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

[scott wurcer]

If you have a set of data: A B C 1 2 3 4 5, with just the numbers being relevant. Then a guy comes and say, "it would be nice to gain tthe space of those letters for something that means something for our process, those letters are useless for our math and everybody else; my process will discard those letter. The numbers will be lossless in the process"
.

This makes about as much sense as, "See Spot run"

 

[RichB]

Although I'd like to take credit, I don't have that much technical skill although still the small amount of common sense to understand that MQA is snake oil.

It was @GoldenOne who did the tests.

Well, thank again for sneaking it by MQA.

- Rich

 

[goldenears]

If you have a set of data: A B C 1 2 3 4 5, with just the numbers being relevant. Then a guy comes and say, "it would be nice to gain tthe space of those letters for something that means something for our process, those letters are useless for our math and everybody else; my process will discard those letter. The numbers will be lossless in the process"
Then the machine outputs: "* / - 1 2 3 4 5". My machine replaced useless letters by useful command to operate over the significant data of the numbers. But then you come and say..."Hey! you said your process would be lossless and it is not, the resulting word differs with the input!". Your algorithm is lying to me! You lied, you are dishonest!.
That's exactly what this test is saying about the issue of noise: the letters in this example. Really, really, I can't figure out how this incredibly obvious thing cannot be understand, other than people here just don't WANT to understand it.

I find it hard to tell which side you are arguing for.

MQA is claiming that their process is lossless. As you've explained, it isn't.

“Extraordinary claims require extraordinary evidence”

MQA are making the extraordinary claim that the letters are unimportant so they are thrown away, even doubling down and making an even more extraordinary claim that replacing those letters actually improves the sound, while simultaneously and falsely claiming that their process is lossless, which by definition would require the letters to be retained.

They want to have their cake and eat it too, all the while charging licensing fees for the privilege.

 

[Chrispy]

Okay my final comment about mqa..... it's just audio bullshit that is no more necessary than silly cabling.

 

[raistlin65]

Labels already adopted FLAC.

Even MQA uses FLAC.

Yep. And MQA doesn't really seem to do anything better than FLAC (unless someone is into ultrasonics). Whether or not it's inferior doesn't really matter.

Seems silly to champion a closed format that has licensing fees when an open format is already widely used in the market that can do the same job.

 

[2poor4speakers]

Can't say I feel anything but disappointment after following this thread for nearly 100 pages. Thanks to those of you who continue to stay true to the goal of objective understanding - it's difficult to wade through the BS in contentious issues like this as I try to grow my understanding of scientific audio concepts.

 

[Chrispy]

Yep. And MQA doesn't really seem to do anything better than FLAC (unless someone is into ultrasonics). Whether or not it's inferior doesn't really matter.

Seems silly to champion a closed format that has licensing fees when an open format is already widely used in the market that can do the same job.

Depends on how much magic you believe is applied by "mqa'ing" it.

ps sorry, dammit, that's my third closing comment now. I'm unsubscribing now....

 

[KeenObserver]

Labels already adopted FLAC.

Even MQA uses FLAC.

I was being facetious.

 

[KeenObserver]

I think all music lovers need to thank GoldenOne for exposing what MQA is so desperately trying to hide.

 

[tmtomh]

If you have a set of data: "A B C 1 2 3 4 5", with just the numbers being relevant. Then a guy comes and say, "it would be nice to gain the space of those letters for something that means something for our process, those letters are useless for our math and everything else; my process will discard those letters. The numbers will be lossless in the process"
Then the machine outputs: "* / - 1 2 3 4 5". My machine replaced useless letters with useful command to operate over the significant data of the numbers. But then you come and say..."Hey! you said your process would be lossless and it is not, the resulting word differs with the input!". Your algorithm is lossy! You lied, you are dishonest!.
That's exactly what this test is saying about the issue of noise: the letters in this example. Really, really, I can't figure out how this incredibly obvious thing cannot be understand, other than people here just don't WANT to understand it.

Your analogy establishes the value of MQA only if the letters are indeed useless - which they are not; and only if what the letters are being replaced with is "meaningful for our process" - which it is not.

So you've actually captured perfectly why and how MQA is simultaneously unnecessary and harmful: it does something harmful to the musical data in order to insert other data that is unnecessary.

I can't figure out how this incredibly obvious thing cannot be understood; I guess you just don't WANT to understand it.

 

[goldenears]

I think all music lovers need to thank GoldenOne for exposing what MQA is so desperately trying to hide.

Personally I'm praying that the Russian hackers are interested enough to break whatever encryption is on the MQA decoder chips and reverse engineer it completely, or the MQA encoder software is leaked.

That would be hilarious as then anyone could independently test MQA, and the whole obvious bullshit house of cards would come crashing down.

 

[Don Hills]

... and still no explanation of how converting a 16/44.1 track to MQA makes it better than the source.

 

[scott wurcer]

... and still no explanation of how converting a 16/44.1 track to MQA makes it better than the source.

Me too, no answer to better than lossless.

 

[gatucho]

... and still no explanation of how converting a 16/44.1 track to MQA makes it better than the source.

Requires no explanation, it just is. And for free... What is not to like...?!
That's what the artist intended.. to confuse you jajaja

 

[blueone]

Me too, no answer to better than lossless.

Me three. No answer to listing any end-user benefit at all.

 

[goldenears]

That's what the artist intended.. to confuse you jajaja

Apparently artists had intentions previously unheard on recordings that were created before MQA existed.

Chewbacca intentions.

 

[Raindog123]

Well, I‘ve just counted, and I probably missed a few...

I personally asked about “benefits of the 16/44 MQA (aka MQA-CD)” within this thread 7 times! Got 0 (zero) answers.
I asked for a “proof of [any] MQA sounding better than non-MQA, through controlled listening or measurements” 11 times! Guess what, 0 (zero) responses.
I asked about why we need ultrasonics at least 5 times (while the hearing curve published by MQA/Meridian stops under 20kHz)... same story.

I must be a cellophane man... Or, maybe there are no answers!?

 

[Cheztumoé]

Amen...

 

[tmtomh]

Well, I‘ve just counted, and I probably missed a few...

I personally asked about “benefits of the 16/44 MQA (aka MQA-CD)” within this thread 7 times! Got 0 (zero) answers.
I asked for a “proof of [any] MQA sounding better than non-MQA, through controlled listening or measurements” 11 times! Guess what, 0 (zero) responses.
I asked about why we need ultrasonics at least 5 times (while the hearing curve published by MQA/Meridian stops under 20kHz)... same story.

I must be a cellophane man... Or, maybe there are no answers!?

As far as we understand, there is literally no place to put the ultrasonic information - the data from the original higher sample rate - on an MQA-CD because the data on the CD is 16-bit, not 24-bit. It would seem that an MQA CD either stuffs a very small amount of lossy encoded ultrasonics in the 14th, 15th, and 16th bits, or else it simply uses those bits to encode "authentication" information and the instruction/trigger to the MQA-compatible playback equipment to employ the MQA "apodizing" digital reconstruction filter upon playback.

In this respect, MQA CDs would seem to be even more fraudulent that MQA high-res digital/streaming files, as the MQA CD packaging routinely contains printed statements about "352.8kHz" resolution/sample rates, which (if you'll excuse the contradiction of trying to talk about degrees of an absolute thing like impossibility) is even more impossible than the resolution/sample-rate claims made for digital MQA files.

 

[gatucho]

As far as we understand, there is literally no place to put the ultrasonic information - the data from the original higher sample rate - on an MQA-CD because the data on the CD is 16-bit, not 24-bit. It would seem that an MQA CD either stuffs a very small amount of lossy encoded ultrasonics in the 14th, 15th, and 16th bits, or else it simply uses those bits to encode "authentication" information and the instruction/trigger to the MQA-compatible playback equipment to employ the MQA "apodizing" digital reconstruction filter upon playback.

In this respect, MQA CDs would seem to be even more fraudulent that MQA high-res digital/streaming files, as the MQA CD packaging routinely contains printed statements about "352.8kHz" resolution/sample rates, which (if you'll excuse the contradiction of trying to talk about degrees of an absolute thing like impossibility) is even more impossible than the resolution/sample-rate claims made for digital MQA files.

Impossibler