are verboten.

Originally Posted by Music Fiend

True experts in HiFi like Nelson Pass understand the limitations of measurements.

Exactly, that's why Mr Pass has never ever designed an amplifier without measurements. That's why Mr Pass designs based on the foundation of his science education. Since it's audio devices, he knows measurements, science and listening are all essential. Exactly as done in the MQA test.
He, like the testers and myself, fully understand the value of all the factors.

Originally Posted by Music Fiend

And I'm under the impression that major orchestras existence preceded measurements, published papers and double blind tests...

Yes, voodoo and witchcraft preceded modern medicine also. Some are doomed to not learn from history and may still go to such practitioners. Luckily, some know better now, so modern medicine is based on science and double blind tests, as are modern orchestras.http://gap.hks.harvard.edu/orchestra...male-musicians
I think we can all be thankful the misogyny, bigotry, biases, etc. that plagued the past is well addressed today, at least for many. I love seeing more female orchestral players!

Originally Posted by Music Fiend

Is it weak? The McNamara Fallacy is not just about data mining. From Wiki:
The McNamara fallacy (also known as quantitative fallacy[1]), named for Robert McNamara, the United States Secretary of Defense from 1961 to 1968, involves making a decision based solely on quantitative observations (or metrics) and ignoring all others. The reason given is often that these other observations cannot be proven.

Yes, because blind tests don't ignore all factors, they account for them, unlike bias filled self assessed abilities uncontrolled listening. The test methods chosen by Stuart is not based solely on measurements, but by actual trust ears/just listening abilities. Red Herrings for rejecting scientific method is extremely weak.

Originally Posted by Music Fiend

I'd say being a professional violinist not only makes him an expert

That is classic Appeal to Authority fallacy. Him being a professional violinist doesn't make him immune from biases, If he can prove that he can tell the differences between the sound of violin by listening free of his beliefs, biases, etc that plague all humans, "expert" or not, then he can demonstrate his "expertise". Just like one has to take a math test and have it graded by another party, or run a 100m dash...and have it measured either against or runners..or measured time.
Being a self assessed/declared "expert" is irrelevant here, only demonstrable ability is. He can either do that, or "talk" about it like that article.

Originally Posted by Music Fiend

it probably allows him not to fall into the McNamara Fallacy or suffer from the Dunning-Kruger effect.

Perfect. Your knowledge of subjects is spot on.
https://en.wikipedia.org/wiki/Dunnin...3Kruger_effect

In the field of psychology, the Dunning–Kruger effect is a cognitive bias in which people of low ability have illusory superiority and mistakenly assess their cognitive ability as greater than it is. The cognitive bias of illusory superiority comes from the metacognitive inability of low-ability people to recognize their lack of ability; without the self-awareness of metacognition, low-ability people cannot objectively evaluate their actual competence or incompetence.

This is exactly why self assessment of ability is really, really bad idea. Thank you!
Yes, to get back to the thread subject, BStuart is acutely aware of this and chose to team with McGill U (a great Higher Ed institution btw) to do non-Dunning-Kruger "self assessment/grade my own math etc" type analysis of MQA. He made a very wise choice! Thanks again for all the info.

cheers,

AJ

Originally Posted by Music Fiend

Does that mean that he concurs with the conclusions that you've reached?

I have no idea if Bob concurs with the conclusions both McGill and I independently reached.
Seer is not among my skills.
You would have to ask him.

Originally Posted by AJ Soundfield

(...)

Self assessed, believed abilities are verboten.

(...)

He, like the testers and myself, fully understand the value of all the factors.


(...)
That is classic Appeal to Authority fallacy.
Being a self assessed/declared "expert" is irrelevant here, only demonstrable ability is. He can either do that, or "talk" about it like that article.

(...)

This is exactly why self assessment of ability is really, really bad idea. Thank you!


cheers,

AJ

I couldn´t have made this up...

Cheers!

Originally Posted by AJ Soundfield

I have no idea if Bob concurs with the conclusions both McGill and I independently reached.
Seer is not among my skills.
You would have to ask him.

I guess he'll react to the paper's conclusions. Although I'm not a seer either, I don't think it's difficult to predict that he'll raise a few objections.

Cheers!

Originally Posted by Music Fiend

I haven't read the entire paper


In my non expert and certainly not scientific approach

Specifically about the violin results and double blind tests I also like to take into account what some true experts have to say:
https://www.thestrad.com/blind-testi...t/6944.article


I tend to use my own hearing ability. Alas, I can't use measurement devices or double blind tests

I'd say being a professional violinist not only makes him an expert it probably allows him not to fall into the McNamara Fallacy or suffer from the Dunning-Kruger effect.

I couldn´t have made this up...

Cheers!

Indeed.

Originally Posted by Music Fiend

I guess he'll react to the paper's conclusions. Although I'm not a seer either, I don't think it's difficult to predict that he'll raise a few objections.

Cheers!

Well, it's of his own creation.

Btw, great game so far today, really enjoying, though felt bad for Egypt, they played hard. And I'm sure, trained hard and had demonstrable abilities to get on the field. Go figure.

cheers,

AJ

Originally Posted by AJ Soundfield

Well, it's of his own creation.

Btw, great game so far today, really enjoying, though felt bad for Egypt, they played hard. And I'm sure, trained hard and had demonstrable abilities to get on the field. Go figure.

cheers,

AJ

All my focus regarding football today is on Portugal Vs Spain!

Originally Posted by AJ Soundfield

Indeed.

I think that you were surprised by how funnily contradictory your post sounded.

Not sure if my words, even in the mix form chosen by you and taken from more than one post, are that much contradictory or funny...

Cheers!

Originally Posted by Music Fiend

All my focus regarding football today is on Portugal Vs Spain!

Well I'm also concerned about Lorenzo running away with another Motogp race.
But yes, Portugal Vs Spain will have me glued

Originally Posted by Music Fiend

I think that you were surprised by how funnily contradictory your post sounded.

Not sure if my words, even in the mix form chosen by you and taken from more than one post, are that much contradictory or funny...

Cheers!

Oh I found it all very funny, including you thinking there was contradiction.
Again, there is a reason BStuart supplied his files for expert double blind listening analysis, by experts at McGill U and not to a violinist blogger or anti-science approach types.
If one wants to know about sound>ears/just listening, he made the right choice, even if some will hate the results. Bob of course, has enough expertise to understand expertise.

cheers,

AJ

Originally Posted by AJ Soundfield

Oh I found it all very funny, including you thinking there was contradiction.
Again, there is a reason BStuart supplied his files for expert double blind listening analysis, by experts at McGill U and not to a violinist blogger or anti-science approach types.

cheers,

AJ

Yes, I must be imagining a contradiction here:

Self assessed, believed abilities are verboten.

(...)

He, like the testers and myself, fully understand the value of all the factors.(...)
This is exactly why self assessment of ability is really, really bad idea.


Happy to see that at least you didn't argue against it being funny!

Cheers!

Originally Posted by Music Fiend

I haven't read the entire paper

In my non expert and certainly not scientific approach

Specifically about the violin results and double blind tests I also like to take into account what some true experts have to say:
https://www.thestrad.com/blind-testi...t/6944.article

I tend to use my own hearing ability. Alas, I can't use measurement devices or double blind tests

I'd say being a professional violinist not only makes him an expert it probably allows him not to fall into the McNamara Fallacy or suffer from the Dunning-Kruger effect.

In the field of psychology, the Dunning–Kruger effect is a cognitive bias in which people of low ability have illusory superiority and mistakenly assess their cognitive ability as greater than it is. The cognitive bias of illusory superiority comes from the metacognitive inability of low-ability people to recognize their lack of ability; without the self-awareness of metacognition, low-ability people cannot objectively evaluate their actual competence or incompetence

Yes, I must be imagining a contradiction here

I don't see any need for imagination unless that is what one does best.
No contradiction either, it's rather consistent!

cheers,

AJ

Originally Posted by AJ Soundfield

I don't see any need for imagination unless that is what one does best.
No contradiction either, it's rather consistent!

cheers,

AJ

Apparently you've quoted the wrong post...

Hope it wasn't on purpose for comic effect...

Cheers!

Originally Posted by Music Fiend

He, like the testers and myself, fully understand the value of all the factors.(...)

Happy to see that at least you didn't argue against it being funny!

Cheers!

No argument that interpretations of it are funny as heck!

Ok, so what don't you like about the MQA McGill U study?

Originally Posted by Music Fiend

Apparently you've quoted the wrong post...

You're the expert here.

So, you're dislike of the MQA results are based on what exactly?

Originally Posted by AJ Soundfield

No argument that interpretations of it are funny as heck!

Ok, so what don't you like about the MQA McGill U study?

As I've said before, I haven't read it all.

I simply think that double blind tests are not effective as there is a considerable risk of making a type II statistical error.

And many who don't understand basic statistics think that when double blind tests fail to show statistical evidence of a difference that you can then conclude something.

Originally Posted by AJ Soundfield

You're the expert here.

So, you're dislike of the MQA results are based on what exactly?

I must defer to you as I don't pretend to fully understand the value of all the factors in any remotely complex matter.

About the MQA tests results, please refer to my post above.


Cheers!

Originally Posted by Music Fiend

As I've said before, I haven't read it all.

And many who don't understand basic statistics think that when double blind tests fail to show statistical evidence of a difference that you can then conclude something.

Well I have read it all...and they made no conclusion whatsoever, in fact the opposite.
This is the difference here...

Originally Posted by AJ Soundfield

Well I have read it all...and they made no conclusion whatsoever, in fact the opposite.
This is the difference here...

That shows that they have a grasp of basic statistics which is hardly surprising.

You appear to say that you had the same results... does that mean that you didn't conclude anything yourself?

If so, what is the point of your posts?

To say that what you and the paper concluded is that there's no conclusions whatsoever?

Cheers!

Originally Posted by Music Fiend

That shows that they have a grasp of basic statistics which is hardly surprising.

You appear to say that you had the same results... does that mean that you didn't conclude anything yourself?

Not only of statistics but testing itself. There is no such thing as absolutes.
There were some results, that mirrored mine. So within the confines of both "tests", the conclusion of those tests were basically the same.
That some stuff is improved by MQA aliasing/EQ, while others are not. It depends heavily on the original mastering, since MQA is a remaster.
There are a lot of ways to improve via remastering...or screw up.
McGill promises further testing.
It's unfortunate that certain types of folks will auto knee jerk go on attack vs blind test results when they don't coincide with their beliefs, but I think we covered that pretty well already, thanks for your links and known effects.
Bottom line, BStuart knew what the risks were when he started this whole mess. You reap what you sow.
If he doesn't like the results, he knows exactly what to do.

Btw, you can't know this, but his own previously submitted work, The Audibility of Typical Digital Audio Filters in a High-Fidelity Playback System caught some blowback in the the AES and worse, his follow up Further Investigations of the Audibility of Digital Audio Filters in a High-Fidelity Playback System has being stuck in "Under Review"..since 2016. Not good.
It is therefore understandable that he wouldn't oppose the McGill study, though obviously it will have no effect on belief. I'm sure Bob knows that.

cheers,

AJ