MQA: A Review of controversies, concerns, and cautions

[abdo123]

I will not, for the following reason: any time the MQA "authentication" light lights up for any sample-rate resolution above 96kHz, it's a lie and has literally been altered from the shape it was received from the publisher: Any 176.4k or 192k or higher sample-rate source file is destructively downsampled by the MQA encoder to 88.2k or 96k before it applies any of its "folding" and lossy encoding. So MQA is in fact that only format that reliably misrepresents an adulterated file as something it is not.

MQA is indeed quite reliable, if you want to be confident that you're being lied to.

No you're incorrect, according to their website the tracks are encoded by the producers of the tracks themselves, that's what 'Master Authenticated Quality' represents, it's authenticated by the masters of the track. so it's not misrepresenting anything when the producers OKed it.

 

[dmac6419]

There are indeed plenty of high-res digital files that have no frequencies in them above 22.05kHz (the Nyquist frequency for rebook 44.1kHz). That's not HDTracks - that's the labels delivering high-res files from non-high-res sources. An MQA file made from the same source will have the same issue. Completely irrelevant to this discussion - the MQA "authentication" light will still go on with such a file.

Um HDTRACKS didn't have MQA files,HDTRACKS upsampled the files that the studios sent them and sold them as higher res.

 

[Tks]

Sure, why not.

Still standardized compression would be nice

One thing I haven't bothered you with, is what you're referring to when you talk about "standardized compression". Completely lost here. Is FLAC a "standardized compression"?

Or do you just mean you wish there was a single de-facto standard of meeting this proof you require for music to be the same untampered file that the producer finally lets loose? (which of course is impossible because the producer doesn't have the final say), but aside from that nitpick, am I correct in understand what you're talking about?

 

[witchdoctor]

For anyone interested the following drops in MQA are outstanding FWIW, anything by:

ZZ Top
The Doors
Talking Heads
Fleetwood Mac/Stevie Nicks

The transfers are so good you can practically tell what type of mic they are using or how they placed the mike in the studio relative to the singer or the instruments.

 

[tmtomh]

No you're incorrect, according to their website the tracks are encoded by the producers of the tracks themselves, that's what 'Master Authenticated Quality' represents, it's authenticated by the masters of the track. so it's not misrepresenting anything when the producers OKed it.

What I wrote above is a fact: MQA files (and MQA CDs) routinely claim to be in resolutions above 88.2k/96k - but by definition no MQA file has a higher resolution than that - yet the hardware light on MQA DACs, not to mention the software displays on Tidal and so on, will falsely indicate higher resolutions because the MQA final "render" upsamples the signal. MQA gives the impression that the original higher resolution is somehow preserved or restored when it is not.

This is a fact. It's not a conspiracy theory, it's not an opinion. It is a technological fact of how MQA works. Educate yourself or kindly take a seat.

 

[abdo123]

One thing I haven't bothered you with, is what you're referring to when you talk about "standardized compression". Completely lost here. Is FLAC a "standardized compression"?

Or do you just mean you wish there was a single de-facto standard of meeting this proof you require for music to be the same untampered file that the producer finally lets loose? (which of course is impossible because the producer doesn't have the final say), but aside from that nitpick, am I correct in understand what you're talking about?

I want FLAC with Authentication basically, if a well designed albeit lossy compression is added on top i wouldn't mind.

Basically the strengths of MQA without its weaknesses, done at the production level.

 

[tmtomh]

Um HDTRACKS didn't have MQA files,HDTRACKS upsampled the files that the studios sent them and sold them as higher res.

I'm not defending HDTracks - if a label supplied them with 96k files and HDTracks upsampled them to 192k and sold the 192k alongside the 96k version for a premium, that's wrong and fraudulent. However, the only information for which there is any proof is that HDTracks received upsampled files - as did other online vendors. When HDTracks has been alerted to this by customers, they have taken down the files in question. But they have kept others up, and like other online vendors have taken a "we just post what we receive from the labels" attitude instead of actively telling the labels they won't accept upsampled material. That's passing the buck and and it sucks, but it's not what you are claiming. You are ill informed and making unwarranted assertions.

So you're missing the point: an MQA file that has similarly been upsampled by the publisher before being sent on to MQA for processing will show as "authenticated" even if the original source started out life at a lower sample rate than what is claimed for it.

 

[abdo123]

What I wrote above is a fact: MQA files (and MQA CDs) routinely claim to be in resolutions above 88.2k/96k - but by definition no MQA file has a higher resolution than that - yet the hardware light on MQA DACs, not to mention the software displays on Tidal and so on, will falsely indicate higher resolutions because the MQA final "render" upsamples the signal. MQA gives the impression that the original higher resolution is somehow preserved or restored when it is not.

This is a fact. It's not a conspiracy theory, it's not an opinion. It is a technological fact of how MQA works. Educate yourself or kindly take a seat.

I never disagred with that, i just wanted to inform you that the producers of the track are scamming you by giving the track in MQA format. MQA is not doing the scamming, they're providing the tools for the scamming. the producers create that track in MQA format, not MQA.

So a track that came directly from the producers / label cannot misrepresent the track, it's illogical.

 

[dmac6419]

You're missing the point. The overwhelming majority of upsampled files are provided by the labels to all the digital file vendors - and an MQA file that is similarly upsampled by the publisher will show as "authenticated" even if the original source started out life at a lower sample rate than what is claimed for it.

Never mind I'm wasting time

 

[abdo123]

even if the original source started out life at a lower sample rate than what is claimed for it.

There was literally a member a couple of pages ago saying that there are Tracks that remain faithful to the original sample rate after unfolding on Tidal. (Which was 16/44.1)

 

[tmtomh]

Never mind I'm wasting time

You are indeed - although you said as much earlier in the thread and kept posting anyway.

 

[Tks]

I want FLAC with Authentication basically, if a well designed albeit lossy compression is added on top i wouldn't mind.

Basically the strengths of MQA without its weaknesses, done at the production level.

Okay, so yeah, then we understand one another. No need for MQA at all, technically it's nonsense (unless they can prove otherwise), and logically, it does nothing a far simpler blockchain implementation wouldn't do. MQA would be palatable if it wasn't a fail pragmatically speaking. And the pragmatic fail follows the technical aspects that people have more than once spoken of here in detail.

 

[tmtomh]

I never disagred with that, i just wanted to inform you that the producers of the track are scamming you by giving the track in MQA format. MQA is not doing the scamming, they're providing the tools for the scamming. the producers create that track in MQA format, not MQA.

So a track that came directly from the producers / label cannot misrepresent the track, it's illogical.

No, you are mixing up two different types of scamming:

1. Publisher takes 44.kHz CD file and upsamples it to 96k, then sends it to HDTracks, MQA, 7Digital, Qobuz, or whomever. In that case the publisher is scamming customers.
2. Publisher sends a genuine, non-upsampled 192k PCM file to MQA. MQA - based on the technological design and limitations of its encoder - downsamples it to 96k, then does all its "folding" encoding, then sells it to the customer (or to a vendor who sells it to the customer). When the customer plays it back on an MQA-enabled DAC, the "192k" light goes on. In that case MQA is scamming customers. If the publisher does the MQA encoding themselves then the publisher is colluding with MQA in the scam - although more likely is that the publisher is unwittingly participating in the scam because one has to listen very closely to Bob Stuart's interviews and pore over the MQA spec to realize that any resolution above 88.2k/96k is fake.

 

[R1200CL]

but by definition no MQA file has a higher resolution than that

So explain this then? And it can even be double. But transfer rate is max 44.1 or 48kHz. (16 or 24 bit). Or did I misunderstand your point.

 

[abdo123]

Publisher sends a genuine, non-upsampled 192k PCM file to MQA.

I'm gonna stop you right there because in principal this shouldn't happen, record companies should in principal own an MQA encoder and do the encoding themselves. Thus the 'Master (of the track) Authenticated Quality' or if it did happen the publisher gives the okay to the final product which is kinda the same thing.

it's not MQA scamming you, they're just providing the tools.

 

[R1200CL]

You're missing the point. The overwhelming majority of upsampled files are provided by the labels to all the digital file vendors - and an MQA file that is similarly upsampled by the publisher will show as "authenticated" even if the original source started out life at a lower sample rate than what is claimed for it.

You may read this.
http://bobtalks.co.uk/blog/provenance/provenance-series-1-madonna/

 

[dmac6419]

You are indeed - although you said as much earlier in the thread and kept posting anyway.

Don't get your feelings hurt,goodbye

 

[tmtomh]

So explain this then? And it can even be double. But transfer rate is max 44.1 or 48kHz. (16 or 24 bit). Or did I misunderstand your point.
View attachment 98837

This is exactly what I'm saying - look right below the second red circle you've drawn: MQA core decoder, 24 bit, 88.2kHz. That's the native sample rate of the file. 3/4 of the samples in the original 352.8k source have been thrown out and the 352.8k final output is a result of upsampling of the resulting 88.2k actual file it just copies each sample 3 times to return the sample rate to 352.8k - but 3/4 of the original samples are gone forever. That's the scam.

 

[R1200CL]

Publisher takes 44.kHz CD file and upsamples it to 96k, then sends it to HDTracks, MQA, 7Digital, Qobuz, or whomever. In that case the publisher is scamming customers.

Any prove this is happening?

 

[abdo123]

Any prove this is happening?

I think there was an incident where HDtracks (not a particular publisher) has been selling upsampled tracks but i'm not sure.