[jagwap]

How would anyone objectively demonstrate that a difference could not be heard?  Those claims cannot be definitively proven, and no rational person should expect anyone to provide any proof.  Now, if there is a difference to be heard, THAT can be demonstrated.

I was involved in such a test. The idea was to see if data identical CDs could sound different from different pressing plants. The test was set up by Prism Sound, a pretty proffesional bunch involved with mastering ADCs and pro DACs.

14-15 CDs, looking identical, with only a number on them. I forget the sample size but it was more than a hundred people.

The statistical conclusion was they sound different I remember. I certainly heard a difference.

So it is possible, but non trivial.

 

[jagwap]

So a troll, then.

Well, at least we know.

I wasn't a troll when I starting discussing this. But the naysaying trolls taught me how this thread is navigated.

 

[watchnerd]

I wasn't a troll when I starting discussing this. 

 
But admitting you're a troll now...
 
As I said, at least we know.

 

[jagwap]

But admitting you're a troll now...

As I said, at least we know.

I'll admit something when you open your mind.

 

[jagwap]

But admitting you're a troll now...

As I said, at least we know.

I'm not trolling. In internet parlance I can be said to be "white-knighting" for the MQA side, and people who think they hear audiable differences.

 

[GuyUnder]

The claim that MQA unpacks high-resolution audio in an MQA DAC is true. Wether using Tidal or streaming an MQA file via other software, the blue "MQA studio" lights up. It even works if you transcode the 24-bit FLAC to ALAC. Others with Mytek DACs have reported MQA files showing up as 88 or 96 kHz on the DAC,

I own the 2L Mozart Violin Concerto in MQA. Since this album was mastered in DXD and the DXD version is available for download direct comparsons are possible. The difference between the DXD master and the 44.1 version is pretty obvious, and when playing it via Roon through the Meridian Explorer 2 it's obviously a high resolution version. 2L claims the MQA is the original master resolution, and 2L is a reputable label.

What I'm really interested in is the promise of advanced psychoacoustic features like corrections to studio recordings and DAC-specific tailoring.

 

[RRod]

I own the 2L Mozart Violin Concerto in MQA. Since this album was mastered in DXD and the DXD version is available for download direct comparsons are possible. The difference between the DXD master and the 44.1 version is pretty obvious, and when playing it via Roon through the Meridian Explorer 2 it's obviously a high resolution version. 2L claims the MQA is the original master resolution, and 2L is a reputable label.
 

 
2L is also the biggest format whore in the classical business (good sounding stuff, though). I tried once to get their test-bench downsamples to match up with what I could make in SoX but I couldn't, which means that they're doing something special (non-linear-phase filter etc.) for the conversion. This is as opposed to say BIS whose hi-res samples I can get to null-out except for dither. So again, companies just bound and determine to obfuscate by pouring special sauce on everything.

 

[asymcon]

2L, while thinking of every possible technical detail in their rec sessions, thousands spent on DXD multitracks and vintage ribbon mics, positioning musicians in various ways to get their Aura 9.1 experience, still manage to sound somewhat boring and repetitive. I bought their Magnificat. While genuinely good music, it doesn't hold up after multiple listenings. 
Their mix is just so perfect, it's boring. 

 
 
@RRod Can you try to null 2L's MQA vs. downsampled 96/24? I had difficulties with it, as MQA doesn't seem to sync properly to the original. 

 

[Joe Bloggs]

The claim that MQA unpacks high-resolution audio in an MQA DAC is true. Wether using Tidal or streaming an MQA file via other software, the blue "MQA studio" lights up. It even works if you transcode the 24-bit FLAC to ALAC. Others with Mytek DACs have reported MQA files showing up as 88 or 96 kHz on the DAC,

I own the 2L Mozart Violin Concerto in MQA. Since this album was mastered in DXD and the DXD version is available for download direct comparsons are possible. The difference between the DXD master and the 44.1 version is pretty obvious, and when playing it via Roon through the Meridian Explorer 2 it's obviously a high resolution version. 2L claims the MQA is the original master resolution, and 2L is a reputable label.

 

2L is also the biggest format whore in the classical business (good sounding stuff, though). I tried once to get their test-bench downsamples to match up with what I could make in SoX but I couldn't, which means that they're doing something special (non-linear-phase filter etc.) for doing the conversion. This is as opposed to say BIS whose hi-res samples I can get to null-out except for dither. So again, companies just bound and determine to obfuscate by pouring special sauce on everything.

I think the salient problem here, rather than what 2L does or doesn't do with their MQA masters, is a breakdown in logical reasoning. A light lights up (simply indicating that an MQA stream has been detected) or a format indicator changes (simply indicating that the input is now formatted at the new sample rate; any old resampling process can cause the same) are taken as evidence that "MQA unpacks high-resolution audio".

DXD is taken and compared against redbook. Putting aside any possible problems with the comparison methodology (starting with the fact that the comparison is sighted), quite what this has to do with an MQA vs Redbook comparison totally escapes me. The 2L labelling of "original resolution" is obviously false--even the MQA decoded stream is formatted at "88 or 96kHz" as he himself states, not 352kHz. Again, the stream being formatted at a certain sample rate shouldn't be taken as evidence that it contains the original resolution of that rate, but if even the rate *labelling* is lower than 352kHz...

Of course it is technically true that MQA unpacks some audio beyond the nominal resolution of the format the FLAC file is in, but the provided proof... are irrelevant.

 

[asymcon]

A light lights up (simply indicating that an MQA stream has been detected) or a format indicator changes (simply indicating that the input is now formatted at the new sample rate; any old resampling process can cause the same) are taken as evidence that "MQA unpacks high-resolution audio".

And how this visual indicator skews subjectivist's judgement over said recording, regardless of any audible differences. 

 

[headfry]

These critiques remind me of the angry, self-satisfied posts about SQ differences of digital cables,
and how a $20 Belkin USB is all you need, after all 0's and 1's travel down the cable the same way
as through a Siltech Anniversary USB orWireworld Platinum.. and that the expensive USB cables in THEIR
OPINION were in no way significantly better sounding and certainly way overpriced snake oil .

Well in my opinion the above argument is nothing but self-righteous, arrogant ignorance - (of course some
higher priced digital cables can give significant SQ improvements in a quality setup) - all done
with the same attitude by several posters on MQA being no better sounding and just looking
to discredit what many hear as a major advance in music streaming delivery SQ.....
which I hear as much better than the non MQA FLAC versions on Tidal and on par with other hi-res formats- perhaps significantly better.

I don't need a DBT to know that MQA/Tidal Masters in general sound obviously much better than their non-MQA
counterparts - myMojo and music have never sounded this good before through Tidal. And this is without DAC MQA decoding,
which I expect to be even better!

Like I said a few days ago.... the verdict on MQA will be overwhelmingly positive -
it's already starting.

 

[pinnahertz]

I was involved in such a test. The idea was to see if data identical CDs could sound different from different pressing plants. The test was set up by Prism Sound, a pretty proffesional bunch involved with mastering ADCs and pro DACs.

14-15 CDs, looking identical, with only a number on them. I forget the sample size but it was more than a hundred people.

The statistical conclusion was they sound different I remember. I certainly heard a difference.

So it is possible, but non trivial.

No, that is your opinion, and was not the conclusion.  The conclusion of resulting paper clearly states, "Listening tests have so far failed to produce convincing evidence for consistent sonic differences among the TD-2 disc sets.", and goes on to explain possible reasons for individual experts opinions that they hear differences.  But the data did not support theory that CDs from different plants repeatably sound different. The paper is actually fairly well done, though the comparison method was hampered by the inability to instantly compare samples, thus affecting final resolution.
 
Also, that was in 1996. One might expect a few refinements in both players and pressing plants in the intervening years.  
 
The paper can be found here.

 

[jagwap]

  No, that is your opinion, and was not the conclusion.  The conclusion of resulting paper clearly states, "Listening tests have so far failed to produce convincing evidence for consistent sonic differences among the TD-2 disc sets.", and goes on to explain possible reasons for individual experts opinions that they hear differences.  But the data did not support theory that CDs from different plants repeatably sound different. The paper is actually fairly well done, though the comparison method was hampered by the inability to instantly compare samples, thus affecting final resolution.
 
Also, that was in 1996. One might expect a few refinements in both players and pressing plants in the intervening years.  
 
The paper can be found here.

My appologies.  I remember the result incorrectly.

 

[Joe Bloggs]

These critiques remind me of the angry, self-satisfied posts about SQ differences of digital cables,
and how a $20 Belkin USB is all you need, after all 0's and 1's travel down the cable the same way
as through a Siltech Anniversary USB orWireworld Platinum.. and that the expensive USB cables in THEIR
OPINION were in no way significantly better sounding and certainly way overpriced snake oil .

Well in my opinion the above argument is nothing but self-righteous, arrogant ignorance - (of course some
higher priced digital cables can give significant SQ improvements in a quality setup) - all done
with the same attitude by several posters on MQA being no better sounding and just looking
to discredit what many hear as a major advance in music streaming delivery SQ.....
which I hear as much better than the non MQA FLAC versions on Tidal and on par with other hi-res formats- perhaps significantly better.

I don't need a DBT to know that MQA/Tidal Masters in general sound obviously much better than their non-MQA
counterparts - myMojo and music have never sounded this good before through Tidal. And this is without DAC MQA decoding,
which I expect to be even better!

Like I said a few days ago.... the verdict on MQA will be overwhelmingly positive -
it's already starting.

Well... I get that it is a popular audiophile opinion that "everything matters" when it comes to sound quality, including which way the wind happened to be blowing at the power plant powering your listening room from a hundred miles away on the day you are listening, to say nothing of whether said power plant happens to be coal-fired, nuclear or hydroelectric (coal-plant powered audio rigs sound warmer while hydroelectric powered rigs sound more liquid, I've been told). Fine. Suppose you are correct. I have the business interests of the company I represent to look after and can only go so far at alienating potential future customers, even in unrelated chitchat.

What I don't get is, why this "everything matters" mentality never extends to the very things that even pro audio practitians agree *do* make huge differences. Compare:

1. The difference in signal between that presented by regular CD audio and that presented by MQA can be quantified in terms of distortion figures at something like 0.000x% THD. If the decoded continuous waveforms of CD and MQA were plotted on top of each other using 0.1mm pencil lead with a 1 meter axis height, you'll still probably have a hard time finding any point where the curves do not completely overlap to the naked eye.

2. The difference between whatever waveform is specified by CD or MQA, and that actually produced by whatever headphones you own, look *nothing alike*. Gross frequency response deviations occur in the audible band to the tune of 10s of dBs, warping the resulting waveform beyond anything but the crudest recognition. Yet any time some technically minded audio enthusiast suggests using equalization to combat these distortions, the most probable result is half the community dismissing him as a green amateur unschooled in the mysterious ways in which EQ will "further degrade the signal", making it "only a bandaid suitable for the worst recordings" or making said "amateur's" "pathetic system sound even worse".

3. This is to say nothing of the fact that audio on headphones sound nothing like the sound on the loudspeakers that most music were mastered for. A few technically oriented companies tout niche HRTF simulation solutions that attempt to compute the way each sound bounces around in a real listening room and enter BOTH the listener's ears with complex frequency and phase relationships, yet the average headphone audiophile is again content to shell out sums of money well in excess of that which could buy him such solutions, to (again) buy more expensive cables and hi-res audio equipment. The difference such HRTF simulation makes goes beyond "huge" and borders on the "infinite": a plain headphone system receiving signal on the left channel will produce no signal whatsoever on the right channel (leading to a classic "left right and centre blobs in your head" soundstage, whereas a HRTF-enabled headphone system will produce delayed, attenuated sound of meticulously computed phase on the right channel to simulate the effect of a left loudspeaker going around your head to reach your right ear. And yet...

Moreover, audiophile-approved solutions get a free "if at first you don't succeed..." pass if one doesn't hear a positive change the first time round: the solution is to buy more and more expensive / different pieces of kit until one finally notices the difference. OTOH, said audiophile will literally give an EQ all of one minute of screen time, throw a few sliders at random, and, if the sound does not change for the better immediately, forever, forever consign EQ to a bin of "perpetual failures".

/rant

 

[Steven Stone1]

TO add actual information to this thread here is a seminar hosted by AES that was held last night in Canada - https://www.facebook.com/AESmontreal/videos/1174201429344035/ - It features Bob Stuart explaining MQA - my fave quote is"The goal of MQA was to add no more distortion to the digital process than one meter of air adds to an analog signal."