MQA: Revolutionary British streaming technology

[britneedadvice]

From 'The Absolute Sound' CES2016 Show Report-
 
"Most Significant Trend
MQA is not quite a bonafide trend but fingers crossed for its widespread adoption. My first experience with this technology in the Vandersteen exhibit room was nothing short of exhilarating."
 
Many other similar reports both Trade and Personal. Where's the doubters now!

 

[watchnerd]

  From 'The Absolute Sound' CES2016 Show Report-
 
"Most Significant Trend
MQA is not quite a bonafide trend but fingers crossed for its widespread adoption. My first experience with this technology in the Vandersteen exhibit room was nothing short of exhilarating."
 
Many other similar reports both Trade and Personal. Where's the doubters now!

 
I think Absolute Sound said similar things about DVD-A, SACD, and HDCD...

 

[ralphp@optonline]

  From 'The Absolute Sound' CES2016 Show Report-
 
"Most Significant Trend
MQA is not quite a bonafide trend but fingers crossed for its widespread adoption. My first experience with this technology in the Vandersteen exhibit room was nothing short of exhilarating."
 
Many other similar reports both Trade and Personal. Where's the doubters now!

Quoting "The Absolute Sound" on anything having to do with audio technology is like quoting Bozo The Clown on quantum mechanics.

 

[watchnerd]

 
Quoting "The Absolute Sound" on anything having to do with audio technology is like quoting Bozo The Clown on quantum mechanics.

 
I think you're being completely unfair to Bozo and his lack of embarrassing statements on physics...

 

[ralphp@optonline]

   
I think you're being completely unfair to Bozo and his lack of embarrassing statements on physics...

True. At least Bozo knows well enough to keep his mouth shut when it comes to quantum mechanics. On the other hand, The Absolute Sounds has knows no bounds when it comes to spreading misinformation.

 

[icebear]

  From 'The Absolute Sound' CES2016 Show Report-
 
"Most Significant Trend
MQA is not quite a bonafide trend but fingers crossed for its widespread adoption. My first experience with this technology in the Vandersteen exhibit room was nothing short of exhilarating."
 
Many other similar reports both Trade and Personal. Where's the doubters now!

 
Here

!
 
when you search the TAS website NEWS sections for MQA you'll get 20+ results.
They obviously like one of their sponsor's/advertiser's product.
That's completely OK, as long as the reader is aware of this positive bias.
 
I am not into streaming and I have a decent disc player, a DAC and a MBP with Audirvana in case I long for highrez downloads to play. I don't need another proprietary new encoding format and licensed decoding hardware. This is like SACD in the download and streaming age, yeah you might be able to listen to the new files w/o the Meridian decoder but you don't get the full SQ ... hybrid SACD anyone?

 

 

[britneedadvice]

I'll take that as only 3 ? (doubters) Not bad for this site !!

 

[Gringo]

No problem with doubters, healthy sceptisism is good. Technology that offers better listening experience is to be welcomed with caution as false promises are not completely unknown and just because something is better does not ensure it will be a great success. Individuals however making comments and expressing them as fact rather than opinion does I feel lower the whole debate. Derisory comments against those with a different point of view only undermines the commenters arguement. Examining all the evidence available to you and that includes listining is the only way to give real weight to the discussion, remembering that making money is not a crime. No one is compelled to buy anything and we can all choose to stick with LP's, CD's, MP3's etc. Existing equipment, expensive or not, is always going to be superceded, (I believe that to be a fact of life), and not being able to freely steal music are factors which should not play any part of the arguement.
(Just a few thoughts not intended to insult or upset anyone, happy listening to all)

 

[watchnerd]

I have no doubt Meridian has done solid R&D on MQA.
 
But I just don't see why I would need this.
 
I can already stream lossless, both at home and mobile.
 
The better impulse response is interesting, but unreproducible with current transducers, including microphones to capture such a fast rise time (at least by any spec I've ever seen).
 
High resolution audio is something I know I can't pass an ABX test vs regular lossless, so don't need that, either.
 
What's the point of MQA?

 

[ralphp@optonline]

  I have no doubt Meridian has done solid R&D on MQA.
 
But I just don't see why I would need this.
 
I can already stream lossless, both at home and mobile.
 
The better impulse response is interesting, but unreproducible with current transducers, including microphones to capture such a fast rise time (at least by any spec I've ever seen).
 
High resolution audio is something I know I can't pass an ABX test vs regular lossless, so don't need that, either.
 
What's the point of MQA?

The point of MQA is to give Meridian a proprietary technology that can then be licensed for use by others. In other words: MONEY, as in money flowing from your bank account into their bank account.

 

[watchnerd]

I can't quantify how much is down to the remastering other than to say the very latest improved digital filtering/timing techniques are being used to compensate for and correct errors, that the original filters that DAC's used and consequently introduced in the first place. 

 

 
What "latest digital filtering" innovation is going to allow you to get around the sinc function of Fourier pairs?

 

[Gringo]

As you have taken the time and effort to go back through the thread you will be aware that I don't pretend to be clued up on the technical side of things. With regard to your question you will have also have noted that much of the technology is being kept firmly under wraps for commercial reasons. I can only assume this is where the important processes are taking place with regard to digital filtering/timing techniques. I make no pretence that I have any understanding of "sinc function Fourier pairs" so would not dream of making any comment, (I rightly or wrongly guess that you assumed that I would not have a clue). As with any technology, nothing stands still and refining and improving is ongoing so much so that as consumers we have to every so often weigh up the benefits and decide if they are worthwhile to us personally and a variable degree of this will inevitably be subjective.  As I have said several times previously, we should all be both sceptical and critical but we should also not dismiss out of hand without using our ears and listening. What I can say with full confidence is that I have listened to MQA on more than one occasion and I think it has a lot to offer. Being just a simple sole, for me the bottom line is that at the end of the day all the numbers ultimately amount to nothing either way, the enjoyment of what I am listening too and if I can afford it is what really counts. If the answer is no it's not worth it then it's not such a big deal because I'm very happy with my existing set up anyway.
As always, these are just my thoughts, no offence is intended and I have no wish to belittle or upset anyone.

 

[watchnerd]

  As you have taken the time and effort to go back through the thread you will be aware that I don't pretend to be clued up on the technical side of things. With regard to your question you will have also have noted that much of the technology is being kept firmly under wraps for commercial reasons. I can only assume this is where the important processes are taking place with regard to digital filtering/timing techniques. I make no pretence that I have any understanding of "sinc function Fourier pairs" so would not dream of making any comment, (I rightly or wrongly guess that you assumed that I would not have a clue). As with any technology, nothing stands still and refining and improving is ongoing so much so that as consumers we have to every so often weigh up the benefits and decide if they are worthwhile to us personally and a variable degree of this will inevitably be subjective.  As I have said several times previously, we should all be both sceptical and critical but we should also not dismiss out of hand without using our ears and listening. What I can say with full confidence is that I have listened to MQA on more than one occasion and I think it has a lot to offer. Being just a simple sole, for me the bottom line is that at the end of the day all the numbers ultimately amount to nothing either way, the enjoyment of what I am listening too and if I can afford it is what really counts. If the answer is no it's not worth it then it's not such a big deal because I'm very happy with my existing set up anyway.
As always, these are just my thoughts, no offence is intended and I have no wish to belittle or upset anyone.

 
The question about sinc and Fourier pairs isn't meant to belittle.
 
However, it is an important question when we talk about signal processing.  Signal processing isn't as simple as throwing more CPU horsepower at the algorithms and magical things will happen if you're clever and have enough compute horsepower.
 
It is governed by some pretty iron-clad mathematical laws and physics (this isn't just for audio, either) that govern the relationship between time domain and frequency domain.  In this case, the relationship between impulse response (and ringing) is tied into the nature of a filter implementation.  There are certain patterns that are established, some common ones look like this:
 

 
 
 
 
So when a company says they're going to improve impulse response (my interpretation of what they mean when they talk about reduced 'temporal blurring'), a person familiar with signal processing theory will say:
 
"Okay, if you want to optimize for the best possible impulse response it means you're going to be using a gentler, minimum phase filter....which is potentially going to leave more noise in the frequency domain or, alternatively, you're going to have to roll off the high-end earlier."
 
Remember, while compute power has increased immensely in the last decades, the signal processing upon which audio is based is decades old and still applies.
 
;TLDR: if MQA really is improving impulse response by using a less-aggressive filter, this a big departure from Redbook specs and would change the spectral and noise content of music unless address via some other unknown means.

 

[spruce music]

   
The question about sinc and Fourier pairs isn't meant to belittle.
 
However, it is an important question when we talk about signal processing.  Signal processing isn't as simple as throwing more CPU horsepower at the algorithms and magical things will happen if you're clever and have enough compute horsepower.
 
It is governed by some pretty iron-clad mathematical laws and physics (this isn't just for audio, either) that govern the relationship between time domain and frequency domain.  In this case, the relationship between impulse response (and ringing) is tied into the nature of a filter implementation.  There are certain patterns that are established, some common ones look like this:
 
 
 
 
 
 
So when a company says they're going to improve impulse response (my interpretation of what they mean when they talk about reduced 'temporal blurring'), a person familiar with signal processing theory will say:
 
"Okay, if you want to optimize for the best possible impulse response it means you're going to be using a gentler, minimum phase filter....which is potentially going to leave more noise in the frequency domain or, alternatively, you're going to have to roll off the high-end earlier."
 
Remember, while compute power has increased immensely in the last decades, the signal processing upon which audio is based is decades old and still applies.
 
;TLDR: if MQA really is improving impulse response by using a less-aggressive filter, this a big departure from Redbook specs and would change the spectral and noise content of music unless address via some other unknown means.

I know I have mentioned it a couple times.  It seems they can manage this with compressive sensing sampling and reconstruction algorithms.  It may just be marketing speak.  I don't have a good understanding of compressive sampling/sensing except in a very general way.  If signals meet certain metrics you can accurately sample and reconstruct them with a very low bit rate which can encompass a very wide frequency range.  So under the right conditions and for the right kind of signal the hard limits and trade offs you show in your post don't necessarily apply any more.
 
Just do a google search for compressive sensing or compressive sampling.  Especially do so using google scholar.  Plenty of short papers giving outlines of how it works.
 
I still think they are mostly trying to gain money for themselves and also am not convinced temporal blur is a problem.  It would have been better to simply apply compressive sensing in place of normal PCM and drop all the other junk.  Perhaps we could have accurately reconstructed audio with parameters equal to 24 bit 60 khz sample rates at the bit rate of 128k mp3.  Now that would be worthwhile.  Of course Meridian couldn't control that or make money from it.

 

[watchnerd]

  I know I have mentioned it a couple times.  It seems they can manage this with compressive sensing sampling and reconstruction algorithms.  It may just be marketing speak.  I don't have a good understanding of compressive sampling/sensing except in a very general way.  If signals meet certain metrics you can accurately sample and reconstruct them with a very low bit rate which can encompass a very wide frequency range.  So under the right conditions and for the right kind of signal the hard limits and trade offs you show in your post don't necessarily apply any more.
 

 
How would reconstruction algorithms help them deal with increased noise in the stopband that result from using a minimum phase filter?