MQA is Vaporware

[asdf1000]

10 minutes ago, The Computer Audiophile said:

Wow, what a joke. Stream-ripping! Recording audio in real time is such a joke. Just like movie studios who go after people using camcorders in movie theaters. The people buying those movies wouldn't have paid anyway. The people stream-ripping aren't paying anyway. The people buying from the stream-rippers aren't paying full price anyway. 

 

Get over it people.

 

It is a joke to us but if one guy stream-rips a popular album and shares it illegally and 1,000,000 people (or more) around the world download it free, then I guess it's not a joke to labels, artists and the RIAA?

 

I guess that's where he's coming from?

 

As you know, I would prefer MQA would go away, but I still try to look at things objectively.

 

The CEO of RIAA wouldn't be talking about these concerns if these weren't also the concerns of the labels... I'm sure they've had lots of private chats. I assume he's just echo'ing the concerns of the major labels...

 

If they perceive it to be a concern, then it doesn't matter what we think?

[botrytis]

26 minutes ago, Lee Scoggins said:

My View on Chris’ Computer Audiophile MQA Seminar at RMAF 2018

 

Here are some notes on the slides in Chris’ presentation that I found to show bias against MQA.  I will comment on select slides using the page numbers embedded in the pdf file.

 

Slide 6:

The “from-to” slide sets up early an anti-MQA theme even before we can discuss the detail on each point.  Then Chris proceeds to present evidence on each and other points that does not include an MQA counterpoint.  This is my biggest beef with the slide deck.  Many quality points and counter-evidence in favor of MQA are ignored in favor of the anti-MQA argument in my opinion.  I will address the most notable points slide by slide.

 

Slide 12:

MQA has presented MRI research that shows humans can perceive time phenomena in incredibly small amounts.  This is one reason that I believe MQA focused on the time smearing aspects so keenly because that will provide the biggest benefit.  My understanding is that this research was sent to Chris and Bob Harley and John Atkinson at the roughly the same time.  So it seems curious that Chris does not mention this research on timing.

 

Slide 14:

Chris mentions that it is not lossless from a standard PCM standpoint but Bob Stuart has admitted that publicly.  But here Chris is ignoring the MQA approach cleverness (imho) and the playback aspects of the file.  The MQA process uses triangular encoding so the bits that are discarded are in an inaudible region.  So the point that those bits don’t represent the music is missing from the slide.  In terms of capturing that music in the “triangle”, the process preserves that bit for bit.   The discarded information is essentially dead space.

 

Further reading:

https://www.soundonsound.com/techniques/mqa-time-domain-accuracy-digital-audio-quality

 

Slide 17:

This studio sound slide implies MQA are remixing the master.  In fact, the differences between the master file and MQA are very small.  MQA is applying filters to correct the time smear of ADC.  So there is a misunderstanding here as well about how studios operate.  Work is done on a mixing “console” and that console outputs (at the end of it all) an analog signal.  That analog signal is then converted to digital using and ADC.  All ADCs introduce time smear.  What happens is that MQA filters correct that time smear so that final analog signal is reproduced better in digital.  It’s a different algorithm than the apodizing filters Stuart has creat in the past but addresses some of the same sound phenomena.

 

If you have different tracks or stems that use different ADCs then they can also be corrected by the MQA process  before being mixed together but that’s extra work and not necessary arguably if you can correct the final analog product coming out of the console.

 

Slide 18:

On the Bruno Mars track, Chris is claiming that MQA is putting a “notch” in this track but MQA doesn’t add a notch in its process in this fashion so it’s unclear what happened in this mastering.  This appears to not be a product of the process so it’s up to the chart creator to explain the process and origin of files.

 

Slide 20:

Regarding MQA taking over all other formats.  Any new audio format wants to be accepted widely so the inventors can make money and pay back their investment.  However, this is not any different from PCM or AAC or other innovations.  Why would this MQA ambition be viewed negatively when every format aspires to the same thing?

 

Slide 23:

This again betrays an understanding of the studio process where a “final” analog signal gets converted at the end (discussed above) for distribution.  The MQA filters eliminate the time smear of the final ADC for the finished product.  MQA has now done this so many times that they were able to build a machine learning algorithm to quickly find the ADC signature and then apply the correct filter to eliminate the time smear.   Now on special recordings, MQA uses a so-called “white glove” process to do a fully custom approach.  The Beatles’ Sgt. Pepper was white glove.

 

Remember, ADC is done off the mixing board.  Every studio you visit, they come out analog to drive effects.  Manley reverb, etc,. then on to digital and that A to D is what is done.   It’s the final ADC unless real-time, wherever ADC is done.  Stems are can be like each of the 48 channels, mix, fade, EQ.  Again, you could MQA each stem if you wanted.  So the concern from multiple DACs being used in the process is unwarranted as the artist hears that final analog signal.

 

Slides 25-27:

The implication is that MQA does not create size compression that is of value.  However, we have all seen situations at home where bandwidth becomes a problem.  I have fast U-Verse fiber and when my wife and I are on the laptops, watching a 4K movie sometimes stutters.  If you look at streaming services like Qobuz at 24/192, you need 9.2 mbps speeds which you often don’t find when traveling on trips or sitting at home.  I have noticed Qobuz, which I otherwise like, sometime stutters or “hangs” in my home office.  MQA, due to its compression, only needs 2.4 mbps to work so a quarter of the bandwidth.

 

MQA compression is also useful when you want to offload the music to a phone.  A 60-minute 24/192 file needs 4 gigabytes versus around 1 gigabyte in MQA form.  There is a limit to the amount of phone storage you can have and so far none have things on the order of terabytes.  So the advantage here is that you are taking up 1/4 of the space with the MQA file.

 

Slide 29:

This slide shows bias in that it doesn’t mention a critical fact: no DRM files have been issued by MQA or their partners.  It also defines DRM to include even MQA’s desire to create an authentication process, not for nefarious reasons of taking away customer freedom but to make sure if the file is certified MQA then the end consumer is assured they are getting the genuine article.  I think there are two reasons for this:

 

1. (audit) It creates a quality check on the overall file integrity.
2. (marketing) It creates a feature on the hardware that tells the consumer that the file is full MQA authenticated via a light and the thinking is that consumers could desire that light to know they were listening to a better sound experience.

 

It seems that many here conflate authentication with DRM but they are two different things.  And this was the point I brought up in the seminar, if no one is issuing DRM on the music then why do we care?  The consumer is protected from DRM.  Files can be copied and shared.

 

Even when we look at outside evaluation, we see a conclusion there is no DRM.  See the Electronic Frontiers Foundation. And as we saw today, the ASA team studied MQA and did not find MQA had many any exaggerated claims.

 

https://www.eff.org/deeplinks/2018/12/closed-proprietary-felonious-toxic-rainbow-locked-technology

 

Slide 30:

To me this slide ignores economic reality.  Each participant must make money to be interested in investing time and money, whether to release new hirez files which require retrieving masters and running the process or redesigning chip implementation at hardware manufacturers.  As my friend JB says, “people gots to be paid.”

 

Slide 33:

“We no longer matter.”  I think this is true on one level as major corporation will decide the success level of MQA in the near term.  But I also believe that, in the long run, MQA will succeed mostly on how the consumer acts in terms of demand.  Will they pay for a premium tier like Tidal Masters to get MQA quality?  Will they reward device makers by buying MQA-embedded product?  Will they pay for another major streaming service to offer better sound quality?  I don’t have the answers to these questions but it’s got to be perceived by the customer to be of enough value to purchase or participate in.  If that falls flat then record label, streamer, and hardware device manufacturer interest will wane and MQA will fade out.

 

Other general comments:

There is also no mention of the success MQA has had in getting all the major record labels on board, the dozens of hardware manufacturers now including MQA in the process, and success in the mobile and automotive worlds.  If Chris is going to present the financial losses the firm has had as a startup then it’s only fair to objectively present their successes too.  We also don’t have a current set of financial statements so it’s impossible to judge what progress they have had over the past several months.

 

My conclusion is that Chris did not give MQA a chance to present their side of each point in the argument and, more importantly, did not present the MQA counterpoints we have been discussing all year or even include some of the relevant studies that were sent to him and other journalists.  So any person perusing these slides is only seeing half the information needed to make an informed judgment.  Of course, one should also partake in arguably the most essential event of all: that of a good MQA demo so one’s own ears can decide.

 

Point one - Chris DID NOT HAVE TO GIVE MQA their time. WHY? It was not a discussion panel. Every discussion panel that was setup for discussion of MQA was stopped due to the fact that NO MQA PERSON WOULD SHOW UP. This happened at AXPONA and RMAF, the previous year. 

 

Point Two - this was Chris' take on MQA, period. As an independent voice, Chris has every right to talk about MQA the way he sees fit.

 

Slide 6 is a setup of what the talk will be about.  THERE IS NO SLIGHT TO MQA. What did you expect. Your bias has already shown it's head, Lee, therefore everything else after that is biased nonsense.

 

As I said before, please stop. MQA IS NOT GOOD FOR CONSUMERS OR AUDIOPHILES. You like it great, but it is not the latest thing since sliced bread. it is just another way to repackage and resell the same music over and over.

[KeenObserver]

All the effort you have gone through because you heard MQA and non MQA on your system.

[The Computer Audiophile]

Just now, Em2016 said:

 

It is a joke to us but if one guy stream-rips a popular album and 1,000,000 people (or more) around the world download it free, then I guess it's not a joke to labels, artists and the RIAA?

 

I guess that's where he's coming from?

 

As you know, I would prefer MQA would go away, but I still try to look at things objectively.

 

The CEO of RIAA wouldn't be talking about these concerns if these weren't also the concerns of the labels... I'm sure they've had lots of private chats.

 

If they perceive it to be a concern, then it doesn't matter what we think?

What we or the consumer thinks matters not to the labels.

 

The person purchasing the stream-ripped files from a Russian site was never going to pay for them from a legit source anyway. The person using a torrent client to get the stream-ripped files wasn't going to pay either. 

 

The labels are public companies and need reasons in their back pockets for when they don't hit quarterly numbers. Stream-ripping is one of their Aces under the table for such occasion. 

[botrytis]

The sad part is, most music streaming sites, play watermarked music files. So recording them would just be annoying as sin, since the watermark would be there also.

[The Computer Audiophile]

26 minutes ago, Lee Scoggins said:

 

Maybe so, but you could say that about any format.  I made no claim on industry profitability.  But keep in mind, this can change rapidly and most of these labels are carrying a heavy debt load, some of which is the result of several buyouts.

Labels carrying heavy debt loads always make the right decisions for consumers correct?

[Lee Scoggins]

8 minutes ago, botrytis said:

 

Point one - Chris DID NOT HAVE TO GIVE MQA their time. WHY? It was not a discussion panel. Every discussion panel that was setup for discussion of MQA was stopped due to the fact that NO MQA PERSON WOULD SHOW UP. This happened at AXPONA and RMAF, the previous year. 

 

Point Two - this was Chris' take on MQA, period. As n independent voice, Chris has every right to talk about MQA the way he sees fit.

 

Slide 6 is a setup of what the talk will be about.  THERE IS NO SLIGHT TO MQA. What id you expect. Your bias has already shown it's head, Lee, therefore everything else after that is biased nonsense.

 

As I said before, pleaser stop. MQA IS NOT GOOD FOR CONSUMERS OR AUDIOPHILES. You like it great, but it is not the latest thing since sliced bread. it is just another way to repackage and resell the same music over and over.

 

This is incorrect.  Chris purported to take an objective look but the points I lay out show bias on almost every slide.  He did at least owe the audience a counterpoint to each argument.

 

Also, MQA is not a repackage of prior work.  It is a new series of algorithms and practices that aim to transmit hirez music in a better way and in smaller file sizes.

[crenca]

2 minutes ago, The Computer Audiophile said:

...

The labels are public companies and need reasons in their back pockets for when they don't hit quarterly numbers. Stream-ripping is one of their Aces under the table for such occasion. 

 

This.  

 

The whole "piracy" angle (i.e. it's a real problem that is fixed with the market acceptance of DRM) has made no sense for at least a decade.  Why is it always mentioned by these executives?  It's reasonable sounding (even if untrue) cover for the investors...

[Lee Scoggins]

1 minute ago, The Computer Audiophile said:

Labels carrying heavy debt loads always make the right decisions for consumers correct?

 

Strawman.  No claim made by me as to the friendliness of the labels to the consumer.

[crenca]

Can we put a limit on Lee's manic spamming of this thread?  Maybe limit him to something a 100 posts per day?

[JoeWhip]

This thread is the gift that just keeps on giving.

[asdf1000]

1 minute ago, The Computer Audiophile said:

The person purchasing the stream-ripped files from a Russian site was never going to pay for them from a legit source anyway. The person using a torrent client to get the stream-ripped files wasn't going to pay either. 

 

I disagree. This may apply moreso to the early Napster days perhaps, where it was apparently so easy for everyone to get stuff free?

 

I wouldn't be surprised if some of the big growth in streaming MAY be due to those people saying 'can't be bothered anymore' with going to the illegal sites anymore. So they stopped getting stuff free and just took up paid streaming...

 

Also young kids getting stuff free eventually turn into working adults and MAY pay later at some point (also the same 'can't be bothered' factor with sourcing illegal stuff, when you're a busier adult).

 

I don't think it's easy to generalise so easy?

 

But anyway, if the labels perceive it to still be a concern (RIAA CEO's words above, not mine and I assume he's echo'ing their thoughts) then obviously they are still looking at ways  to further reduce illegal downloads. 

 

 

[The Computer Audiophile]

1 minute ago, Lee Scoggins said:

 

Strawman.  No claim made by me as to the friendliness of the labels to the consumer.

But you continue to claim the friendliness of labels to their artists. Just replace consumer in my comment with artist and it's identical. 

 

You claim labels will pay more to artists because of MQA. Never going to happen.

[Sonicularity]

It's like trying to have a discussion with a television commercial.

[botrytis]

1 minute ago, Lee Scoggins said:

 

Strawman.  No claim made by me as to the friendliness of the labels to the consumer.

 

It is not a strawman argument. I mean, the same can be said about the labels to the artists. The labels seem to care more about themselves than they do the artists.

 

Your dismissal is condescension pure and simple.

[Lee Scoggins]

Just now, The Computer Audiophile said:

But you continue to claim the friendliness of labels to their artists. Just replace consumer in my comment with artist and it's identical. 

 

You claim labels will pay more to artists because of MQA. Never going to happen.

 

That's the artist, not the consumer.  I think if they get a new revenue stream, they will share some of that with the artist.  I've talked to a few enlightened label execs and they realize the current streaming payout is untenable.

[The Computer Audiophile]

1 minute ago, Em2016 said:

I wouldn't be surprised if some of the big growth in streaming MAY be due to those people saying 'can't be bothered anymore' with going to the illegal sites anymore. So they stopped getting stuff free and just took up paid streaming...

Exactly. The only people still going through the hassle of sketchy download sites are those who aren't going to pay. 

 

Anyway, good side discussion :~)

[The Computer Audiophile]

1 minute ago, Sonicularity said:

It's like trying to have a discussion with a television commercial.

I just fell off my chair laughing. Comment of the month!

[The Computer Audiophile]

1 minute ago, Lee Scoggins said:

 

That's the artist, not the consumer.  I think if they get a new revenue stream, they will share some of that with the artist.  I've talked to a few enlightened label execs and they realize the current streaming payout is untenable.

Comical Lee. 

[asdf1000]

Just now, The Computer Audiophile said:

The only people still going through the hassle of sketchy download sites are those who aren't going to pay. 

 

Again, that 15 year old illegally downloading may not be paying today... but in 5 years , when they start earning and can no longer be bothered with the effort to source stuff illegally for free, maybe they do end up paying...

 

Don't think you can generalise the way you have...

 

PS: I would prefer MQA go away....

[botrytis]

1 minute ago, Em2016 said:

 

I disagree. This may apply moreso to the early Napster days perhaps, where it was apparently so easy for everyone to get stuff free?

 

I wouldn't be surprised if some of the big growth in streaming MAY be due to those people saying 'can't be bothered anymore' with going to the illegal sites anymore. So they stopped getting stuff free and just took up paid streaming...

 

Also young kids getting stuff free eventually turn into working adults and MAY pay later at some point (also the same 'can't be bothered' factor with sourcing illegal stuff, when you're a busier adult).

 

I don't think it's easy to generalise so easy?

 

But anyway, if the labels perceive it to still be a concern (RIAA CEO's words above, not mine and I assume he's echo'ing their thoughts) then obviously they are still looking at ways  to further reduce illegal downloads. 

 

 

 

If you want, there are always ways around any system. People, who I know stream, just want music. They don't want to have libraries of music. Many of them have already sold the physical copies of the music they owned. The RIAA tried to stop that also (the sale of used physical copies of music) and that went nowhere.

 

[botrytis]

7 minutes ago, Lee Scoggins said:

 

This is incorrect.  Chris purported to take an objective look but the points I lay out show bias on almost every slide.  He did at least owe the audience a counterpoint to each argument.

 

Also, MQA is not a repackage of prior work.  It is a new series of algorithms and practices that aim to transmit hirez music in a better way and in smaller file sizes.

 

It was an objective look. He was pretty laid back about the whole thing. If it wasn't an objective look, he would have presented, just like the MQA shills did, when they interrupted them. What you want is a subjective look.

[The Computer Audiophile]

8 minutes ago, Lee Scoggins said:

 

This is incorrect.  Chris purported to take an objective look but the points I lay out show bias on almost every slide.  He did at least owe the audience a counterpoint to each argument.

 

Also, MQA is not a repackage of prior work.  It is a new series of algorithms and practices that aim to transmit hirez music in a better way and in smaller file sizes.

There is no truthful counterpoint. I looked long and hard. I did the research. I don't present alternative facts. 

 

 

[firedog]

11 minutes ago, Lee Scoggins said:

 

That's the artist, not the consumer.  I think if they get a new revenue stream, they will share some of that with the artist.  I've talked to a few enlightened label execs and they realize the current streaming payout is untenable.

 

Really? Please give us any data in the past behavior or record labels that would make us think this is a believeable scenario? They are the ones who helped establish the whole low level of royalty payments to artists in for non physical music media.

 

And you are still parroting the MQA falsehoods about how MQA fixes the time smear of ADCs. It simply isn’t true and since we now know how the “renderer” works, we know that it’s one of the MQA lies. 

[KeenObserver]

WOW!

 

This thread may reach a million views just on the refutation of Lee's ridiculous diatribe!

This whole MQA thing is getting tiresome.  But the fact is,  MQA tried slipping in without the public knowing what it was really about.  Shining a bright light  on MQA is doing the music consumer a great service.  I do not want to be screwed by MQA.  And that is what will happen to the music consumer if MQA gains wide acceptance.