MQA is Vaporware

[audiventory]

MQA was launched on December 2014. It is now January 2, 2017 so let’s see where MQA is at today.

 

There is closed loop: manufacturers for label owners > label owners for exists devices.

 

In my opinion, main problem of MQA that it is closed standard.

 

I don't see reasons for label owners to wide support the format.

 

May be compatibility with traditional CD only. But there exists competitors - hybrid technologies of CD/DVD.

 

 

 

FLAC give about 40-50% compression and it's free and open source.

 

Currently super compression interesting for movie rather than for audio.

 

 

In the future DRM may be will reworked for modern realities. But currently it is not enough suitable for conscientious end users.

[audiventory]

High bitrates and low noise are all very well, but, next to the air between a musician and audience, it's open formats that present the lowest barrier to transmission.

 

I suppose, monopoly of one format is impossible.

 

So plurality of formats will supported as before.

[audiventory]

I don't think it can all be done by software. Certainly from reading the patents the emphasis is on a end-to-end system which includes the ADC and DAC. Reasons I can think of include:

 

- how can MQA select the correct reconstruction filter in the DAC if its not got any control over the DAC?

- (guessing here) perhaps there is a potential for the DAC upsampling to mess up the MQA unfolding.

- with certain DACs the USB interface is the limiting factor (like the Dragonfly). If it was all done in software then 96 kHz would

be the max input. But with MQA firmware in the DAC it can support unfolding to at least 192 kHz (to be confirmed).

 

The basic idea is that MQA is an end to end system that wants to control the ADC and the DAC. This is why many of the DAC manufacturers don't like it. Does the MQA system work? We'll find out this year, as we experiment with various DACs etc.

 

If I understand correctly, you think that hardware can more that software for MQA decoding, isn’t it?

 

If so, I can say that MQA decoding very probably (nobody except its developers exactly know how it work) is pure math.

 

If there pure math, PC, as rule, have more math capabilities than hardware decisions.

 

For super fast math may be used FPGA or one-chip processors. But there may be limitations in application kinds, price and precision.

But it is general reasoning. Need consider each implementation.

 

However, most audio applications may be solved on PC faster (in development) and with maximal reasonable precision.

 

In PC resampling may be performed in 64-bit float point. But, I suppose, it is too hard task for hardware decisions.

 

Even if MQA use some analog processing (I very doubt that), then analog processing may be modelled in PC.

[audiventory]

A firmware update won't rewire the analogue parts of a DAC.

 

Analog filter there will do not updated

[audiventory]

I wonder if you are going to have the debate of the likes of FLAC vs WAV where the decoder is blamed for the degradation in SQ?

 

FLAC decoder may be checked right now. Because encoder is available for anybody. But discussion continued

[audiventory]

Have you heard MQA?

Even Darko thought MQA sounded clearly better than the original HD files in his comparison.

 

Does you meant, original file after encoding and decoding sound better than before?

[audiventory]

You could easily add a standard cryptographic signature (e.g. PGP) to a FLAC file.

 

Like Mac’s software. Files there do not encrypted. For checking launched system utiliy.

[audiventory]

It’s easy for me to believe that MQA can help many DACs get better sound from a 16/44.1 or 16/48k bit stream than they do now.

 

How it is possibly?

[audiventory]

I think what he is saying is it brings a kind of "best practice" to the heretofore poor overall filter/reconstruction design of *most* DAC's. He is not saying that MQA is necessary or brings something undiscovered to the table - so we are back to the overall (end to end, etc.) package argument. I look at this argument/angle and then I look at the actual MQA implementation we have currently (mostly Tidal of course) and judge it a hit or miss...a big "depends", but then we are back to the "mastering" angle...

 

I think, that I understand, what you meant. But let's look to example:

 

1. We have compressed as MQA original high-res (as example 192/24).

 

2. We compare it with the original hi-res 192/24.

 

3. Both cases we use single DAC's mode (sample rate/bit depth - 192/24).

 

Where's improving?

[audiventory]

He didn't say that. You have to read more carefully.

"It’s quite another thing to claim that MQA betters 24/192 or DSD – on the surface it explicitly doesn’t have the same S/N or bandwidth… And 24/192 and DSD don’t have the problems that Redbook does. IMO they solve Redbook’s problems much more elegantly than MQA and with less damage to the signal, tho not at the same bit rates.

 

 

It’s easy for me to believe that MQA can help many DACs get better sound from a 16/44.1 or 16/48k bit stream than they do now.

 

I understand now: it is about of compression audio. Not about improving sound quality.

[audiventory]

I cannot consider MQA as format for high end system... IMO MQA want to address an average customer and we audiophiles are not important in the MQA business plan. MQA is about music content, not about audiophiles.

 

Unfortunatelly, we don't know exactly: what transformed in an audio signal after the decoding-encoding.

 

May be there range 0 ... 20 kHz is transmitted without transformations.

[audiventory]

what is currently possible. I don't see a general solution how to apply DSP on fully decoded MQA content - played on usual software player like JRiver or Foobar2000 (and many others) to for example Meridian MQA DAC. Do you know such a solution? Was it anywhere explained, what has a player developer to do to enable DSP on MQA content? Was such a general possibility for software players anywhere mentioned from the side of MQA company?

 

Read here (question «In addition to hardware products, do you also provide solutions for software applications?»)

MQA | For playback providers

 

If I understand correctly, the decoder provided for streaming services apps only.

If player have status [streaming service app], it can use the decoder and process unpacked audio stuff.

[audiventory]

Why bother with specialized encoding "schemes" that are closed source when FLAC gives you everything you NEED...Of course, FLAC does not have any "cool" aid to sell you ;-)

 

FLAC is lossless, but have lesser compression ratio, than MQA. For broadcast/streaming applications it may be critical.

[audiventory]

But regardless... Anyone able to stream MQA could stream a 24/96 file. Bandwidth of this sort is generally not a problem today.

 

For home network - no problem, for public network - bandwidth is critical, as rule.

 

Here I done some calculations (for unfolded MQA to 192 kHz bandwidth), that may be interesting http://www.computeraudiophile.com/f8-general-forum/why-universal-music-partnered-mqa-article-31755/index2.html#post639602

[audiventory]

And they claim the 8-bit undecoded version still sounds better than real CD. No way in hell.

 

We can't deny other people's subjective perception - we have no proofs. Otherwise, claims, that based on subjective perceptions, are not proofs too

[audiventory]

critical??? don't think so:

 

from:

 

MQA

 

 

 

 

So flac it up and you get even more bandwidth savings without PROPRIETARY "schemes of unfolding"

 

Let calculate bitrate for 1000 users simultaneously for abstract audio streaming service, as example. If you save band 2 times it will good

 

In my post above link to my calculations for WiFi in other thread.

[audiventory]

2 hours ago, Nikhil said:

Of course "apodizing filters" sounds very impressive.  Fixing Time distortion? Less said the better ... Mitch's article pretty much nailed that lie.

For using "time distortion", "timing" and other same terms need define it exactly.

All information, that I read about "what is time distortions" is information about minimized amplitude of ringing and fixing of pre-ringing, if I correct understand links below:

1. http://www.aes.org/e-lib/browse.cfm?elib=17501

2. https://patentscope.wipo.int/search/en/detail.jsf?docId=WO2014108677&recNum=1&maxRec=&office=&prevFilter=&sortOption=&queryString=&tab=PCTDescription

Also pay attention to claims #21 of the patent (link #2). As I understand, there described using resampling filters, that allow achieve lesser total filters impact to signal in pair (encoder and decoder).

When digital audio signal compressed, such way allow reduce signal damaging (ringing) due resampling.

If digital audio signal is not compressed, there reduce nothing.

[audiventory]

13 minutes ago, Sal1950 said:

No, not all.

I'm in no hurry to see my lossless CD quality downloads replaced with MQA's lossy facsimile of 16/44

Prefer to see MQA disappear in a puff of Vaporware smoke.

 

I want only lossless downloads.

[audiventory]

2 minutes ago, jhwalker said:

I'd hate to kill a nascent technology that *may* provide better outcome(s) just because "I only want lossless downloads".

 

For broadcast the technology is ok.

[audiventory]

22 minutes ago, jhwalker said:

 

Every stream that is done in MQA vs. lossless is a net savings in bandwidth.  Storage is another story

 

John,

Many years storage for audio is not story

[audiventory]

2 hours ago, Samuel T Cogley said:

I know of two software MQA Core implementations: Tidal desktop and Audirvana+ 3.0.  Are there any others?

 

MQA Core is not full MQA ability? As I remember, full MQA ability allowable with apparatus only?

[audiventory]

45 minutes ago, mrvco said:

With a bit of momentum, I expect that MQA makes a lot of sense from a record label perspective.  The labels are struggling with the economics of Hi-Res music production and distribution, as well as their inherit "master tape" piracy fears.

 

How MQA help in struggling with piracy?

[audiventory]

3 minutes ago, mrvco said:

 

It helps offset the (real or imagined) losses attributed to piracy  Not getting a cut of Apple's iPod hardware sales was a big complaint (piracy enabling device!!!) of the labels once they realized where Apple was really making its money.  This motivation was also behind the "piracy tax" on blank CD media... even though paying the tax didn't mean you could legally pirate content.

 

How loses from piracy will come back to authors/copyholders via MQA?

Could you decribe it step-by-step?

 

 

[audiventory]

8 minutes ago, mrvco said:

 

I'm speaking hypothetically of course since I'm not worthy of knowing what MQA's business model really is, but revenue is revenue for the record labels, whether it comes from record sales or ancillary (e.g. MQA royalty) revenue.  If I buy the MQA album and an MQA DAC, then presumably they would generate revenue on both transactions.  If I pirate the MQA album, but I still have to buy the MQA DAC to take advantage of the MQA encoding, then at least they make money on the MQA DAC royalty.

 

Any MQA royalty get MQA patent holder. How these money come to copyright holder?

Does somewhere noted that the patent holder will pay to copyright holder?

[audiventory]

4 minutes ago, The Computer Audiophile said:

 

I'm not saying this is the case, but it can easily be done. For example, see the private copying levy that kicked back money to the rights holders based on other peoples' patents. And, music CD-Rs

 

https://en.wikipedia.org/wiki/Private_copying_levy#Blank_music_CDs_and_recorders

 

Yes. I know this scheme. I want to say, that I don't stumbled on information, that MQA will provide it.