MQA is Vaporware

[Fokus]

51 minutes ago, Jud said:

Vote with your dollars (as in, the absence of them from MQA).  It's a language that gets listened to.

 

Doesn't work. I purchased an Explorer 2 in order to analyse it.

 

[Jud]

1 minute ago, Fokus said:

Ironically, if MQA had been a bit more open from the beginning they would by now have been in a position with a much better chance for long-term success. But no ... they chose to be driven by maximum greed.

 

This is easy to say in hindsight.  But I think it was much more difficult to project at the time several years ago what would eventually allow for a return on investment.  In particular, I think they likely underestimated progress in streaming speeds, moved more slowly than they'd anticipated so streaming speeds had ramped up by the time they commercialized their ideas, or both, which negatively affected the value proposition of compression-while-retaining-nice-sound and left them to market on the basis of Fabulous Fidelity that really wasn't there.

[semente]

On 16 July 2017 at 3:04 PM, Digital Assassin said:

And now Mr. Analog, Michael Fremer has jumped on board the Atkinson MQA train with ridiculous over the top

statements.

 

According to Archimago:

 

"Most recently (August 2017 Stereophile, reviewing Brinkmann's Nyquist DAC) I read that mister analogue himself Michael Fremer endorses MQA - "Had this been CD sound in 1983, I'd still be an LP guy - but I'd also be all in with digital." Wow... Really? Consider that later on in the article he used the analogy of "Grand Canyon of analog-vs-digital" to describe the sonic divide to describe the difference; did MQA make that much difference!?"

 

Please note, he said the identical nonsense a few months ago while being interviewed at a show.

 

This tells me he had a pre-determined agenda.

 

Now the entire staff at Stereophile is officially on board. They got the boss's memo.

 

 

 

MQA Vinyl Edition 220g

 

[Digital Assassin]

38 minutes ago, semente said:

 

MQA Vinyl Edition 220g

 

post of the week so far! 

[fung0]

5 hours ago, Jud said:

Vote with your dollars (as in, the absence of them from MQA).  It's a language that gets listened to.

 

Exactly! In our eagerness for what is new and surrounded by hype, we often forget this 'golden' rule...

[Rt66indierock]

5 hours ago, Jud said:

 

They could.  Currently, they seem to be taking a long time to succeed.

 

Vote with your dollars (as in, the absence of them from MQA).  It's a language that gets listened to.

 

The MQA folks are certainly making it easy to vote with your pocketbook. I can only stream TIDAL something Sprint is finding hard to give away.

[fung0]

5 hours ago, Fokus said:

Ironically, if MQA had been a bit more open from the beginning they would by now have been in a position with a much better chance for long-term success. But no ... they chose to be driven by maximum greed.

 

Absolutely!

 

MQA is seeking to launch not just a product, but a whole new way of distributing music. The right way to do that would be to gain expert acceptance for the technology, and create consensus among all the participants. Or simply forward the spec, as 'bread cast upon the waters,' and let it succeed or fail on its merits. Clearly, this is not an easy way to make big profits.

 

Instead, MQA chose to limit expert testing of its new technology, and surround the specs with patent barriers, non-disclosure agreements and bafllegab. Worse, it tailored its proposed new format to appeal not to consumers, but to the basest short-term instincts of large publishers and distributors. The 'authenticated' part, in particular, has been shown to have no connection whatsoever to the claimed benefits of 'audio origami.' It's an anti-feature which has no obvious purpose other than to encourage corporate buy-in and maximize licensing revenue.

 

Forging new standards is dreary, thankless work - if done the right way. Designing and marketing a new standard with profit as the primary consideration is at best a perversion of the process. It isn't proof that the underlying technology is without merit, but it should certainly raise a lot of skepticism, if only on procedural grounds.

[james45974]

2 minutes ago, fung0 said:

Instead, MQA chose to limit expert testing of its new technology, and surround the specs with patent barriers, non-disclosure agreements and bafllegab. Worse, it tailored its proposed new format to appeal not to consumers, but to the basest short-term instincts of large publishers and distributors. The 'authenticated' part, in particular, has been shown to have no connection whatsoever to the claimed benefits of 'audio origami.' It's an anti-feature which has no obvious purpose other than to encourage corporate buy-in and maximize licensing revenue.

 

From a consumer standpoint I am eagerly awaiting the many filters that will no doubt be offered to the marketplace as "MQA like".  I have no issue with furthering the pursuit of perfecting digital playback, but MQA seems to be intent on alienating what may be a good portion of their potential customer base with their practices.

[JoeWhip]

3 hours ago, semente said:

 

MQA Vinyl Edition 220g

 

Back from MF's days as a stand up comic?

[Shadders]

On 17/07/2017 at 1:34 PM, Shadders said:

Hi,

Thanks - yes - just looked up the book i have - constant time delay, but varying phase delay. The group delay graphs for the various orders are constant. I need to examine more closely.

Regards,

Shadders.

Hi,

I have worked out my mistake, group delay is the differential of phase change across the frequency band. So constant group delay is linear phase change.

The time differences between a 100Hz and 20kHz signal for a 4th order bessel filter with cut off frequency of 22.05kHz is approximately 3.5uS - so not very much.

I have yet to simulate an 8th order elliptical filter - so i assume that this is what MQA is supposed to address - the temporal smearing of the signal due to the phase changes across the frequency band ?

Given the higher sample rates used today - i assume that temporal smearing is not an issue so we don't need MQA ?

Regards,

Shadders.

[Jud]

2 hours ago, Shadders said:

Hi,

I have worked out my mistake, group delay is the differential of phase change across the frequency band. So constant group delay is linear phase change.

The time differences between a 100Hz and 20kHz signal for a 4th order bessel filter with cut off frequency of 22.05kHz is approximately 3.5uS - so not very much.

I have yet to simulate an 8th order elliptical filter - so i assume that this is what MQA is supposed to address - the temporal smearing of the signal due to the phase changes across the frequency band ?

Given the higher sample rates used today - i assume that temporal smearing is not an issue so we don't need MQA ?

Regards,

Shadders.

 

An experiment:  Set up a free test installation of Audirvana Plus.  Set upsampling preferences as you like, but leave the “pre-ringing” setting at 1.0 (linear phase).  Listen to whatever you like.  Then adjust pre-ringing to 0.0 (minimum phase) and see if you hear a difference.  (You may wish to upsample to the maximum input your DAC will accept, so as much of the upsampling as possible is performed by iZotope 64-bit SRC bundled with Audirvana Plus.)

[opus101]

2 hours ago, Shadders said:

I have yet to simulate an 8th order elliptical filter - so i assume that this is what MQA is supposed to address - the temporal smearing of the signal due to the phase changes across the frequency band ?

What are you using for simulation? I've simmed some elliptic filters (7th order) in LTspice, one or two examples are shown in this thread - http://www.whatsbestforum.com/showthread.php?8858-Digital-that-sounds-like-analog&p=154399&viewfull=1#post154399

[Fokus]

8 hours ago, Shadders said:

I have yet to simulate an 8th order elliptical filter - so i assume that this is what MQA is supposed to address - the temporal smearing of the signal due to the phase changes across the frequency band ?

 

No.

 

MQA's nemesis is the ringing visible before and after the main lobe in the impulse response of a linear phase filter.

 

They solve this by starting from a linear phase high sample rate recording, preferably 4x or higher, and then reduce this to 2x with an extremely leaky downsampling (anti-aliasing) filter, a filter with a very narrow impulse, but, obviously, with a lot of aliasing. They claim they analyse the source signal and then pick a downsampling filter so that the amount of aliasing that effectively hits the audible band does not exceed the natural noise already present in the audible band (i.o.w. the noise is to mask the aliasing). The 2x signal is then folded into 1x with the origami trick, for distribution (streaming, download, MQA-CD).

Upon replay the 1x signal is first unfolded to 2x (inverse origami), and then upsampled to the original rate with the leaky filters documented by Mansr here on CA. These filters obviously cause a lot of imaging, as can be seen here for a '192kHz' example:

 

 

 

====

 

As for the audibility of pre-ringing. You can follow Jud's advice. But even more telling would be this:

 

-reverse the file (i.e. front becomes back)

-convert with a minimum phase filter (see Jud's post)

-reverse the file again

 

Now you have a maximum phase file: all pre-ringing, no post-ringing. Have a listen.

 

 

[esldude]

16 minutes ago, Fokus said:

 

No.

 

MQA's nemesis is the ringing visible before and after the main lobe in the impulse response of a linear phase filter.

 

They solve this by starting from a linear phase high sample rate recording, preferably 4x or higher, and then reduce this to 2x with an extremely leaky downsampling (anti-aliasing) filter, a filter with a very narrow impulse, but, obviously, with a lot of aliasing. They claim they analyse the source signal and then pick a downsampling filter so that the amount of aliasing that effectively hits the audible band does not exceed the natural noise already present in the audible band (i.o.w. the noise is to mask the aliasing). The 2x signal is then folded into 1x with the origami trick, for distribution (streaming, download, MQA-CD).

Upon replay the 1x signal is first unfolded to 2x (inverse origami), and then upsampled to the original rate with the leaky filters documented by Mansr here on CA. These filters obviously cause a lot of imaging, as can be seen here for a '192kHz' example:

 

 

 

====

 

As for the audibility of pre-ringing. You can follow Jud's advice. But even more telling would be this:

 

-reverse the file (i.e. front becomes back)

-convert with a minimum phase filter (see Jud's post)

-reverse the file again

 

Now you have a maximum phase file: all pre-ringing, no post-ringing. Have a listen.

 

 

 

 

Funny is it not.  The solution to ringing that is at an inaudible frequency is all this rigmarole which creates audible band aliasing which we can cover with noise masking, and feel completely okay about as being higher fidelity when played back with leaky imaging filters that might also intermodulate into the audible band. 

[Fokus]

1 hour ago, esldude said:

Funny is it not.  The solution to ringing that is at an inaudible frequency is all this rigmarole which creates audible band aliasing which we can cover with noise masking, and feel completely okay about as being higher fidelity when played back with leaky imaging filters that might also intermodulate into the audible band. 

 

That is too stern.

 

The original MQA papers make clear that it is the intention, at all stages, to assess signal levels and ensure that aliasing and imaging remain within acceptable bounds.

 

It really is rather clever.

 

But it does not solve a real problem.

 

(But then, is a clever solution to an imagined problem really clever?)

[Shadders]

8 hours ago, Jud said:

 

An experiment:  Set up a free test installation of Audirvana Plus.  Set upsampling preferences as you like, but leave the “pre-ringing” setting at 1.0 (linear phase).  Listen to whatever you like.  Then adjust pre-ringing to 0.0 (minimum phase) and see if you hear a difference.  (You may wish to upsample to the maximum input your DAC will accept, so as much of the upsampling as possible is performed by iZotope 64-bit SRC bundled with Audirvana Plus.)

Hi,

Thanks. Seems to be a Mac only program. Also, do not have a DAC, but do have a surround sound processor - so testing is limited.

Regards,

Shadders.

[Shadders]

7 hours ago, opus101 said:

What are you using for simulation? I've simmed some elliptic filters (7th order) in LTspice, one or two examples are shown in this thread - http://www.whatsbestforum.com/showthread.php?8858-Digital-that-sounds-like-analog&p=154399&viewfull=1#post154399

Hi,

After not being involved in this area of design for over a decade - just starting to get back into the area. I want to become more familiar with the subject, in the analogue domain using op-amp filters, then later in the digital domain to gain experience in DSP techniques.

Simulation will allow me to remember the theory and build the templates for filters etc.

I am interested in what the problem is that MQA is stated to solve - seems to be temporal blur as per many "pseudo technical marketing" texts - so just want to understand rather than repeat others verbatim.

Regards,

Shadders.

[Shadders]

2 hours ago, Fokus said:

 

No.

 

MQA's nemesis is the ringing visible before and after the main lobe in the impulse response of a linear phase filter.

 

They solve this by starting from a linear phase high sample rate recording, preferably 4x or higher, and then reduce this to 2x with an extremely leaky downsampling (anti-aliasing) filter, a filter with a very narrow impulse, but, obviously, with a lot of aliasing. They claim they analyse the source signal and then pick a downsampling filter so that the amount of aliasing that effectively hits the audible band does not exceed the natural noise already present in the audible band (i.o.w. the noise is to mask the aliasing). The 2x signal is then folded into 1x with the origami trick, for distribution (streaming, download, MQA-CD).

Upon replay the 1x signal is first unfolded to 2x (inverse origami), and then upsampled to the original rate with the leaky filters documented by Mansr here on CA. These filters obviously cause a lot of imaging, as can be seen here for a '192kHz' example:

 

 

 

====

 

As for the audibility of pre-ringing. You can follow Jud's advice. But even more telling would be this:

 

-reverse the file (i.e. front becomes back)

-convert with a minimum phase filter (see Jud's post)

-reverse the file again

 

Now you have a maximum phase file: all pre-ringing, no post-ringing. Have a listen.

 

 

HI,

Thanks - still progressing through the basics - so will examine in more detail later. Thanks.

Regards,

Shadders.

[Fokus]

39 minutes ago, Shadders said:

Thanks - still progressing through the basics - so will examine in more detail

 

 

Some useful background

 

https://www.stereophile.com/features/106ringing/

[Jud]

57 minutes ago, Shadders said:

Hi,

Thanks. Seems to be a Mac only program. Also, do not have a DAC, but do have a surround sound processor - so testing is limited.

Regards,

Shadders.

 

You can also use a free test version of iZotope Rx itself and do the upsampling offline, then use mansr’s SoX fork if you wish to sigma-delta modulate to DSD format; or just use the SoX fork for the whole shebang.

 

You sure you don’t have a DAC, like reasonable sound card and headphones?  The point is to have not only something you can see, but something you can hear (or can’t).

[Don Hills]

Has anyone looked at the filter used in the core decoder to join the 0-24 KHz band with the 24-48 KHz band? It'll be the complement of the filter used to split them when encoding. I'll bet they're quite sharp and will ring. (Although, if you split with a ringing filter and rejoin with the complement of that filter, I suspect it cancels the ringing.)

[Jud]

3 hours ago, esldude said:

 

 

Funny is it not.  The solution to ringing that is at an inaudible frequency is all this rigmarole which creates audible band aliasing which we can cover with noise masking, and feel completely okay about as being higher fidelity when played back with leaky imaging filters that might also intermodulate into the audible band. 

 

At least if you manage to have a filter that leaks *and* rings, it can be quite audible.  With A+ I’ve managed to make myself oversampling filters that made music sound like someone had turned up the reverb way too high.  I haven’t gone back to do the experiment I recommended, otherwise using the default A+ settings, which should be non-leaky, and just play with phase (called pre-ringing in the A+ settings).  And I don’t have a free trial of iZotope Rx, though that route is open to others.  Or there’s the SoX fork.

 

Using A+ just makes it more convenient to adjust phase nearly “on the fly” with the very good iZotope resampling software, though it is (currently) Mac-only.

[Fokus]

52 minutes ago, Don Hills said:

Has anyone looked at the filter used in the core decoder to join the 0-24 KHz band with the 24-48 KHz band? It'll be the complement of the filter used to split them when encoding.

 

I asked for this before.

 

They need two quadrature mirror filter pairs for the split and the join.

 

The following (pretty dire)  filter response always surfaces when an MQA DAC is tested with non-MQA signals:

 

 

 

I would like to know if this filter is one in the two QMF pairs.

 

If it is, then the recording side downsampling filter has to pass on a lot of aliasing into the 1x band. This aliasing then invariably corrupts the signal when it is not replayed by an MQA decoder. This would make non-decoded MQA files markedly inferior to their CD-rate equivalents, contrary to MQA's claims.

 

[mcgillroy]

On 7/13/2017 at 7:33 PM, crenca said:

 

Well stated, however I think it is important to remember that there are at least two categories of "Audiophile":

 

1)  The traditional 2 channel guy who has $money$ but no electronic, engineering, digital background and to which you can sell $5k USB and CAT cables to (to say nothing of boxes of rocks).  This guy is likely to be subjectivist because what else can he be - he understands to some degree what he does not understand.

 

2)  The younger, largely "personal audio" guy who has less money, is more likely to have passed calculus, programmed something, or worked in the server room (or even set up his home network) and who is ad/marketing adverse, skeptical of theory and trusting of physics.  He is more likely to be balanced or an outright objectivist

 

The traditional audiophile press serves only #1, and does very little (other than "sounds like" reviews and new product announcement) for #2.  #1 is an aging and shrinking demographic - although because high end is also "life style" product for the rich and famous it is also expanding.  #2 is an expanding (how much is debated) and is the majority future.  The audio press and the entire "high end" industry is however wholly dependent on #1 because that is where the real money is - selling things for orders of magnitude their actual cost of goods/manufacture/distribution/etc.

 

The traditional press/industry knows the gravy train can't last forever, and are actively courting #2.  However, look how they are doing it - witness the "Jana Dagdagan" phenomenon over at Stereophile (i.e. hiring smart, young, pretty girl and see what happens) and recent efforts of so many traditional brands to offer some overpriced HP amp of some sort.  However, even traditional #2 manufacturers can't help themselves and are pressing personal audio prices up and up with ever more ethereal, "subjectivist" offerings such as HP's in the 6-50K range.

 

Also, while Atkinson is an electronics guy of some capability he is not a digital/computer/software guy as such (MQA is first and formost software), though I have noticed he has learned a thing or two through this MQA debacle.  

 

Thx and yes that is an interesting development happening in the audio press. Noticed Dagadan's writing too and she is the only one on Stereophile who managed to sneak in a critical take on MQA into the publication:

 

Jana Dagadan: Any opinions on MQA? It doesn't make you want to revisit digital?

 

Jim Hagemann: No opinion other than it looks like a really clever technology. I don't know how it sounds but it's a clever way to combine everything and be compatible. I've heard some horrible stories about MQA and licensing. They're brutal with their licensing for manufacturers who want to use it and employ it. They sting you. For the small guy like me, I don't think I could afford to get into that business.

 

https://www.stereophile.com/content/industry-profile-jim-hagerman-hagerman-audio-labs

 

That's pretty damming and nowhere on Stereophile any follow up on it.

 

Secondly I think you are spot on with your typology of audio buyers. To expand on your description of #2 (young audio buyer) and my earlier notion that digital audio draws a different crowd with a different skillset I like to add the following observation:

 

There are plenty of well educated MINT-graduates rendered open-floorplan-dwellers by the tech-industry.

 

Their only chance of privacy are their headphones. These are well-paid kids and the explosion of headphone offerings in the past decade not least caters to this demographic with their ample purchasing power and appetite for distinction through gadgets.

 

Once people regularly started to pay $300.- and more for a headphone a subset of these became Head-Fi enthusiasts and/or audiophiles.

 

Now try to think of an audio-"innovation" that would challenge the curiosity, political sensibilities and technical competence of this demographic?! Basically software engineers familiar with tool-chains which are a mix of proprietary, FS and OS-software.

 

Let's see - it should include:

 

1. hyperbolic claims about shortcomings in information theory

2. ditto for the related sampling-theorem

3. obfuscation of the "science" behind these claims with meaningless graphs and numbers published only in high-quality journals like Stereophile and TAS.

4. disavowment of peer-reviewed research

5. ditto for independent listening tests

6. reliance on proprietary and closed source software

7. a licensing regime openly aiming to lock-in vendors

8. DRM hidden in plain-sight

9. NDAs galore to cover 1-8.

 

That's a brilliant marketing strategy to reach digital natives! Especially the well-educated ones with interest in audio-hard & software and enough discretionary income.

 

Bob S. and J. Atkinson really got themselves into some fun here. Let's make sure we all are stocked up well on Popcorn

 

 

 

 

 

[soxr]

27 minutes ago, Fokus said:

 

I asked for this before.

 

They need two quadrature mirror filter pairs for the split and the join.

 

The following (pretty dire)  filter response always surfaces when an MQA DAC is tested with non-MQA signals:

 

I recently asked Mytek if they always enforce the minimum phase upsampler MQA is using.
Mytek was against it, but MQA tried to force it anyway.

So I believe MQA is making sure redbook will sound worse than MQA because of those crappy filters being active.

But Mytek refused, and you can turn off the MQA decoder and turn off the crappy minimum phase upsampler.

It's crappy because it's limited in #taps and cannot be compared to minimum phase resampling such as in libsox (which auralic is using, just dump their firmware and run ldd on lightningserver and it will reveal libsox) or the soxr library, which is the resampling-ony part of libsox.