MQA is Vaporware

[jabbr]

11 minutes ago, The Computer Audiophile said:

 

Perception is reality. If it sounds worse to people, then it's worse. If it measures worse, but sounds better, then those who look at music will be mad while those who listen will be happy. 

 

 

 

Reminds me of the "looks bad, feels good" joke  

[jabbr]

On 5/25/2017 at 7:39 AM, Jud said:

Two comments:

 

- I think we still have a way to go in finding out what causes us to hear a reproduction as accurate versus the original.

 

- The commercialization and confidentiality involved with various efforts that we are told aim toward that goal (for example MQA, Schiit Audio's "Manhattan Project") hinder our ability to find out if there's really anything worthwhile in them.

 

Totally agree ... to the point where its actually more important to me that we have *open* and *documented* media storage formats, rather than the "ultimate" in psychoacoustic experience.

 

MQA doesn't teach us (the collective us) anything, and life being short, I'll defer on spending my increasingly precious time, considering it. Published research, open research is the only way to move *toward* the goal of "the absolute sound". 

 

If I want real accuracy I can either pay $$ to listen to a real live concert, or bring good wine over to my friends' places and listen in person ... perhaps to a Strad if I bring really good or really interesting wine  

[jabbr]

12 minutes ago, Jud said:

 

I want to know the wine that can get you your own chamber concert with a Strad.

Ha that's a secret: We ABX the strad & wine together  

[jabbr]

Wiener deconvolution: https://en.wikipedia.org/wiki/Wiener_deconvolution has long been applied to the audio domain (http://www.cnbc.cmu.edu/~samondjm/papers/Recio-Spinosoetal2005.pdf was found offhand as an example) and is widely used as a way to model and correct "error". In image analysis, a 3D model of the error is deconvolved with the 3D image to produce a corrected result. Similarly for 2d or 3d audio domain. The error may be measured from a system impulse response, or estimated. "deblurring" may refer to a sharpening in "linewidth" of fourier series peaks, but since MQA is proprietary who knows. The point I am making is that there are very well known and widely used techniques that are generally called "deblurring". In the audio domain, if one were producing, for example, a sonar map, then such techniques might be employed to sharpen such a map. This is analogous to "soundstage" (as an example).

[jabbr]

3 minutes ago, Jud said:

 

Nothing in the MQA patents, white papers, or histories of the principals to indicate Wiener deconvolution or anything similar is going on, and lots in all of those to indicate it's good old Meridian apodizing filters in a new dress.

 

I'm not surprised. Norbert Wiener's contributions have in many circles been ignored or have not been given the credit they are due. I think you would find his biography interesting. In many ways both Wiener (who at the time did not get the credit due) and Shannon both developed "information theory" independently and from different angles -- though both were at MIT. I am referring to the concept of a filter correcting error -- or eliminating noise see: https://en.wikipedia.org/wiki/Generalized_Wiener_filter -- given a modelled error (or noise). This is where the term "apodization" comes from. We used this technique: https://svi.nl/BeadsDeconvolutionExample in the 1980s. While Meridian may use a type of apodizing filter, they did not invent apodizing filters. I'd say that the *old* Meridian apodizing filters were, even at that time, Wiener filters in what at that time was a new dress

 

My suspicion is that the reason MQA remains "secret" is that the mathematical techniques are actually in very very wide use.

 

[jabbr]

3 hours ago, PeterSt said:

 

Why ?

I strongly believe in open unencumbered data archival formats. Imagine if, for example "Sgt Pepper's" were only available on SACD ... when will the last SACD player be made? What a shame that would be to lose a piece of our cultural heritage (insert any other example).

[jabbr]

22 hours ago, kumakuma said:

The data on a SACD may be in DSD format but the discs themselves are heavily copy protected and we were unable to retrieve or archive the data on these discs until someone figured how to do it on certain models of the Sony Playstation.

Yes 100% and as long as I can rip the ISOs to HD and then convert to DSF with AuD, then I purchase SACD ... I think SACD unfortunately is dying out ... perhaps if it weren't encrypted (like CD isn't) then it would have a greater usage??

 

my absolute bottom line is that I have to archive my music in an open format -- if you disagree then do whatever you want but that's what I do and will do 

22 hours ago, kumakuma said:

This is undoubtedly how the record labels would like to use MQA as it puts the piracy genie back in the bottle and forces us to pay for yet another version of the same music we've already paid for many times in the past.

 

To be clear however Now that I can rip my SACDs first to ISO with a PlayStation and then to DSF with AuD, I purchase lots of SACDs. The ability to rip has thus increased SACD sales -- at least from me

Its the new stuff that seems to come out as digital download.

 

Nowadays artists make their $$$ with live concerts. 

[jabbr]

3 hours ago, AJ Soundfield said:

IOW, absolutely nothing like MQA sound quality claims.

 

Interesting, of course with HQPlayer or XXHighEnd (which I can upsample to 768k on a celeron), the fixed cost is very low, and it works with my entire library... 

[jabbr]

12 hours ago, AJ Soundfield said:

Why is that interesting?

Because I upsample all my PCM before sending to DAC, there should be no SQ difference with MQA according to that graph.

[jabbr]

Ah ok well I'm upsampling regardless  

 

My point is that it's really really cheap to upsample PCM in the computer - I'm using a board that's $70  

[jabbr]

47 minutes ago, Fitzcaraldo215 said:

I am no right wing capitalist, far from it, but I do not believe that a world in which every product was free, or easily duplicated exactly from public information or patents to the molecular level, would be a good one to live in.  There would be scarce incentive for innovation.  It is economics 101.

I'm no Marxist . I've lived through the transitistion from the single source single vendor closed network (ATT) to the open sourced open protocol Internet & Web as we now know it. The rise of the Web as an economic force has been the most astonishing driver of innovation I have witnessed. Despite the formats and protocols being open -- no. because the formats & protocols have been open. Some other folks agree with me btw in case you are wondering  

[jabbr]

Microsoft is the best counter example. Apple put an absolutely terrific GUI on an open source kernel, built great hardware and the rest is history. ... before Jobs came back with that open source OS, Apple was at death's door.

 

I never said the processing or what you do with the data need not be open source or not proprietary. I guess you'd never invest in IoT then ... or self driving cars ... or most anything with an ARM chip because increasingly there's some linux running somewhere inside...

 

So you have an investment in something to do with MQA? Because file formats are a dime a dozen... data compression has become uninteresting ...

 

 

[jabbr]

1 minute ago, Fitzcaraldo215 said:

Note also that CD for a long time was not open.  Players and music discs were made under license agreements with royalties to Sony/Philips.  Not sure about blank discs.  Did any of that stop it?  Of course not.

 

No, I am in no way even remotely affiliated with MQA or any of its licensees or anyone in the audio business.  I do not own it, and I have no immediate plans to.  I have never myself owned a Meridian product, either.  My interest in MQA is purely out of audiophile curiousity and annoyance at ignorance (including my own), cheap shots, unfounded claims, distortions, mob rule, etc. that may obscure the truth about it. 

Ok fair enough. I have a very long interest in open data formats, not audio, dating back decades. I have followed the technology industry very closely. I have been involved in standards committees, invited expert to the W3C, government bodies and other entities that have interests in creating and maintaining large, long term data archives.

 

CD: Players and discs proprietary of course but underlying data is WAV/AIFF. I don't mind paying for software nor hardware, I just don't want my data locked up. I won't pay for that. That's not a cheap shot, unfounded claim, nor distortion. That is my opinion.

[jabbr]

http://s1.downloadmienphi.net/file/downloadfile4/206/1391106.pdf

[jabbr]

3 hours ago, mansr said:

The input to the renderer consists of 23-bit PCM data and a 1-bit metadata stream. The metadata is actually only a few bytes that are repeated continuously.

 

Do you think there is an analogy to DEM or the operation of the AD5791 which uses a 14 bit R2R along with 6 bits of "metadata" (hope that is a good way to describe that). What I am seeing is the possibility of a hybrid PCM/multibit SDM. Is this a reasonable analogy? They aren't the same of course.

[jabbr]

5 hours ago, Charles Hansen said:

... The only difference between the original Berkeley DAC (your long time reference) and your even more beloved "Reference Series" (at nearly 3x the price) was upgraded passive parts quality - things that shouldn't make a difference yet obviously do to a trained listener with a familiar system playing familiar music. Specifically, a chassis machined from solid billet instead of bent sheet metal and a change from "standard" FR-4 PCB material to Rogers 4000 series, a low-loss material designed for GHz range circuits.

...

 

That is great info thanks ... seriously. Details like perhaps alu chassis vs steel (shielding) component selection ... even PCB material ... when you say these "shouldn't" make a difference, I really suspect that's because most folks are looking at the wrong characteristics of what "should" influence sound.

[jabbr]

1 hour ago, The Computer Audiophile said:

There's actually much more that's different between the Alpha and RS. 

 

Ha ha, that's fine. We have such an amazing ability to measure small differences that it really interests me regarding what the limits of human ability to distinguish really are. The suggestion that Rogers 4003 or whatever might sound different than FR4 PCB opens up a lot of questions for me, like : wow, how do we hear that?

 

On the other hand, if really good modelling was done you ought be able to account for the dielectric differences so perhaps its just a more optimized design?

[jabbr]

10 hours ago, Charles Hansen said:

I think what we will find is that our current understanding of the ear/brain hearing mechanism is woefully incomplete. A good example of this can be read in a book called "The Secret of Scent" by Luca Turn. A fascinating read (and possibly a bit of a struggle if you've never studied chemistry, especially organic chemistry (which involves carbon and is the basis for all life). In it he shows that the commonly-accepted "lock and key" model for how the nose works is completely incorrect, even though that has been "conventional wisdom for nearly a century.

 

Instead it turns out that olfactory nerves do not sense the shape and/or charge on molecules - they "listen" to the vibrational frequencies of the molecule's side chains, using tiny spectrometers. These spectrometers are so small that to fit into a human cell that they only work on the principles of quantum mechanics. (QM was only proven to be correct in the mid-'90s and is the really weird physics where solid objects can pass through each other, two particles can be in the same place at one time, and the transmission of information is not bound by the speed of light - it can transfer instantaneously across any distance.)

 

Whoa!

First, a book by a scientist does not equate with a change in the scientific under standing of a principle. A colleague of mine a long time ago was also working on an "artificial nose"  ... I worked on some very early speech and sensory recognition projects (c 1980s -- my very first paying job !) so am perhaps much more familiar with these issues than you might expect.

 

Luca Turn's model made predictions that did not pan out nor to my knowledge did his artificial nose (our defense dept / homeland security has extraordinary interest in this) .

 

olfactory nerves respond to chemical inputs. All chemicals have wave equations and can be thought to vibrate -- there is nothing inside that actually "listens" nor is there an actual miniature spectrometer that isn't just a description of molecular machinery -- the body is composed entirely of molecular machinery and this all has quantum descriptions and all interactions are described by EM/strong & weak nuclear forces (gravity not being too relevant here). Yes that is all biophysics and all solids are not solid at the olecar level ... so what?

 

This isn't "really weird physics" just 20th century physics that every graduate student learns (we'd hope undergrads too if they pay attention)

 

I don't think we need to bring quantum entanglement into this discussion ...

 

Having actually worked in this area, the issues in understanding how the nose or ear work are elsewhere -- we don't understand the layers above the basic senses. How does the cortex process information? 

 

Regarding Fourier uncertainty-- it has been universally accepted since ?1970s that our sensory systems including cochlear are highly nonlinear. The article is yet another piece of information that discusses this ... no big deal.

 

Regardless Maxwell's equations have persisted quite nicely  in the quantum era -- we have great ability to measure down below -160 dB both voltage & phase from 0 to GHz range -- my question is: what are the real limits of human audibility? Perhaps focus on very fine grained phase error which isn't always looked at...

[jabbr]

16 hours ago, Charles Hansen said:

As nature doesn't like to waste good, usable mechanisms, using the same things not only across different species but even across different kingdoms, I will be surprised if it turns out that the ear/brain does not use QM. Here is an article that shows us just the tip of the iceberg on how incomplete our current understanding is: Human Hearing Outsmarts Physical Limits | Evolution News

 

 

The organization of cells is, to a first order approximation, defined by the DNA codon language (this is just an approximation). DNA codes for sensory molecules such as rhodopsin which itself changes state as a result of interaction with a photon. The interaction is clearly a quantum phenomenon and this discovery dates back to the 1930s http://www.ghuth.com/images/waldlecture.pdf ... so in the visual system there are molecules that have selective propensity to interact with photons having certain frequencies (vibrations?) ... this has been studied in great deal and is not new.

 

Of course the auditory system and brain is based on molecular interactions which themselves are defined by quantum mechanical phenomena -- how else could it be? I wouldn't say that nature has a "desire" to employ QM, rather that nature as we currently know it, is fundamentally based on QM. Two "solids" being in the same place at the same time can be best understood as waves that are superimposed -- the equations that define a "solid" have nonzero point probabilities.

 

All of this is highly nonlinear. The rhodopsin molecule either receives/accepts a photon, or doesn't. The nonlinearity of the auditory system was first recognized in the 1970's: http://physicsworld.com/cws/article/news/2013/jan/31/human-hearing-is-highly-nonlinear but agreed, this is not well understood by audiophiles, nor folks who insist that the  20-20kHz limit is absolute.

 

I'd like to point out the www.arxiv.org site which is a tremendous resource for freely available physics research articles: https://arxiv.org/pdf/1208.4611.pdf

[jabbr]

12 minutes ago, Ralf11 said:

Cliff's Notes version:  Quantum Mechanics is not required to understand biology.

 

LMK if you want the Dummies Guide.

 

Absolutely, sorry if I made seem otherwise. Even more importantly for us here, electromagnetism as defined by Maxwell's equations is entirely compatible with the quantum version and so neither is quantum mechanics necessary for an understanding of electronics (under our normal circumstances). Things can be nonlinear without invoking quantum mechanics.

[jabbr]

34 minutes ago, PeterV said:

That's what results in a very convincing audible improvement of these albums ans all others I am listening to dor just 20,- euro per Month. No need to download music anymore and the soundquality is better than ever before.

So if I listen to:

 

 

Does that “ca Ching” get felt in your pocket? I’m not feeling the hi fidelity...

[jabbr]

9 minutes ago, FredericV said:

Tubes break the end-to-end analog promise

That promise wasn’t broken by your tubes

[jabbr]

22 minutes ago, PeterV said:

What is your point jabbr..? Youtube has no MQA encoded albums available (yet) and if they would, you will need to decode it properly. Only then you will hear what it has to offer

I’m just tryin to learn, so would MQA (if it did YouTube) convert:

Into: 

? 

Or vis versa

??

[jabbr]

6 hours ago, PeterV said:

Technical analysis without complete information how the encoding is being done, why the focus of MQA is totally focussed towards the improvement of temporal blur and to deliver a very fine time description is incomplete and inadequate as well.  Emulation is per definition a surrogate and will always be an incomplete and incorrect representation of what MQA is capable of.  But you are right, being happy with MQA or without it is fine with me as well.

 

Probably best for you not to have an opinion about how technical analysis ought be done. The determination of lossy/non lossy does not depend on the encode/decode algorithm. The real question for MQA seems to be whether it is better than MP3 analogous to whether JPEG vs JPEG2000 is "better" than RAW/TIFF/PSD ... in any case having been around the data formats data analysis field for, oh a few decades, and seen so many formats, improvements, etc come and go and come and go and come and go,

 

I don't even look at any formats that either aren't *standardized* or have readily/freely available format converters.

 

The reason I buy (considerable number) of SACDs is because I have a PS3 that can rip to ISO, and I can then extract to DSF which is the format I use to archive, alternatively AIFF/FLAC.

 

Blah blah blah "temporal blur" blah blah blah ... been there, seen that, done that, yaaaawwwwnnn. I vote with my pocketbook. Run along and do whatever you please.

[jabbr]

1 hour ago, PeterV said:

My need for ripping and archiving is gone now I can stream MQA albums and standard FLAC in Tidal. More time to discover new musical Gems

Quote

Blah blah blah "temporal blur" blah blah blah ... been there, seen that, done that, yaaaawwwwnnn. I vote with my pocketbook.

 

You have not been here where MQA has proven to do exactly what is promised "

I have no reason to care about MQA promises. You admit not to have any real technical knowledge and you listen with your head between the speakers, so what if why should I care? 

Temporal deblurring is so 1970s. Good to see it being applied to home audio. But if you want to get a millisecond of my attention, the description needs to be both accurate and well written.

 

Proprietary voodoo BS is a waste of my time, and really of everyone’s. I’m not going to bother.

 

PS I stream DSD512 and PCM768 around my network every day. I have no need for MQA and my kids tell me they have no need for Tidal. They get all their new music on the Internet or in the in person concerts of all types we go to. MQA is a (non) solution looking for a problem.