MQA is Vaporware

[The Computer Audiophile]

2 minutes ago, Tony Lauck said:

nor will I purchase any recordings or deal with streaming services that support MQA

Does this extend to not purchasing music from the labels who signed on with MQA?

[Fair Hedon]

19 minutes ago, Shadders said:

Hi,

I think MQA could be investigated by the relevant Advertising Standards Authority here.

 

1. The first unfold is the extraction of higher frequency information from the baseband 0Hz to 22kHz (?) bandwidth, and as such the first unfold adds lossy higher bandwidth information (22kHz to 44kHz OR 24kHz to 48kHz).

 

2. The second unfold does NOT include any embedded information, but is upsampling. No new information. This means by MQA's own definition, they are not unfolding a second time - they are upsampling which is a completely different mechanism.

 

As such, MQA are in breach of any advertising standards since stating a second unfold occurs, means that they are stating NEW information is being added between 48kHz to 96kHz, when in fact it is not, and there is NO new information in this frequency band.

 

Regards,

Shadders.

I think you are very much on to something.  Along with the misleading technical nonsense, their other big whoppers, like "authentication", and "correcting" for the "original" ADC could be addressed.

 

However this is all far too esoteric for any mainstream consumer protection agency.

 

Our first line of defense, the self crowned so called golden eared authorities in the "press" have failed us.

 

Thankfully, and we should be thankful, others in the industry made a lot of noise...

[Dr Tone]

7 minutes ago, The Computer Audiophile said:

Does this extend to not purchasing music from the labels who signed on with MQA?

 

Ugggh, what if they start downsampling and dithering the MQA master to cut the CD.   We need some way of stopping this mess.

[The Computer Audiophile]

4 hours ago, Fair Hedon said:

Which shows you are being, what was it you said...a useful what?

 

MQA upsamples anything above 96 to the original sample rate. If you get 192 it is coming form an upsampled process.

 

LOSSLESS.

 

Do your homework then come back. Maybe next time you won't get a knew corn hole the size of a comet crater torn

into your backside.

Tone it down.

 

You're calling him a ??? because of repeated a technical claim he was told? That's not the test I'd use to determine this. 

 

The corn hole comment. A bit abrasive, unneeded, and borderline against the rules. 

[Fair Hedon]

3 minutes ago, The Computer Audiophile said:

Tone it down.

 

You're calling him a ??? because of repeated a technical claim he was told? That's not the test I'd use to determine this. 

 

The corn hole comment. A bit abrasive, unneeded, and borderline against the rules. 

I understand your intent here, and you are right. However, Mr. Scoggins in clearly engaging in classic trolling.

 

Playing dumb, ignoring corrected information, floating opinions as facts, etc.

 

I do see, that he also did receive a similar warning, so your note to me is fair.

[The Computer Audiophile]

2 minutes ago, Fair Hedon said:

I understand your intent here, and you are right. However, Mr. Scoggins in clearly engaging in classic trolling.

 

Playing dumb, ignoring corrected information, floating opinions as facts, etc.

 

I do see, that he also did receive a similar warning, so your note to me is fair.

Cool. I'm trying to follow along and keep the correct context etc... it's tough :~)

[Fair Hedon]

I hear you. Perhaps what makes posters over react or get nasty more than anything else is hypocrisy.

 

And thankfully, that gets called out pretty much every time. And I am sure the list of banned posters who

could not handle being exposed is a mile long.

 

 

[semente]

20 hours ago, crenca said:

 

Ha!  You used the word "elegant". This in no way helps your cause to convince us that you are not an anti-consumer shill for the industry.

 

I noticed that too.

[semente]

20 hours ago, crenca said:

 

 

Lee, I gave you a thumbs up because your explicitly honest about who is manipulating who in your scenario. 

 

On this board however you're dealing mostly with audiophiles

 

@GUTBill come along in a minute to set things straight.

[Tony Lauck]

40 minutes ago, The Computer Audiophile said:

Does this extend to not purchasing music from the labels who signed on with MQA?

Depends.  Unfortunately, many labels have victimized their artists, who are generally innocent of all of this skullduggery.   But some labels have been on my do not buy list for decades, generally because they have a track record of poor sound quality.  I suspect quite a bit of overlap, here. 

[hsmeets]

unfold, upsample.....trying for myself to imagine how one can pack 196khz sample rate information in a file that has a 44khz base sample rate.

 

Creating the packed file:

 

So you take the original 196/24 file and downsample it to 44/24 by removing 3 out of 4 samples. But you need to store some information about those 3 samples in that 1 surviving sample.

 

Okay, given that the LSB's (least significant bits) are noise you could truncate them, let's assume you could free up to 6 bits, and store information about this 3 removed samples here. Obviously you can't store 3 x 16 bits of data in 6 bits.....

 

A method could be to encode the 3 samples as delta's to the surviving samples. 6 bits available so, for each sample you could use 2 bits. But even under the assumption that sample to sample delta's are small, 2 bits are more than likely not enough to store the delta. Also mind that we defined 6 bits as noise and now we encode data in 2 bits....

 

Another approach could be that you encode some information of the 3 samples by a coefficient that describes a curve-fit between the surviving samples. 6 bits might be enough to add 1 coefficient with sufficient precision. Let's assume that this is the method.

 

Unpack the file for replay:

 

So now you start to unpack the file and the DAC has to restore the 196/24 file from the 44/24 file.......it takes the 6 bits from each survived sample and calculates for each survived sample 3 extra samples by curve fitting over multiple survived samples using the coefficients stored in those samples. 

 

Obviously the output file will never be a bit-perfect copy of the 196/24 we started with. Small details (above 44Khz) will be smoothed out by the curve fitting. Lossy thus.

 

This leaves only one question open: does this scheme perform significantly better (less distortion) with the help of the stored coefficients versus an upsample that does curve fitting without the help of the encoded coefficients?

 

 

 

 

[Lee Scoggins]

I think people are misreading my use of the term "elegant".  It's a phrase used in the management consulting industry to describe business models.  Elegant means the business model is well thought out, it doesn't imply a particular love for the company implementing it.  It is a recognition of a business model that aligns incentives which means that all the participants are likely to individually have financial incentives to participate and that the alignment is such that good profits should result.  "Business model" is a high level framework of how the money flows.

 

For instance, in my day job I build business models for data and analytics.  One of the more powerful models is a "contributory data pool" whereby companies contribute data to a third party (which may be owned by the participants) and the third party mines the data for insights that can help their business such as providing better customer prospect lists, more data on risk for insurance or banking clients, and other ways to improve revenue.  Some of these models started from just collecting data for meeting government regulations at scale and then developed into machine learning based insights.

 

I believe the MQA model is elegant because it aligns the incentives of the record labels, hardware vendors, and the consumers toward a model for streaming music which you can think of as monetizing the music library away from one-time purchases into subscription revenue.

 

I also think that HDTracks is elegant too as it provides a portal for high quality downloads for purchase and the artist keeps around 65% of the proceeds.  Streaming has work to do in paying the artist properly.

[Dr Tone]

10 minutes ago, Lee Scoggins said:

I believe the MQA model is elegant because it aligns the incentives of the record labels, hardware vendors, and the consumers toward a model for streaming music which you can think of as monetizing the music library away from one-time purchases into subscription revenue.

 

An elegant solution would be streaming the original hi-res PCM master in a FLAC container, using our existing hardware and still generate subscription revenue.  No need for new MQA hardware, licensing fees etc etc.

 

This would truly be a win for the consumer.

[mansr]

3 minutes ago, hsmeets said:

unfold, upsample.....trying for myself to imagine how one can pack 196khz sample rate information in a file that has a 44khz base sample rate.

 

Creating the packed file:

 

So you take the original 196/24 file and downsample it to 44/24 by removing 3 out of 4 samples. But you need to store some information about those 3 samples in that 1 surviving sample.

MQA only tries to preserve one of those 3 samples. To do so, it relies on the fact that most music has very little energy above 24 kHz, so together with some compression, the 8 low bits of the 24-bit samples are sufficient. That's what the "triangle" talk is all about. If there's too much high-frequency content, more than 8 bits per sample may be allocated at the expense of base band resolution.

[crenca]

8 minutes ago, Lee Scoggins said:

 

 

I believe the MQA model is elegant because it aligns the incentives of the record labels, hardware vendors, and the consumers toward a model for streaming music which you can think of as monetizing the music library away from one-time purchases into subscription revenue.

 

 

 

Interesting.  The word has a history in relation to MQA that indicates the group think in the audiophile press.  

 

I agree that the labels have desire or "incentive" for an DRM, proprietary format as the standard.  I do not believe it will give them what they want however (i.e. a reversal of their calamitous decline).  Only the ever popular short term thinking so often found in busines would lead a hardware vendor to believe that turning over part of their innovation and design freedom to MQA would be good for them in the long run.  Some of them of course have recognized this (I suspect most).  The idea that consumers have an incentive to be locked in to an IP/DRM format, or the choices in (e.g. in their digital worlds, in equipment, etc.) that would result from that is rather absurd.  It is the kind of thing someone would say if they were trapped in the industries perspective such that they either had lost touch with the consumer completely or simply did not care. 

[Lee Scoggins]

1 hour ago, Tony Lauck said:

All the money spent buying new DAC hardware is going to vultures.  Consumers have limited funds and any money that doesn't go to the musicians and song writers and the engineers who made the original recordings is going to parasites.

MQA is nothing but a parasitical scheme involving the legal system to extract rent from music lovers at the expense of everyone else. (Legal aspects include proprietary formats backed up with non-disclosure agreements and patents and the use of encryption that can't legally be "circumvented" due to fascist legislation such as the US DMCA law.)

 

IMO the invention of a proprietary digital audio format should be treated as a capital crime under Napoleonic law, where the defendant is presumed guilty unless he can prove his innocence.  But since I'm not an emperor I will just have to apply what little market power I have.  For starters, this means I will not purchase any DAC from a manufacturer who sells MQA products, nor will I purchase any recordings or deal with streaming services that support MQA.  To be specific,  once I learned that Mytek has signed on to MQA, I have removed them from my "acceptable vendor list" despite my being satisfied with Mytek DAC that I have owned for several years.

 

 

 

Hi Tony,

 

I am sensitive to the customer having limited money for new DACs as well. We have all been there.  But I think you are looking at this a bit wrong.  I'll explain.

 

In a free market system that music exists in (albeit with some oligopolistic dimensions), the Customer is truly King.  If MQA creates this ecosystem and the Customer doesn't buy the streaming subscription (either hirez or no hirez) then the format will collapse and record labels will fade it out.  We have seen this happen to SACD where Sony turned off the music titles tap after a while and in DVD-Audio where most labels stopped creating new discs.

 

The Customer really has control now. So I think you need to look through that lens than the lens of big corporate players creating evil schemes.  That may have been the case in the CD era where the business model for a label was "we have the master tapes and you don't and we will put out what we want when we want."

 

Fortunately that world is long gone.  The web and mobile has concentrated power into the hands of the consumer.

 

So the consumer has to get some value out of MQA and money has to flow.  To do that, there has to be a compelling value proposition to buying MQA files or streaming access.

 

As for consumer options, there are several sub-$300 MQA DACs from Audioquest, ProJect, and Meridian.  It's not a burden on the consumer either way.

 

For wealthier audiophiles, it will likely be decided by whether their favorite DAC maker integrates MQA/Roon and whether their own personal listening shows them that MQA files sound better or not.

[hsmeets]

1 minute ago, mansr said:

MQA only tries to preserve one of those 3 samples. To do so, it relies on the fact that most music has very little energy above 24 kHz, so together with some compression, the 8 low bits of the 24-bit samples are sufficient. That's what the "triangle" talk is all about. If there's too much high-frequency content, more than 8 bits per sample may be allocated at the expense of base band resolution.

 

mhmm, is that all so novel that it can be patented? Sounds (pun not intended) as very straightforward to me that someone would own a patent already long time before.

 

And taking your comment in account that the base band resolution is varying......does that not imply that maybe even the <22Khz is lossy ?

[Shadders]

1 hour ago, Fair Hedon said:

I think you are very much on to something.  Along with the misleading technical nonsense, their other big whoppers, like "authentication", and "correcting" for the "original" ADC could be addressed.

 

However this is all far too esoteric for any mainstream consumer protection agency.

 

Our first line of defense, the self crowned so called golden eared authorities in the "press" have failed us.

 

Thankfully, and we should be thankful, others in the industry made a lot of noise...

Hi Fair Hedon,

Taking this further, then this may have serious ramifications for MQA.

 

Assume that the Master is 192kHz, with 96kHz bandwidth - 24bits depth.

1. If the file contains NO information for the frequency band 48kHz to 96kHz (is upsampled), then this is NOT a Master quality file, since the information from 48kHz to 96kHz is missing.
2. Therefore the selling/streaming of said file as Master Quality is a false sale - Fraudulent.
3. As such, MQA cannot sell any file derived from a 192kHz Master as Master Quality, if half the frequency band information is missing. (fraud).
4. If the majority of MQA files are as above (192kHz based), then their system is misrepresented by calling it MQA = Master Quality Authentication, since it is not a Master Quality.
5. Therefore, they may then need to stop using the acronym MQA and name Master Quality, as it is a false representation of what they are selling,

Regards,

Shadders.

[Lee Scoggins]

4 minutes ago, crenca said:

 

Interesting.  The word has a history in relation to MQA that indicates the group think in the audiophile press.  

 

I agree that the labels have desire or "incentive" for an DRM, proprietary format as the standard.  I do not believe it will give them what they want however (i.e. a reversal of their calamitous decline).  Only the ever popular short term thinking so often found in busines would lead a hardware vendor to believe that turning over part of their innovation and design freedom to MQA would be good for them in the long run.  Some of them of course have recognized this (I suspect most).  The idea that consumers have an incentive to be locked in to an IP/DRM format, or the choices in (e.g. in their digital worlds, in equipment, etc.) that would result from that is rather absurd.  It is the kind of thing someone would say if they were trapped in the industries perspective such that they either had lost touch with the consumer completely or simply did not care. 

 

I remain unconvinced so far that DRM is a big driver for any of this but let's assume it is in place for arguments sake...

 

If the consumer enjoys the sound of MQA files and feels they sound better than non-MQA files then they may ignore DRM concerns for the convenience of better sound, library access, or both.

 

Let's further assume that MQA is not quite 24/96.  Let's say that James Guthrie convinces Pink Floyd to do MQA.  If the MQA version of Animals is an "almost 24/96" version of Animals, then that would be the best digital experience and I think people would get intrigued and might up their $10 Tidal subscription to a $20 subscription for Masters access.

 

{Note: Guthrie is a big DSD fan so he might opt just to put out the catalog with Chad as Acoustic Sounds but I am ignoring that for sake of argument.}

 

In both cases, the Customer is King and decides whether the product lives or dies.

[mansr]

Of a bit over 100 MQA tracks (from their featured playlists) I checked on Tidal, roughly half were from 48 kHz masters. Where's that high-res now?

[Samuel T Cogley]

17 minutes ago, Lee Scoggins said:

 

In a free market system that music exists in (albeit with some oligopolistic dimensions), the Customer is truly King.

 

"The Customer is King"  Hmm.  Where have I heard that before?

 

Oh yeah, this little gem that predicted crowd funding would forever change the face of audiophilia.  I wonder where Gavin Fish is now...

[mansr]

3 minutes ago, Lee Scoggins said:

If the consumer enjoys the sound of MQA files and feels they sound better than non-MQA files then they may ignore DRM concerns for the convenience of better sound, library access, or both.

What will the consumer do when MQA eventually goes out of business, the format details are lost, and compatible hardware is no longer made? How will his MQA files sound then?

[GUTB]

58 minutes ago, semente said:

 

@GUTBill come along in a minute to set things straight.

 

Responding to the call. Yes, I'll correct the statement: "mostly non-audiophiles".

[semente]

4 minutes ago, GUTB said:

 

Responding to the call. Yes, I'll correct the statement: "mostly non-audiophiles".

 

I've been meaning to ask you if you agree with the following:

 

IF MQA is lossy AND lossy is in many cases not as good as the high res Master THEN MQA is not high-end (thus not audiophile either)

[tmtomh]

11 minutes ago, Lee Scoggins said:

 

Hi Tony,

 

I am sensitive to the customer having limited money for new DACs as well. We have all been there.  But I think you are looking at this a bit wrong.  I'll explain.

 

In a free market system that music exists in (albeit with some oligopolistic dimensions), the Customer is truly King.  If MQA creates this ecosystem and the Customer doesn't buy the streaming subscription (either hirez or no hirez) then the format will collapse and record labels will fade it out.  We have seen this happen to SACD where Sony turned off the music titles tap after a while and in DVD-Audio where most labels stopped creating new discs.

 

The Customer really has control now. So I think you need to look through that lens than the lens of big corporate players creating evil schemes. 

 

Lee, I'm trying really hard to believe that you're not trolling - I'm not talking about conventional trolling, but rather promotional trolling, keeping yourself at the center of the conversation by playing whack-a-mole with the various claims and arguments being made here, selectively picking and choosing what you want to respond to.

 

But I will put that aside to see if you are willing to be accountable for what you yourself have stated here. In this latest comment you write that the "customer is king," and that MQA will die if customers don't buy streaming subscriptions.

 

But you've repeatedly stated in prior comments that:

1. Audiophiles are not the main customer for streaming services;
2. MQA is a convenient way to deliver high-res music to a broader audience not primarily interested in that; and
3. MQA is just "along for the ride" (your words) and streaming is the main product.

Your own arguments clearly show that the customer is not king when it comes to MQA, because streaming subscribers are not paying for MQA - they're paying for streaming subscriptions in which parts of the streaming library happen to be MQA-encoded.

 

Customers will "choose" (or not choose) MQA only if all the labels and streaming services make MQA-encoded streams a separate, higher priced or add-on subscription produce. But if MQA becomes integrated into the base subscription price (or a lossless subscription tier), then the customer has no choice.

 

This is the classic problem of vertical integration and oligopolistic practices, and it's the #1 myth about free markets: Consumers routinely are presented with clusters of choices that can't (or won't) be unlinked, meaning that consumers have no choice. Your own posts have shown that MQA's real customers are the record labels. And there is not a one-to-one pass through of costs and profits from the labels to the consumer.

 

So if you want to be taken at all seriously, you need to either recant your earlier claims about the MQA market and how streaming services work, or else you need to recant your more recent claim that the customer is king when it comes to MQA. I am not optimistic that you will do either, but I am happy for my pessimism to be proven unfounded.