MQA is Vaporware

[KeenObserver]

6 minutes ago, John Dyson said:

I agree -- what can we do about it?   1)  cannot legally redistribute 'good' versions that are seldom available  2) no matter how people complain, the answer is to 'remaster', which doesn't fix the problem.

 

I can reverse the evil compression even on many remasters!!!   That is *bad*.  There shouldn't be an ability to correct a problem -- because it shouldn't be there to begin with.

 

The "MQA"  kind of thing has, in a distant way, been with us since CDs came out.   The manifestation was that 'digital sound' back in the '80s that was misattributed to digital technology.   Just chatted with a friend who was also alive during the time of CDs coming out, (an AT&T Bell Labs co-worker), and he quit buying CDs in disgust also.   People like me aren't alone -- I knew what recordings sounded really sound like, and I had to finally reject almost anything that I purchased.    There wasn't enough people like me and my friend Harold -- more people should have simply quit buying the CDs.    People generally don't hear 'sound quality', but they have a 'listening experience' ...   As long as the 'sound quality' isn't too terrible that it can be adapted to/accomodated, then it will only minimally interfere with the 'listening experience'.

 

This is why the industry has gotten by without needing to fully expose their intellectual property, still making the majority of the population adequately satisified.   The industry has gotten the best of both worlds.

 

John

 

 

There were some of the smaller houses that were putting out outstanding CD's.  The problem is that the larger studios could put out garbage and tell you: "take it or leave it".  They could always dangle that carrot of something better down the road, in front of the music consumer. I always picture the bigwigs at the majors going: "WAIT! WHAT!" when the advertising for CD's said it was perfect sound forever.

Warner is in a position of having 700 million to a billion dollars in cash reserves.  This puts them in a better position than some of the other studios to weather out the theater closing effects.  It may also put them in a position where they think they can tell the music consumer that they are implementing MQA, take it or leave it.

The only thing that would get their attention is a worldwide "boycott Warner" movement, like the "boycott Sony" movement after the Sony rootkit fiasco.

[John Dyson]

5 hours ago, KeenObserver said:

 

There were some of the smaller houses that were putting out outstanding CD's.  The problem is that the larger studios could put out garbage and tell you: "take it or leave it".  They could always dangle that carrot of something better down the road, in front of the music consumer. I always picture the bigwigs at the majors going: "WAIT! WHAT!" when the advertising for CD's said it was perfect sound forever.

Warner is in a position of having 700 million to a billion dollars in cash reserves.  This puts them in a better position than some of the other studios to weather out the theater closing effects.  It may also put them in a position where they think they can tell the music consumer that they are implementing MQA, take it or leave it.

The only thing that would get their attention is a worldwide "boycott Warner" movement, like the "boycott Sony" movement after the Sony rootkit fiasco.

Message not primarily about my project, but about some characteristics of the 'audiophile' consumer -- 'audiophile' does not necessarily mean 'audio quality' phile though.

 

This is all the main reason why I started the FA project.   Over the years, some people have gotten frustrated when working with me, and some people simply like the compressed sound - but  won't admit it to themselves, but it is now possible to produce near perfect results if one is patient enough.   The problem is:  Not everyone has the skill and patience to use command line Windows on a program that runs betwen 3X realtime to slower than realtime.   The DA version (a small component of FA) has even been requested by a major studio (not just a small studio -- one of the biggest) -- the quality of the admittedly tedious to use (but not really difficult) expander/decoder is excellent.

 

So, if one is REALLY motivated, and REALLY DOES want near studio sound, it IS possible to improve the quality *very noticeably* on probably 2/3 of the POP garbage and alot of the damaged classical also.   The improvement isn't 100% -- because frankly, many recordings aren't all that great to begin with -- but can be better than the CDs and downloads really are.  There is always more 'interest' in better quality than actual motivation to do something about it.   It seems like there is a real market for 'spending money' on something that partially corrects the 5% that the consumer can EASILY fix, but little interest even something FREE that corrects 2/3 of the 95% that is intentionally damaged by the distribution & industry.  (That is, consumers will spend $10K on hardware to help mitigate problems with the 5% of quality problems, than spend TIME to mitigate perhaps 2/3 of the 95% problem with recording quality.)

 

This is all why Warner and other garbage merchants get by with what they are doing -- they find that people are *interested* in better quality, but won't do anything to correct it.   Not only that, the garbage quality helps to motivate the 5% improvement market, where people are willing to SPEND MONEY instead of SPEND EFFORT.

 

There is so much lip-service from a big part of the 'elite' consumers who claim they want better quality, but instead there is more  of a hoarding instinct.  It is so sad to see people  spending money in passive desperation.   Fortunately or unfortunately -- many years ago, I detected the sucky quality, and couldn't let myself spend much money in order to attempt perfection.   That is, until I found the technical quality problems.

 

Is is better to have given up on the hobby in the late 1980s like I did, or to enjoy -- but hear the noxious compression while spending many $ks trying to fix the problem in futility?   I ended up passively collecting some recordings, but I was honest with myself that I was buying garbage, and a system that cost more than a few $k would be polishing the proverbial excrement.

 

MQA is yet another way to further control IP and somehow marginally to extract money.   It has NOTHING to do with quality -- they know that most consumers will put up with the abuse, unless they really make a big stink.   The consumers didn't make a 'big stink' back in the 1980s, and now we have garbage that it took 30yrs to figure out how to fix.

 

I hope that the publicity from this group will be enough push-back -- but eventually, the bait&switch merchants will win.  (Here is your nice, better recording -- Garbage merchant:  oh, I didn' tmean it, sorry, the consumer already spent their money - and I know that they'll spend more, bye!!!)

 

 

John

 

[asdf1000]

13 minutes ago, John Dyson said:

 

This is all the main reason why I started the FA project.  

 

Hi John, can you share a brief description of what your FA software does?

[John Dyson]

33 minutes ago, asdf1000 said:

 

Hi John, can you share a brief description of what your FA software does?

*The FA decoder mostly successfully undoes the ubiquitous dynamic range compression on most commercial consumer digital recordings.

 

*What motivated it:   Disgust about the quality of consumer recordings (ca. 1990), and random notice about consistency in the characteristics (ca. 2011/2012)

 

*My qualifications:  I am a forward looking EE/CompSci/DSP/Systems engineer who got some of the most obnoxiously difficult projects at Bell Labs.   If it was a complex multi-disciplinary project, I was either a developer on the project or systems engineer.   

 

*My process: I noticed a very 'regular' kind of compression being used, and took a long random walk to try to see what could be done to fix it.

 

*I found the type of compression (which has been very common in pro-circles since 1966), but it took a very long, tedious reverse engineering effort to determine the general design of the compressor complex.   The compression doesn't come from just one compressor, but a network of multi-band compressors with EQ.

 

*This form/shape of compression is not undone by normal dynamic range expansion.  (A simple expander in a DAW will do VERY LITTLE to actually correct the recordings.)

 

 

 

* SO, how often have you heard a normal, commodity recording with normal sounding instruments?  Not very often, perhaps with hiss also?  Here are two snippets from cleaned up copies (using the FA decoder) of a *commodity* very compressed recording from London Symphony (drop box player is very poor -- so maybe a download would be good for the reveal).   The piano on the 'snippet1'  might be a good demo -- I use pianos for one of my own indicators because I know what they sound like.

 

https://www.dropbox.com/s/qtunbpnfc9tku0y/01 - Rhapsody in Blue-snippet.flac?dl=0

https://www.dropbox.com/s/cyj8kd8h89t15v4/01 - Rhapsody in Blue-snippet1.flac?dl=0

 

 

*What is the FA decoder?  It is made up of layers of VERY ACCURATE DolbyA decoders (better than true DolbyA units, or it wouldn't work very well.)   They are layered in 10dB steps, and because of the shape of the compression curve, a normal expander mostly just chokes on more than 10dB of expansion.   However, as you can tell by the hiss amplification on consumer materials, there is much more than 10dB of compression on consumer recordings.   Making actual and conforming inroads into cleaning up consumer recordings hasn't been done, until now.   The very accurate DolbyA decoder software is being used by pro recording studios -- it really IS better than a true DolbyA for decoding purposes.

 

* The program can be called a 'decoder' or a very special purpose 'expander', but it has been a painfully tedious project that has been mostly successful at undoing the compression that damages MOST commodity consumer recordings.

 

* In fact, the hiss reduction by the project is nearly perfect, and there is actual correction for the dynamics.   There is no other way that I know of, even FFT NR techniques that are as effective at NR on the distorted materials.

 

There is a 'FeralA' message list on Audiophile Style.

 

The decoder is essentially finished, and works as expected.   The bad news -- it does require patience to use, is Windows/Linux command line and because using the decoder is a form of mastering, it is hard to make everyone happy with your personal results.   It is best to use the decoder to correct the recordings in your collection for yourself -- you know how the recordings should sound for you.

 

Peoples hearing has been distorted by the ubqiutious compression for so very long -- that it is best just to help fix ones own personal collection.

 

John

 

 

[asdf1000]

5 minutes ago, John Dyson said:

The decoder is essentially finished, and works as expected.   The bad news -- it does require patience to use, is Windows/Linux command line and because using the decoder is a form of mastering, it is hard to make everyone happy with your personal results.   It is best to use the decoder to correct the recordings in your collection for yourself -- you know how the recordings should sound for you.

 

Thanks! Can you point me to where I can find the command line instructions, so I can give it a whirl ?

[John Dyson]

1 minute ago, asdf1000 said:

 

Thanks! Can you point me to where I can find the command line instructions, so I can give it a whirl ?

We should take this to the 'FeralA' group -- people get irritated with off topic discussions :-).

 

Anyway, there is a Usage doc in the package, and there is a big manual that describes the install.   The install is very primitive, originally intended as a toy, but then  pros got interested...   It needs .wav file I/O, and works with 16bit,24bit,FP in,  24bit,FP out,   44.1k though 96k, 176k 192k, 352k, 384k  (it adapts to sample rates, so need not be dead-on.)    There is a real-time play option (--play or --splay -- needs the 'aoplay' program in the same place as the da-avx program, and it is included.)

 

Do the install as described in the big manual (the Dyson Hess manual), then -- to decode classical materials is usually simple like this:

 

da-avx --infile=in.wav --overwrite --outfile=out.wav  --info=2 --tone=-50  --fcs="6,auto,fce" --fw=classical --pi=1k

(Possibly --tone=-60 or --tone=-54.5 also -- you'll hear compressor effects if the setting is wrong.)

 

For POP, it is a little more tricky.   You use essentially the same kind of command line...   New pop recordings are pretty closely decoded by:

da-avx --infile=in.wav --overwrite --outfile=out.wav  --info=2 --tone=-50  --fcs="6,auto,fcf"  --pi=v3 --pi=hf3 --pi=1k

 

The --v3 or --hf3 might be needed to be changed, e.g. --v0 or --v5.   --hfN (N=0 to 6) is the same.   Some recordings, esp old ones, might need --pi=*v3 instead.   Also, some might need --pi=-hf3 instead.    Basically, the various punctuation marks (perhaps needed 10-20% of the time) are differences in EQ.   They are explained *inadequately* in the usage guide.

 

I have been so busy on writing/debugging/testing the decoder while also learning how to use it, that the docs have fallen behind.

It is now much easier to use than the previous crazy myriad of options that used to be needed.  I added a lot of internal automation and and adaptation.

 

It does appear that  the *compression* is very consistent, but NOT the EQ.   All of these variables are forced by NEEDED EQ.   The EQ specified in the command line is NOT ad-hoc, but is a reversal of the FA encoding process.   Here is the pointer to the small repository, but lets move to the FeralA messages (or PM me) so that not-interested people don't get irritated :-).

 

https://www.dropbox.com/sh/1srzzih0qoi1k4l/AAAMNIQ47AzBe1TubxJutJADa?dl=0

 

PM me if you need help -- esp until I update the docs!!!!

 

John

 

[UkPhil]

After some analysis of a lot of these Warner releases on Tidal  looks like they are conversions of 16/44.1 copies, so not even hi res par say, so what’s the point nothing to preserve above 20khz regardless if we can hear it or not and no saving on space, plus most of them are not authenticated as a studio source. So they are left with their “deblurring” and proprietary filtering which I am sure can be created by personal eq, readily  available filtering or room treatments. 

[bambadoo]

And people pay

[The Computer Audiophile]

On 11/15/2020 at 7:59 AM, asdf1000 said:

 

Hi @The Computer Audiophile

 

Where you able to test these albums on Amazon Music HD?

 

 

2L recordings :

 

Hoff: Innocence
Hoff Ensemble

 

Arnesen: MAGNIFICAT 4. Et misericordia
Nidarosdomens jentekor & TrondheimSolistene

 

These albums are MQA from Amazon Music HD.

 

[dmackta]

Just now, The Computer Audiophile said:

These albums are MQA from Amazon Music HD.

 

 

[KeenObserver]

What label?  Are they posted as being MQA?

[UkPhil]

14 minutes ago, KeenObserver said:

What label?  Are they posted as being MQA?

2L are known for supplying MQA versions to other streaming platforms by default, they have been backers of the format since its inception 

[The Computer Audiophile]

5 minutes ago, UkPhil said:

2L are known for supplying MQA versions to other streaming platforms by default, they have been backers of the format since its inception 

Yes. It's a great selling point for the MQA team. "We have people delivering a single file for all platforms, just like we promised." It's exactly what MQA did by starting with HiFi companies to stack their deck before bringing their pitch to larger companies. "We have the most discerning audiophile companies onboard." 

 

 

[UkPhil]

13 minutes ago, The Computer Audiophile said:

Yes. It's a great selling point for the MQA team. "We have people delivering a single file for all platforms, just like we promised." It's exactly what MQA did by starting with HiFi companies to stack their deck before bringing their pitch to larger companies. "We have the most discerning audiophile companies onboard." 

 

 

Yes a sonically crippled file with no keys to get in unless you pay the organ grinder 

[asdf1000]

5 hours ago, The Computer Audiophile said:

These albums are MQA from Amazon Music HD.

 

 

Same issue as Qobuz have.

 

2L recordings

[sp25]

I'm fairly new to the "party" round mqa and the loudness war in general. My misconceptions have been a) that the latest remaster is the best b) that louder is generally better c) that streaming music could replace my own collection d) that its worth investing in mqa hardware. What a dogs dinner that our beloved music, which to us are works of art, are generally being treated as product to re-sell over and over and with no consideration to hi-fi outside of marketing. If the original work is being compromised I see little value in spending thousands trying to upgrade my hi-fi system, its a dead end. There's still great music to listen to fortunately, just have to make the best of a bad job and try and spread the word about what hi-fi is!

[John Dyson]

4 hours ago, sp25 said:

I'm fairly new to the "party" round mqa and the loudness war in general. My misconceptions have been a) that the latest remaster is the best b) that louder is generally better c) that streaming music could replace my own collection d) that its worth investing in mqa hardware. What a dogs dinner that our beloved music, which to us are works of art, are generally being treated as product to re-sell over and over and with no consideration to hi-fi outside of marketing. If the original work is being compromised I see little value in spending thousands trying to upgrade my hi-fi system, its a dead end. There's still great music to listen to fortunately, just have to make the best of a bad job and try and spread the word about what hi-fi is!

Back in the 1980s, I loved my HiFi hobby (it was both high fidelity and audiophile.)   Even though I am not a vinyl-phile, truly the audio quality started substantively diminishing when CDs came out.   It was not the fault of digital, but that misconception against 'digital' helped to perpetuate the selling of inferior quality recordings in digital forms.  For example, I used to have a totally beautiful 'I have the music in me' vinyl from Sheffield, when it was released!!   It was niiiiceeee!!!   It was exactly the Sheffield CD version back in the 80's/early 90's that was the very last straw for my hobby, CDs would never be produced correctly.   I gave up.   I typically quote approx 1990 as the time that I quit, but the final blow as that CD.

 

Recently, had a correspondent sent me a digital copy of the recording from Thelma Houston (no relation to the Dionne W, Whitney family) -- and I was TOTALLY disgusted (this is with my even more discerning hearing.)   Speaking of Dog's Breakfast -- that recording was a total mess, yet touted as coming from a premium company.   The vinyl was beautiful -- it is all about damaging processing in the signal chain.   (CDs can be made to be perfect, so can digital downloads...   Someone is cr*pping on the recordings.)

 

The knowing pushback that couldn't happen back in the middle 1980s needs to happen NOW.   We didn't know what they were doing back in the 1980s, but we DO know now.  This should be a component of the MQA pushback -- tell the distributors that we will NOT tolerate any more garbage being sold as recordings.

 

John

 

[sp25]

30 minutes ago, John Dyson said:

tell the distributors that we will NOT tolerate any more garbage being sold as recordings.

 

The push really ought to come from the artists but they'd mostly sold out by 1975! As Frank Zappa humorously played "We're only in it for the money"! Seriously its also a matter of education and being able to hear how better music sounds done well. The irony of the revival in vinyl sales is that again its been driven by marketing as a cool thing to do and then its played on a cheap turntable with rubbish gear and the real value of vinyl missed.....

I should add to my list of misconceptions earlier e) that Hi-Res recordings would be the best available. Bit of a rude awakening after investing time and money in project Hi-Res. The good ones remain good just no guarantee....

[yahooboy]

Well we are all wrong

MQA IS LOSSLESS (Well at least to MQA's marketing)

We are just stupid not to accept Bob's "facts

[The Computer Audiophile]

2 minutes ago, yahooboy said:

Well we are all wrong

MQA IS LOSSLESS (Well at least to MQA's marketing)

We are just stupid not to accept Bob's "facts

Wow, alternative facts. 

[KeenObserver]

6 hours ago, sp25 said:

I'm fairly new to the "party" round mqa and the loudness war in general. My misconceptions have been a) that the latest remaster is the best b) that louder is generally better c) that streaming music could replace my own collection d) that its worth investing in mqa hardware. What a dogs dinner that our beloved music, which to us are works of art, are generally being treated as product to re-sell over and over and with no consideration to hi-fi outside of marketing. If the original work is being compromised I see little value in spending thousands trying to upgrade my hi-fi system, its a dead end. There's still great music to listen to fortunately, just have to make the best of a bad job and try and spread the word about what hi-fi is!

 

There are a lot of very good recordings out there, you just have to search them out.

 

I think John does a disservice when he implies that all CD's are terrible.  The CD platform is capable of producing outstanding results.  There are people that understand this and produce outstanding CD's.  There are others that keep pushing out compressed, poorly mastered garbage.  And there are people like Warner that want to jam MQA down your throat.

And "Hi Rez" is the same.  Just because something is 24/96, or 24/192, or even 32/384, does not make it a good recording.

People buying discs, downloading music, or streaming music should shun the music that is deliberately crippled and seek out those music providers that produce top quality recordings.  If enough people refused to buy crippled music, perhaps the majors might start supplying quality music.  However, with an entire generation that listen to low level MP3 over junk earbuds, that might not be a possibility.

[yahooboy]

10 minutes ago, The Computer Audiophile said:

Wow, alternative facts. 

Well it's out there in spades, licensees that keeps promoting the third unfold etc.

 

[AudioDoctor]

11 minutes ago, The Computer Audiophile said:

Wow, alternative facts. 

 

When most people don't even care, you can say pretty much anything you want. I guarantee I could talk to most of the people I know about this and their eyes would gloss over before they walked away. Their only concern is does sound that vaguely resembles music come out of whatever when they hit play? Good, done.

[John Dyson]

6 minutes ago, KeenObserver said:

 

There are a lot of very good recordings out there, you just have to search them out.

 

I think John does a disservice when he implies that all CD's are terrible.  The CD platform is capable of producing outstanding results.  There are people that understand this and produce outstanding CD's.  There are others that keep pushing out compressed, poorly mastered garbage.  And there are people like Warner that want to jam MQA down your throat.

And "Hi Rez" is the same.  Just because something is 24/96, or 24/192, or even 32/384, does not make it a good recording.

People buying discs, downloading music, or streaming music should shun the music that is deliberately crippled and seek out those music providers that produce top quality recordings.  If enough people refused to buy crippled music, perhaps the majors might start supplying quality music.  However, with an entire generation that listen to low level MP3 over junk earbuds, that might not be a possibility.

I NEVER wrote that all CDs are terrible, most are terrible.   Few are good, unless boutique.

 

However, I am especially susceptable to excessive compression-hatred, especially the insidious kind used typically on consumer CDs.

Give me a .wav file image of a master tape, and a blank audio CD, then my results will be wonderful.  However, I will not add some nonsense processing to the recording -- just a small amount of EQ, or a small amount of artful compression if the dynamic range is much wider than the listener expects.

 

Artful compression can be good.  (I just ran into a recording where the nasty compression was preferred by the listener, but could probably have been much nicer if not done with the 'one size fits all' botch that is normally used.)  

 

Good, artful compression is much less distasteful than the commonly used compression on BY FAR most CDs.   This includes a recent set of tests on some Telarc stuff.

 

If you find a good quality (I mean, properly done, good quality, master tape playout) CD, then be VERY happy about it.

 

John

 

[KeenObserver]

The FLAC container is a lossless system.  The MQA file inside is lossy.  And corrupts the original music, no matter what Bob says. 

 

I hope we are not seeing a resurgence of the MQA BS.  My hip boots are in storage.