MQA Deep Dive - I published tracks on Tidal to test MQA

[BS5711]

It is giving them an unfair advantage. But it is you who is giving it to them, by doping. The files prefer chilled white wine. Red wine only affects you.

Freudian slip ??

 

[Audiophiliac]

Remember records left on the dashboard in sunlight? CD's will eventually go that way, too. A thumb drive is tougher temperature-wise, is your waterfall manmade, or just mildly naturally filtered? A thumb drive actually gives a quieter read than a cd, but possibly more importantly, a more reliable read. Remember how magnified a recordable cd's read problems were, and they had to be written at your slowest speed possible for best results? My USB ssd is silent and reliable, although I need to get a better I2S interface to my DAC than usb going on.
It certainly does sound like a good place to listen, though. Especially compared to a downtown apartment. Wait, next to a waterfall?
If your subscription to the OCD guy channel is boring, have you seen his latest video? He probably just takes a little more time to get used to, stay tuned.

Ok, I understand, if you can't hear any difference, why would anyone care about this thread in the first place? Well, as long as nobody wants to say I have to be crazy for hearing a difference, I had long since forgotten about this thread...

 

[BS5711]

I wish more people would give appropriate consideration to the psychology of audio perception

 

[Elegiac]

It's a very small ornamental waterfall.

I don't say I could fall asleep because it's boring, but because it's this amusing droning. Total surety coupled to a large serve of cognitive incompetence. Not total incompetence, but a big heaping serve nonetheless. I enjoy it under circumstances where it can't do any great harm. Helps me feel comfortable with the reality of the condition of the human species. The sound of dunning-kruger being utterly negated. Sometimes it's possible to go from peak to peak, and bypass the valley of despair entirely by shooting yourself out of a cannon or something. These are the realms of subjectivity.

 

[BS5711]

sometimes I wish we had more than just a like button

 

[Jose in Jersey]

So, you didn’t even try Wikipedia? https://en.m.wikipedia.org/wiki/Master_Quality_Authenticated

It qualifies MQA codec as lossy and references, among several, this application note from the respectable Benchmark company: https://benchmarkmedia.com/blogs/application_notes/163302855-is-mqa-doa

Even my standard AI-enabled browser answers the question “Is MQA lossless?” by saying “MQA uses lossy compression…”

This prong goes to the honesty and integrity which people are holding MQA's feet to the fire on. Both of your links cite OP/Goldensound's work and it is cited by many others across the web. His was one of the louder seminal "lossy" claims, but it was based on dishonestly fabricating, manufacturing and making up his data out of thin air. He admitted it, first test tones (caught and rejected) and then altered "music." When you make up your own facts and data, you can prove anything. There is no situation in the real world (academia, journalism, medical, scientific, legal, etc...) where that type of behavior is acceptable.

Get this. MQA doesn't care about you. They didn't create MQA to benefit you, that's purely incidental to the goal: make money. Your defense of MQA signals your intellect/virtue about as much as a defense of Disney as a company would. Or Sony.
It's silly, it's cringey. Let it go.

MQA distributes a miniscule amount of music which I may or may nor enjoy. But it is an available option. The "cringey" thing is actually the confused audio justice warrior anti-MQA pile on that 1) has no idea when to be offended unless a blue light flashes, 2) bends knee themselves to lossy streaming mystery meat which they rail against and 3) beat their chest and yell loudly about honesty and integrity but then condone this type of behavior? Either way, MQA lives, MQA dies, but life goes on. Enjoy your music all.

 

[gregorio]

This prong goes to the honesty and integrity which people are holding MQA's feet to the fire on.

You don’t seem to get that a lack of honesty and integrity is precisely why “MQA’s feet are/were being held to the fire”!

His was one of the louder seminal "lossy" claims, but it was based on dishonestly fabricating, manufacturing and making up his data out of thin air.

A few points, which should be obvious but apparently aren’t for you, for some reason:
1. GoldenSound’s test was certainly the one which the audiophile community took most notice of (was “louder”), however it was certainly not the only lossy claim or even the most convincing. The most convincing was from MQA themselves! The patent application by Meridian Audio for MQA specifically stated it was lossy and even gave some detail regarding how it was lossy but then in their marketing to consumers they stated/implied the opposite, that it was not lossy. Clearly some consumers were suckered by that marketing and in fact so suckered they argued about it online without checking the actual facts.
2. What do you mean “dishonestly”? Everything submitted for encoding for MQA encoding is “manufactured and made up of data out of thin air”. The only thing GoldenSound did that was different to EVERY other submission for MQA encoding was the deliberate inclusion of some signals in a way that made it easier to analyse the results, rather than them being buried unintentionally within music recordings. This allowed him to bypass the restrictions on MQA testing, which had been uniquely imposed by Meridian because such testing would clearly demonstrate that MQA’s marketing claims were false. The dishonesty and lack of integrity in this instance was Meridian’s, if their marketing claims were true, they would have allowed users/professionals to test and demonstrate that fact, just as every other codec on the market was!
3. GoldenSound’s test demonstrated in practice not only that the marketing was false but that MQA had some relatively serious issues compared to other lossy codecs. This is why his was the “louder” voice, audiophiles could easily see the difference, without combing through a moderately complicated patent application that most wouldn’t understand anyway.

The "cringey" thing is actually the confused audio justice warrior anti-MQA pile on that 1) has no idea when to be offended unless a blue light flashes, 2) bends knee themselves to lossy streaming mystery meat which they rail against and 3) beat their chest and yell loudly about honesty and integrity but then condone this type of behavior?

No, the truly “cringy thing” is when someone invents a bunch of strawman arguments to defend a belief in false marketing! Namely:
1. A flashing blue light is completely irrelevant to anyone being offended.
2. Lossy streaming is not “mystery meat” and no one is railing against it. What we’re railing against is lossy streaming falsely marketed as lossless and a company trying to gouge fees out of the industry based on that false marketing!
3. Who is condoning their lack of honesty and integrity, besides you apparently?

G

 

[Luckyleo]

You don’t seem to get that a lack of honesty and integrity is precisely why “MQA’s feet are/were being held to the fire”!

A few points, which should be obvious but apparently aren’t for you, for some reason:
1. GoldenSound’s test was certainly the one which the audiophile community took most notice of (was “louder”), however it was certainly not the only lossy claim or even the most convincing. The most convincing was from MQA themselves! The patent application by Meridian Audio for MQA specifically stated it was lossy and even gave some detail regarding how it was lossy but then in their marketing to consumers they stated/implied the opposite, that it was not lossy. Clearly some consumers were suckered by that marketing and in fact so suckered they argued about it online without checking the actual facts.
2. What do you mean “dishonestly”? Everything submitted for encoding for MQA encoding is “manufactured and made up of data out of thin air”. The only thing GoldenSound did that was different to EVERY other submission for MQA encoding was the deliberate inclusion of some signals in a way that made it easier to analyse the results, rather than them being buried unintentionally within music recordings. This allowed him to bypass the restrictions on MQA testing, which had been uniquely imposed by Meridian because such testing would clearly demonstrate that MQA’s marketing claims were false. The dishonesty and lack of integrity in this instance was Meridian’s, if their marketing claims were true, they would have allowed users/professionals to test and demonstrate that fact, just as every other codec on the market was!
3. GoldenSound’s test demonstrated in practice not only that the marketing was false but that MQA had some relatively serious issues compared to other lossy codecs. This is why his was the “louder” voice, audiophiles could easily see the difference, without combing through a moderately complicated patent application that most wouldn’t understand anyway.

No, the truly “cringy thing” is when someone invents a bunch of strawman arguments to defend a belief in false marketing! Namely:
1. A flashing blue light is completely irrelevant to anyone being offended.
2. Lossy streaming is not “mystery meat” and no one is railing against it. What we’re railing against is lossy streaming falsely marketed as lossless and a company trying to gouge fees out of the industry based on that false marketing!
3. Who is condoning their lack of honesty and integrity, besides you apparently?

G

Very well said, and a great recap. Thanks for your time!

 

[castleofargh]

This prong goes to the honesty and integrity which people are holding MQA's feet to the fire on. Both of your links cite OP/Goldensound's work and it is cited by many others across the web. His was one of the louder seminal "lossy" claims, but it was based on dishonestly fabricating, manufacturing and making up his data out of thin air. He admitted it, first test tones (caught and rejected) and then altered "music." When you make up your own facts and data, you can prove anything. There is no situation in the real world (academia, journalism, medical, scientific, legal, etc...) where that type of behavior is acceptable.


MQA distributes a miniscule amount of music which I may or may nor enjoy. But it is an available option. The "cringey" thing is actually the confused audio justice warrior anti-MQA pile on that 1) has no idea when to be offended unless a blue light flashes, 2) bends knee themselves to lossy streaming mystery meat which they rail against and 3) beat their chest and yell loudly about honesty and integrity but then condone this type of behavior? Either way, MQA lives, MQA dies, but life goes on. Enjoy your music all.

https://patentimages.storage.googleapis.com/pdfs/c9207ae3fbb2bdf725cc/US20150154969A1.pdf
There are more options than they ever used or planned to use(it's a patent, after all), but you can see that nearly all options declare a lossy portion, and the option to attenuate the amplitude in case there is high energy ultrasound content, so it can fit inside the allocated space.
Was the fully lossless option(technically lossless, the lower bits are still discarded to make room for the ultrasonic data) ever used? Would it make any sense to use it for streaming, where the only allege benefit is to use smaller files?
To me the conclusion was clear enough when I first read this and another one for the filter in the DAC from Meridian all those years ago. But that's not important for the format, IMO. Lossy as used for formats is simple. You take a signal, you encode it, then decode it, and what comes out must be the same as what went in. Then that's a lossless format.
So dither used when decoding= not lossless.
Noise shaping= not lossless.
Changing the attenuation of the ultrasonic signal to fit inside the allocated space=not lossless.
Actually using a format that use lossy encoding or decoding or both depending on the scheme and resolution desired= most likely not lossless for the files you used on a streaming service as that's where size matters the most.

So, I don't know what you think, but it seems like MQA might not be lossless.
I never cared about that myself, plenty of lossy formats are my friends.

What people didn't appreciate, one bit, me included, was the BS marketing! Personally, I got hooked on their false comparison between the famous 5μs from a paper about the best temporal delay perceived by some guy, and the period between 2 samples. At some point, they would bring that up in every interview.
But here's the thing, the temporal resolution of a PCM file is not the delay between 2 samples! Anybody who pretends to care about that can find out it's wrong by several orders of magnitude(or more for 24bit).

Just as a joke, I loved how insistent they were about temporal resolution, but never talked about the discarded bits at the encoding process that reduced temporal resolution ^_^.
It's the risk when you try to milk gullible people and manipulate them. When some of them realize what's going on, they might hold a grudge.

 

[m-i-c-k-e-y]

Couldn't find a thread, this is the closest one:

https://x.com/mqalabs/status/1864671282333176278

Whitepaper: https://mqalabs.com/wp-content/uploads/2024/12/MQA-Labs-QRONO-White-Paper.pdf

 

[GoldenSound]

but it was based on dishonestly fabricating, manufacturing and making up his data out of thin air.

Please elaborate on what data was fabricated, manufactured or made up

 

[GoldenSound]

He admitted it, first test tones (caught and rejected) and then altered "music."

My tests used a variety of different methods, the 'rejected' stuff was files that only contained test tones (no surprise MQA wouldn't want that being put through as it'd expose issues in the encoder). Then I published my own music with test tones hidden inside.

But in the videos I also used readily available mainstream music such as from Sam Smith to show that the same issues were present even when no test tones were present in the file. I've also shared info on how to repeat my tests with any track on tidal and others have done so.

 

[GoldenSound]

So, I don't know what you think, but it seems like MQA might not be lossless.

The biggest indicator that would show this to be the case is probably just the fact that as soon as my video came out and before they made any response they went through and scrubbed the term "lossless" from all of their marketing/website materials

Or changed the claims from being lossless to "better than lossless!"

 

[Audiophiliac]

It was true after Tidal bou

The biggest indicator that would show this to be the case is probably just the fact that as soon as my video came out and before they made any response they went through and scrubbed the term "lossless" from all of their marketing/website materials

Or changed the claims from being lossless to "better than lossless!"

It true in the "After" that finally MQA came from a rights owner, except that second, shut up about the music label, liars.
In the first place, MQA is trying to be coy about how if just tell a processor to "un-hide" the samples inside of the original samples, to play multiples of the same sample to reach the original bitrate again, how could you argue that your new compression scheme is losing any playback bandwidth? Oh, but it actually got worse than that, because it turns out your dog wants to kill your favorite artists, for making them have to live with that NOISE that your bandwidth got replaced with, for 2x$=$$. Yup, MQA was the noise-injected plan, that made me have to have Qobuz also, for proper listening of my game-changing 8 custom smart AI playlists. However, if someone comes out with a 10x custom playlists streaming service that doesn't use Audirvana, I don't know what will happen besides still using Tidal for proper audio monitoring. I just need to get on the "headphones prefer the electrostatic driver type's trade of impact for speed" plan, exactly. Except these were so the best ones for the money, until +3k. Then also, obviously my amp won't be showing me what my headphones could all have been actually doing better, next. Nobody wants it to have to be a better DAC, since reproduction's player will always just be cover versions by your crappy gear, that will never make it to anywhere near equal to the original instruments plugged straight into your hi-fi amp, anyways.
What if one of your gear designers starts to seem like his job goes with you better than most music does? Oh man, I bet the most successful gear makers all seem like they get to be slackers who think whatever is people's favorite stuff, is their favorite stuff for them to be into, also, all the time, lol! Shut up, they have a right to be as mean to everyone else as possible, any time they want, same as everyone does in the first place!