Another major look at MQA by another pro.

[lucretius]

18 minutes ago, ShawnC said:

Can't you take a Dragonfly or Mytech Brooklyn and run the output into a reciever/preamp and use the room correction there, like Anthem,  Audyssey (Marantz) or Dirac (Emotiva)?     

 

Yes but that would be a  subsequent A/D and D/A conversion.

[Fitzcaraldo215]

9 minutes ago, ShawnC said:

Can't you take a Dragonfly or Mytech Brooklyn and run the output into a reciever/preamp and use the room correction there, like Anthem,  Audyssey (Marantz) or Dirac (Emotiva)?     

Yes, you can.  But, the signal is then converted from d-a-d-a, robbing it of transparency.  What we want is DSP room correction entirely in the digital domain with one final d-a conversion in the DAC.  But, with MQA, that DSP correction would screw up MQA's metadata.

[Rt66indierock]

2 hours ago, The Computer Audiophile said:

I hear you. I know many people "working" with MQA now. It's moving slower than the tectonic plates. 

Good to know.  There certainly seem to be a wide variety issues for a limited number employees.

[PeterV]

On 25-7-2017 at 4:22 PM, soxr said:

 

PeterV clearly did not fully understand the article or just highlights those parts that confirm his bias.

MQA is not the studio sound. PeterV does not understand the typical studio workflow, and the fact that MQA is not encoded in the studio but in a remote MQA facility.

 

2

I received confirmation from MQA that meanwhile 3 MQA encoders are in use in recording studio's worldwide. Encoding and 'white-glove' provenance according to MQA training is being performed in those recording studio's

[PeterV]

On 25-7-2017 at 0:22 PM, Shadders said:

Hi,

Specifically "Nobody over here knows how the MQA encoding and decoding process really works."

What we do know, is that the actual 24bit sample is dithered down to 17bits, BUT is stated to recreate near 24bit sound through its process.

The filter applied to reconstruct is one of 32.

So we are not obtaining an accurate audio sequence, but something that is made to sound nice, or we are told is more accurate (but how can that be if we drop 7 bits from the sample - simple analogy).

You might as well process the file and add even order harmonics to make it sound nicer.

MQA is an effects processor, that is all. (my simplistic interpretation)

Regards,

Shadders.

 

Your technical analysis totally overlooks the aspect of deblurring, which is the essence of the encoding process. The whole MQA debate should focus much more on this important aspect, since that seems the reason why MQA end-to-end improves the sound quality ( to my ears and opinion and others..)

[firedog]

25 minutes ago, PeterV said:

Your technical analysis totally overlooks the aspect of deblurring, which is the essence of the encoding process. The whole MQA debate should focus much more on this important aspect, since that seems the reason why MQA end-to-end improves the sound quality ( to my ears and opinion and others..)

So far it seems "deblurring" is  a euphemism  for a filter without pre ringing. You don't need MQA for that.

[PeterV]

On 26-7-2017 at 9:57 AM, Fokus said:

 

You are aware that at least some MQA files have been derived from totally different masters, perhaps even different mixes, than the standard releases that are out there?

1

 

Yes, of course. There are many Masters out there and it is as with HDtracks and the Mobile Fidelity remasters.. I would like to be sure that there is only 1 all-time best MQA version, but I expect this will be not feasible.

On 26-7-2017 at 9:57 AM, Fokus said:

 

You are aware that undecoded MQA contains the hidden signal crud that contaminates the base signal, potentially in an audible way?

 

Un-decoded MQA files do not sound worse than the original native file and in some cases, the un-decoded MQA version sounds better, but this is just my personal experience

On 26-7-2017 at 9:57 AM, Fokus said:

 

What do you need measurements for? You can derive a decent guess of the overall impulse response of an MQA chain quite easily with the information that is now publicly available. Assuming that you start from a 8x recording (DXD), concatenate the IRs of the ADC and two of the leaky filters Mansr/Archimago have published. That's about it.

2

I would like to have scientific proof that the impulse response of a recording and DAC improves by a fator of 10 as MQA claims. Some CA members require an MQA encoded test tone for this. Since they are not available, I suggested if such a test-tone is potentially somewhere in an original album present, so a comparison can be made. Is this impossible? 

On 26-7-2017 at 9:57 AM, Fokus said:

Yes. And ironically, if MQA have their way all these systems are obsolete, and MQA have to be paid dues (and royalties) in order for these systems to function again.

2

Both my a NAD C390DD with BluOS module and the portable Pioneer XDP-100K Flac, Wav, DSD and MQA player enable EQ settings. I was quite surprised but it works very well. So why all the concerns that MQA will not allow DSP..? It is a proven fact that it does work on some of the MQA devices out there

On 26-7-2017 at 9:57 AM, Fokus said:

No. You are only interested in whatever confirms your notion that MQA is better, and that this is because of something you call time smear (and which you do not understand, at all).

 

No, this is incorrect. I am fascinated by the sound quality improvement on my home system and what I heard on several other, much higher quality audio systems. I did read many publications on MQA, amongst them one of the better ones in Sound On Sound magazine where the Time Smear is described like this: 

 

Quote

 

The third and final pillar of the MQA system is the format’s radically improved time-domain performance, which really sets it apart from all previous digital audio systems. The company have established a challenging target of delivering a perceptual time smear of about 10µs for existing digital recordings (it’s around 100µs for a conventional 24/192 system), with the aim of achieving an astonishing 3µs for newly recorded material. To put some qualitative numbers on all this, 10µs of time smear is roughly equivalent to that encountered by sound waves passing through 10m of air.   see:  http://www.soundonsound.com/techniques/mqa-time-domain-accuracy-digital-audio-quality

 

 

 

On 26-7-2017 at 9:57 AM, Fokus said:

 

You want people to listen to one note and conclude from this that MQA is better? Or do you want people to look at the trace of a single note???? If the latter: you won't see ringing or blurring. Really.

 

No, as mentioned earlier, I am not an electrical or acoustical engineer... I am a (probably.) overenthusiastic audiophile searching for answers. Since MQA seems to work as an end-to-end system and the end result is in improvement is Time-Smear reduction, this deblurring effect in AIR should be measurable I would say. Or do you believe our ears are so much more sensitive than nowadays acoustic measurements?

On 26-7-2017 at 9:57 AM, Fokus said:

 

Oh, I did. I do have an MQA DAC. I do have listened to it. In many cases the MQA was just as good/bad as the original. In some cases the MQA was better, but clearly from a different master, so the gains would pass equally well through a non-MQA chain.

3

Do you realize that most digital Masters intrinsically contain anti-aliasing errors? Quote from the Sound-On Sound article:  Every digital audio chain involves a cascade series of filters — such as anti-alias, band-limiting (if sample-rate conversion is involved), and reconstruction filters — and the overall system’s frequency response is the product of their individual responses. However, in the time-domain, the overall impulse response of the chain is the convolution of their individual responses

On 26-7-2017 at 9:57 AM, Fokus said:

I know perfectly well what MQA is about.

 

 

I that is true, then you will be able to confirm or deny if the algorithm is capable to reduce the overall impulse response of the chain or not!

[Fitzcaraldo215]

6 minutes ago, firedog said:

So far it seems "deblurring" is  a euphemism  for a filter without pre ringing. You don't need MQA for that.

True, but actually, MQA claims to minimize both pre- and post- ringing.  And, it claims to do so at both the ADC and DAC ends of the chain. 

[PeterV]

On 26-7-2017 at 2:43 PM, The Computer Audiophile said:

@PeterV and @soxr - Any more talk about exposing people's identities and you're banned from CA. It's not funny. 

 

I don't care if someone posts their home address on another site. If they don't want it here, I respect that. 

I understand your point, but got carried away by the post on the subject of upsampling DXD files elsewhere and assumed it was now allowed to quote and relate this. 

[PeterSt]

9 minutes ago, PeterV said:

So why all the concerns that MQA will not allow DSP..? It is a proven fact that it does work on some of the MQA devices out there

 

Because it can't work in software.

Who wants the "DSP" in some hardware device. Maybe Home Theater freaks (and then still they want the wrong thing because that should be done in software as well).

[PeterV]

On 25-7-2017 at 6:51 PM, mcgillroy said:

Oh PeteV got deployed again. Seems like someone at MQA feels the impacts coming closer.

1

 

Funnu, but you are giving me and my potential influence in this thread and group way too much credit. I am just very pleased with the sound quality of MQA and hope the record companies will release many new albums ASAP. It is just a large improvement of my listening pleasure at home, that's all.

On 25-7-2017 at 6:51 PM, mcgillroy said:

There's nothing like boots on the ground spreading FUD, especially if your grunt is a jolly fellow happily absorbing the incoming and keeping everybody busy.

 

 

 

 

 

[PeterSt]

6 minutes ago, PeterV said:

I understand your point, but got carried away by the post on the subject of upsampling DXD files elsewhere and assumed it was now allowed to quote and relate this.

 

I actually don't understand it much, if first someone posts an "article" where Mr xyz soxr announces himself to the world. Maybe Chris missed that (OP) ?

I suppose it is Belgian hoopla to get angry about that and on a tone which is most certainly not Belgian-worthy (says me as the neighbor). 

[PeterSt]

Just now, PeterV said:

Funnu, but you are giving me and my potential influence in this thread and group way too much credit.

 

Peter, something doesn't work here. or is too strange.

You are new here and all you get thrown at you is triple sh*t. Still you stay.

Any normal human being would have left the building by now, but you don't.

 

Why do you care so much ?

Why can't you think along the lines of people who "know" a bit more about it technically, than you (you seem to know nothing about it). Why being persistent "tegen de klippen op" ?

So my hopefully good advice : it doesn't work out like this, so maybe stop posting about this subject and maybe find something you're good at and join those discussions.

Or try to be a tad more neutral. Or act like that. 

[PeterV]

3 minutes ago, PeterSt said:

 

I actually don't understand it much, if first someone posts an "article" where Mr xyz soxr announces himself to the world. Maybe Chris missed that (OP) ?

I suppose it is Belgian hoopla to get angry about that and on a tone which is most certainly not Belgian-worthy (says me as the neighbor). 

If MQA and DSP software are truly incompatible with each other, then I start to understand much better now the hostility ( and fear.) by some DSP manufacturers out there... This was clearly noticeable in the Xivero paper a few Moths ago, where MQA was criticised heavily. When I checked the website, it became clear that Xivero is a well-established manufacturer of DSP software. So MQA is a potential threat to their business, as with companies who manufacture streamers with pitch software.. But who knows, there will be an opportunity in the future? I am curious how MQA will deal with this, since DSP is becoming increasingly popular.  Maybe MQA will allow DSP plugins..? who knows?

[PeterV]

8 minutes ago, PeterSt said:

 

Peter, something doesn't work here. or is too strange.

You are new here and all you get thrown at you is triple sh*t. Still you stay.

Any normal human being would have left the building by now, but you don't.

 

Why do you care so much ?

Why can't you think along the lines of people who "know" a bit more about it technically, than you (you seem to know nothing about it). Why being persistent "tegen de klippen op" ?

So my hopefully good advice : it doesn't work out like this, so maybe stop posting about this subject and maybe find something you're good at and join those discussions.

Or try to be a tad more neutral. Or act like that. 

PeterSt,  To be honest, I do not mind at all if persons react like this.. I seem to be touching a nerve, otherwise, they would ignore me don't you think..?  Action is reaction and I am not the type of guy to walk away. Although interesting points are sometimes raised, I have the impression that most arguments against MQA are business related, not Sound Quality related. I have a lot of respect for the inventors of MQA and I have witnessed fierce personal attacks to Bob Stuart last year. These improper 'hate' reactions intrigue me, since it means that something important is going on. I am probably naive enough to enjoy the vast literature out there written by established journalists in Stereophile, Absolute Sound, Sound Ons Sound, Audiostream Dar_Ko and more. I also do lots of desk research and serahc the patents etc. It just intrigues me and I am not disappointed by the end result at all. So MQA is for my hobby an improvement which I do not want to see lost because of hostile, not sincere non-objective research. What I read here and elsewhere is insufficient proof that MQA is a 'scam' although many would like that it is regarded like that. 

[PeterSt]

16 minutes ago, PeterV said:

So MQA is a potential threat to their business, as with companies who manufacture streamers with pitch software..

 

Peter, you may take it too lightly. I think on this forum the three better players for the highest quality SQ players are around (I deliberately use "player" in two contexts). I don't know about you and whether you use Foobar or possibly even nothing of the software sort, but those players can not work with MQA when a hardware renderer is in order.

And might you be the last of Mohicans using a CD Player, then you are the last one indeed. In other words :

 

MQA is a thread to ALL of us (but you with the CD player). If you are only into the best SQ possible and don't use Foobar for it (and a dozen more).

 

Do you start to understand now ?

Computer Playback these days is about FILTERS. The filters reside in software. They can not be applied when MQA hardware is in order.

 

And so MQA will lose and you will lose your investment in MQA. Everybody will. Even me, supporting it in software.

[Shadders]

1 hour ago, PeterV said:

Your technical analysis totally overlooks the aspect of deblurring, which is the essence of the encoding process. The whole MQA debate should focus much more on this important aspect, since that seems the reason why MQA end-to-end improves the sound quality ( to my ears and opinion and others..)

Hi Peter,

What i was trying to infer is that MQA is not accurate and not lossless.

With regards to the deblurring - this is just a mathematical process - an inverse of the blurring (obvious statement).

So the function that deblurs is a kernel/filter ?.

This kernel/filter is not patentable - it is just a mathematical function which is the inverse of the system that caused the blurring. It will be an approximation too.

So, what you are hearing is neither accurate, nor is it lossless, nor do we need this as a patentable process - surely it is just basic engineering (equalisation is an approximation of what it is ????)

Regards,

Shadders.

[jhwalker]

54 minutes ago, PeterSt said:

 

Peter, something doesn't work here. or is too strange.

You are new here and all you get thrown at you is triple sh*t. Still you stay.

Any normal human being would have left the building by now, but you don't.

 

Why do you care so much ?

Why can't you think along the lines of people who "know" a bit more about it technically, than you (you seem to know nothing about it). Why being persistent "tegen de klippen op" ?

So my hopefully good advice : it doesn't work out like this, so maybe stop posting about this subject and maybe find something you're good at and join those discussions.

Or try to be a tad more neutral. Or act like that. 

 

So anyone who likes MQA must either stop posting or be "more neutral", but those who don't are free to post anything negative they like?  What kind of thinking is that?

 

I'm hoping I missed your point.

[PeterV]

8 minutes ago, Shadders said:

Hi Peter,

What i was trying to infer is that MQA is not accurate and not lossless.

With regards to the deblurring - this is just a mathematical process - an inverse of the blurring (obvious statement).

So the function that deblurs is a kernel/filter ?.

This kernel/filter is not patentable - it is just a mathematical function which is the inverse of the system that caused the blurring. It will be an approximation too.

So, what you are hearing is neither accurate, nor is it lossless, nor do we need this as a patentable process - surely it is just basic engineering (equalisation is an approximation of what it is ????)

Regards,

Shadders.

 

Hi Shadders, thanks for clarifying. Yes, I understand that it is mathematical and I also see your point that it will probably not be a perfect neutralisation of all past errors, but altogether it seems to improve the overall time behaviour of the original recording significantly. The inaudible part is indeed lossy, but this is a sacrifice required to enable streaming the package as you know. So if the original Master intrinsically contains errors, but a new technology is capable of bringing the sound closer to the original, what is then the problem..?  For archiving, MQA might not be the way to go, I do not know. DXD recording seems to be the ultimate we reach nowadays and I suppose blurring is insignificant with such masters... BUT if the DAC itself contributes to an audible distortion, even when playing DXD files and MQA is capable of compensating for these errors, why not use the technology? I suppose uncompressed MQA will exist as well ( would be nice to compare with folded MQA) , and the first MQA certified A/D converters still have to be built.. As long as digital audio is not representing the truth of the sound due to distortions, I prefer to get rid of them before I listen to it. We all notice the raise of high-speed analog recording. The sound of such high quality analog recording beats the best digital recording, MQA follows very close to this some experienced recording experts told me last weekend. I have not been able to witness this yet.

[PeterSt]

Just now, jhwalker said:

I'm hoping I missed your point.

 

John, Yes, you did somewhat. 

Blame my Dutch, spoken to a Dutchman.

 

The gist of it was being helpful and if you read PeterV's response you can see it actually at work. Please keep in mind that "we" are a bit different as all cultures are not alike.

What you maybe also missed (elsewhere) is that I am not against MQA (yet) so go figure.

 

Reading your response, it could be more helpful for people like PeterV to openly support him. The crux is in that part though, because I myself do not see arguments from him to explicitly support (him). That is, I am from the world that needs arguments instead of believing brown eyes. So really, all I think when I read "but it sounds better to me" is some system which coincidentally shows off better with MQA. Just because I am confident this is possible (just like it started to happen for me with the Lush USB cable).

So that does not help me and it does not help PeterV either.

If a 1000 of you start to shout that it sounds better indeed, then this shifts. But sadly I count them on one hand and it includes myself.

 

The biggest problem seems to be that those doing the math and claim that MQA is a sheer hoax, don't listen themselves. At least I can not find them in this forum. Or it could be me myself, but I am useless because I may think that MQA sounds better. And I can not help myself.  Or at least not yet. I must first know why it could be sounding better.

 

I hope this clarifies a bit.

Thanks,

Peter

 

 

[mansr]

1 minute ago, PeterV said:

The inaudible part is indeed lossy

If it is inaudible, why include it all?

[PeterSt]

3 minutes ago, mansr said:

If it is inaudible, why include it all?

 

7 bits of dither.

[PeterV]

5 minutes ago, mansr said:

If it is inaudible, why include it all?

I cannot exprees this correctly with my own wpords or understanding... I can only refer to some of the publications done by MQA why they see the need to do this:  " Modern insights from psychophysics therefore imply we should preserve temporal structure at a finer scale for our ears to take advantage of the original sound; a distribution system that permits end-to-end resolution of 8 μs implies a Gaussian bandwidth of around 44 kHz. A chain is as strong as its weakest link, and one limit on resolving events will be the minimum-phase transducers at each end and amplifiers in between, but as Fig. 1 shows, to achieve this target through a uniform cascade of eight stages, each needs a bandwidth closer to 100 kHz"    see also :  http://www.aes.org/tmpFiles/elib/20170727/18046.pdf 

[PeterV]

56 minutes ago, PeterSt said:

 

Peter, you may take it too lightly. I think on this forum the three better players for the highest quality SQ players are around (I deliberately use "player" in two contexts). I don't know about you and whether you use Foobar or possibly even nothing of the software sort, but those players can not work with MQA when a hardware renderer is in order.

And might you be the last of Mohicans using a CD Player, then you are the last one indeed. In other words :

 

MQA is a thread to ALL of us (but you with the CD player). If you are only into the best SQ possible and don't use Foobar for it (and a dozen more).

 

Do you start to understand now ?

Computer Playback these days is about FILTERS. The filters reside in software. They can not be applied when MQA hardware is in order.

 

And so MQA will lose and you will lose your investment in MQA. Everybody will. Even me, supporting it in software.

Hi Peter,  I tried Foobar about 5 years ago and then switched to Jriver + Jpay platform This 'true'computer streaming adventure was frustrating, since every change and setting and off-the-fly upsampling and playing in hibernated mode was audible.. I stopped with the search when Bluesound and NAD ofered a suitable solution for me.. peace of mind.. Then MQA came and now I am looking at the developments in ROON.... This post might surprise you:  https://community.roonlabs.com/t/eq-and-effects-on-mqa/19481

[Shadders]

9 minutes ago, PeterV said:

 

Hi Shadders, thanks for clarifying. Yes, I understand that it is mathematical and I also see your point that it will probably not be a perfect neutralisation of all past errors, but altogether it seems to improve the overall time behaviour of the original recording significantly. The inaudible part is indeed lossy, but this is a sacrifice required to enable streaming the package as you know. So if the original Master intrinsically contains errors, but a new technology is capable of bringing the sound closer to the original, what is then the problem..?  For archiving, MQA might not be the way to go, I do not know. DXD recording seems to be the ultimate we reach nowadays and I suppose blurring is insignificant with such masters... BUT if the DAC itself contributes to an audible distortion, even when playing DXD files and MQA is capable of compensating for these errors, why not use the technology? I suppose uncompressed MQA will exist as well ( would be nice to compare with folded MQA) , and the first MQA certified A/D converters still have to be built.. As long as digital audio is not representing the truth of the sound due to distortions, I prefer to get rid of them before I listen to it. We all notice the raise of high-speed analog recording. The sound of such high quality analog recording beats the best digital recording, MQA follows very close to this some experienced recording experts told me last weekend. I have not been able to witness this yet.

Hi Peter,

I can see how one might want to remove the blur from previous recordings. What has been reported is that the final DAC stage effect is minimal - so no need to have a MQA approved DAC.

If MQA can reverse the blurring of the audio file, then MQA Ltd could offer a service where they provide a program and requisite file that deblur's a specific CD recording. So once the CD has been parsed - what remains is a CD file that has been deblurred. No need for patents and MQA in the chain. Anyone can implement the program to deblur their CD collection.

If all new recordings are DXD based - no blurring - then MQA is not needed. Final DAC effect is minimal.

Analogue tape - this is just an amplitude modulated high frequency carrier ?. Hmmm, how accurate is that ?

Maybe, the MQA process and Analogue Tape process is inherently inaccurate - and we like this inaccuracy - since our ears/brain like it ?.

All we need is a recoding that has been deblurred to see if that is all we require - and if it proves to be the result we want - then no need for MQA.

Or is it the process of MQA (lossy, aliasing etc) that the ear/brain likes, and deblurring is a red herring ?

Regards,

Shadders.