[Gringo]

 
Woe slow down a minute and let's backup a bit.
 
I've been following high end audio for more years than I care to admit so let's just say at least 30 years. And during all those years I've seen and heard many, many claims about Audio Nirvana finally being reached. Everything from magic clocks to super thick cables and wires to mp3s that sound better than wav files to ultra high resolution digital audio to magic jitter reducers to DSD. And each and every time the claims did not match the reality. That's what is known as marketing.
 
And as I've stated before MQA is being marketed to those who want to believe in the impossible. The impossible in this case is that once something is recorded, either digital or analog, that original recording will always be the best possible source. Sure the original recording can be "improved" by using equalization but in the end it is the original recording that sets the limits on the final sound quality. Add to this the fact that in order for MQA to succeed the superiority of high resolution audio (24bit and 88.2 kHz and above) has to be established as an irrefutable fact. And until the high end audio world can prove this last statement then processes like MQA will also remain unproven. Which is not to say that MQA doesn't do exactly what it claims to do but rather more of so what, at least until that time.
 
It's kind of like all those AC power conditioners - each and every power supply in each and every piece of audio equipment "conditions" the power since they all convert the incoming AC into DC and can easily handle the normal deviations of voltage and frequency.
 
Basically the high end audio game is all about building these complicated belief systems (by they be about cables, power conditioning, bit rates, jitter, etc.) on totally unproven foundations. Kick away the rotten foundation and it all comes crashing down like the house of cards that it actually is.
 
By the way, I will go out on a limb here and state that MQA will never be "officially" subjected to double blind testing. Oh sure, there will be tons of testimony from all the usual high end audio mouthpieces but no real testing. Sighted listening tests have little to no value.

Firstly, I stand by the complaints I posted regarding untrue comments about myself, if by "Woe slow down a minute and lets backup a bit" you are retracting the comment against me then that's all fine.
 
The comments you offer regarding your past experiences, as far as I am concerned, are given in a far more fair and unbiased manner and I completely concur with the vast bulk of what you say and have had many similar experiences.
 
In all my postings I have tried to strike a balance, eg

• someone posts a graph with detailed explanations which I then acknowledge that I am not in a position to counter and will accept unless I discover something better.

2.     The graph and comments are latter in my opinion proved by MQA to be completely incorrect misinterpretations.
3.     The same graph is then posted with an incorrect conclusion again so I posted the detailed MQA rebuffal.  
4.     Another poster then accuses MQA of using terminology not appropriate for a white paper when they knew it wasn't a white paper in the first place and again I challenge that.
 
Yes companies can make false claims, guild the lily etc. but so can the opposite camp.  All I have ever done, where my limited knowledge will allow, is to urge people to offer balanced comment to the best of their ability. If you or I believe a person or company is being biased then responding with equally bias comments is just as reprehensible.
 
If someone proves me wrong, I will be the first to concede the fact
If someone holds different views to mine, then I am happy to simply acknowledge that we disagree
If someone counters my arguments and I am not confident in my ability to refute them, I will stay quiet or acknowledge I am unable to refute them
If someone makes factual inaccurate statements, (deliberate or otherwise and in particular about myself), I reserve the right to say so but will refrain from insults, slurs or the use of abusive language
I don’t believe it putting people or companies on pedestals and that includes MQA
 
If tests, advertising etc, in whatever form they are conducted say one thing but my ears tell me something different over extended multiple auditions then rightly or wrongly I will always go with my ears because they are what I enjoy both live and recorded music with.  That said I believe I don’t have a closed mind and am interested in what others might have to contribute to any given topic
 
When offering my opinion I hope I make it clear it is only my opinion and do not assert it is better than anyone else’s, (I would be one of the first to acknowledge that they are perfectly entitled to their opinion)

 

[gregorio]

From what you state I think it would be more reasonable to say that MQA's claims are still suspect and need more constructive proof than to use the term hyperbol but then that is your perogative.

 
From what I state, I don't think that would be "more reasonable" to say that "MQA's claims are still suspect". From what I've stated, MQA's claims are largely based on incorrect "facts", inaccuracies, misrepresentations, confusion of the issues and double standards. "Hyperbole" is a lenient term to use IMO, for what appears to be deliberate deceit, rather than just exaggeration. Having said this, it is possible that this is just an extremely misguided marketing strategy designed to compete with all the other deliberately deceitful marketing in the audiophile world and that there is actually something of real benefit to MQA. However, I think this is unlikely and my position (and what I've tried to state) is that my concerns about MQA's claims are far more serious than just "still suspect". Likewise, ralphp@optonline's characterisation of what I've stated isn't completely accurate either, as I'm not absolutely certain there's nothing to the "whole MQA thing".
 
My actual position is somewhere between "still suspect" and it's definitely all just BS but from what's been presented so far, I am closer to ralphp's characterisation of what I've stated than to yours. I can completely sympathise with your position though, from your perspective, "from what state" is not even half of it, it's obviously also about what the interviewee has stated and without an expert level of knowledge, how can judge who to trust?
 
G

 

[Joe Bloggs]

They put forward that image, yet their own image shows that the TPDF dithering of CD is below the noise floor of both the MQA AND CD master throughout the spectrum.

Since there are no higher-res versions of 2L-120 track 01 available for comparison, I downloaded the file set for 2L-111 track 15 instead (another track they produced graphs for).

I compared the MQA file noise floor with both that of the CD-quality download provided and that of the 352kHz master manually remastered to CD quality 16/44.1 format via TPDF dither. The results are as follows:



As you can see the TPDF encoded CD format is lower in noise than the MQA file entirely throughout the range.

IN OTHER NEWS, the same 352kHz master, converted to 24/88.2 and saved as FLAC level 8, yielded a 13.1MB file (compared to the MQA FLAC's 15.9MB). The noise spectrum of the 24/88.2 file is included for comparison...
edit: please disregard this last part as it turned out I'd screwed up the project settings for the file export.

 

[sonitus mirus]

 
If tests, advertising etc, in whatever form they are conducted say one thing but my ears tell me something different over extended multiple auditions then rightly or wrongly I will always go with my ears because they are what I enjoy both live and recorded music with.  That said I believe I don’t have a closed mind and am interested in what others might have to contribute to any given topic
 
 

 
This is the fundamental issue that will always be at odds with the other side of these kinds of debates.  If sighted bias or other factors are influencing your opinion, it will continue to do so unless you can definitively show to yourself that you are actually hearing what you think you are hearing.  If it is not possible to conduct a valid ABX test or you simply do not believe in them and you don't understand the mathematics and biology well enough to be swayed or at least be skeptical, these discussions go absolutely nowhere.  Each party leaves in a huff thinking that others are stubborn and ignorant.
 
I think people are listening to wonderfully recorded music and it happens to be in an MQA format.  Someone already tested an MQA file and PCM and there didn't appear to be any significant audible differences shown in a null test.  The noise from MQA seems to be greater than from a CD.  I could take a well-encoded MP3 file and boost the bass and treble a bit and probably get several people to believe this format is better than Red Book if I used marketing magic.  Because MP3 is not proprietary and can be dissected and verified, we would scoff at this idea.  I'm seeing red flags with MQA.  I can't simply listen to MQA and make up my mind, because there is nothing to accurately compare it with until neutral parties have an opportunity to test this format more rigorously. 

 

[Gringo]

Gregorio - thanks for the useful comments

 

[Gringo]

   
This is the fundamental issue that will always be at odds with the other side of these kinds of debates.  If sighted bias or other factors are influencing your opinion, it will continue to do so unless you can definitively show to yourself that you are actually hearing what you think you are hearing.  If it is not possible to conduct a valid ABX test or you simply do not believe in them and you don't understand the mathematics and biology well enough to be swayed or at least be skeptical, these discussions go absolutely nowhere.  Each party leaves in a huff thinking that others are stubborn and ignorant.
 
I think people are listening to wonderfully recorded music and it happens to be in an MQA format.  Someone already tested an MQA file and PCM and there didn't appear to be any significant audible differences shown in a null test.  The noise from MQA seems to be greater than from a CD.  I could take a well-encoded MP3 file and boost the bass and treble a bit and probably get several people to believe this format is better than Red Book if I used marketing magic.  Because MP3 is not proprietary and can be dissected and verified, we would scoff at this idea.  I'm seeing red flags with MQA.  I can't simply listen to MQA and make up my mind, because there is nothing to accurately compare it with until neutral parties have an opportunity to test this format more rigorously. 

Agreed, all sorts of factors can influence us such as suggestions/experiences/prior knowledge/advertising etc etc and I totally agree with the idea of waiting and making your mind up after rigorous testing. How neutral the testing is just like everything else open to debate. I could construe that were you eventually to get to conduct whatever test you wanted that you would not be unbiased, (I hasten to add that I am not for one moment doing so), because of your already expressed concerns  as I will maintain that looking at graphs does not tell us what it sounds like.  I personally am not worried about originals, remasters, equalisation etc. but rather what sounds the best to me which may well be quite different to you. We can agree to differ on my idea of ultimately using your own ears. As far as I am concerned the jury is still out on MQA

 

[sonitus mirus]

  Agreed, all sorts of factors can influence us such as suggestions/experiences/prior knowledge/advertising etc etc and I totally agree with the idea of waiting and making your mind up after rigorous testing. How neutral the testing is just like everything else open to debate. I could construe that were you eventually to get to conduct whatever test you wanted that you would not be unbiased, (I hasten to add that I am not for one moment doing so), because of your already expressed concerns  as I will maintain that looking at graphs does not tell us what it sounds like.  I personally am not worried about originals, remasters, equalisation etc. but rather what sounds the best to me which may well be quite different to you. We can agree to differ on my idea of ultimately using your own ears. As far as I am concerned the jury is still out on MQA

 
I used a mic and software with graphs to set up my speakers.  Otherwise, why not just pour a few glasses of wine and turn the volume up?   I'm only attempting to remove emotion from the equation as best as I reasonably can.  My past experiences are what led me to think this way.  I've shown to myself that I was easily tricked when not appropriately isolating my hearing.

 

[castleofargh]

there is no need to go so far as to dress up as goths to remove emotions from the listening test. we have plenty of examples of people who prefer the lesser sound.
when I enjoy vinyls more, when I enjoy a colored tube amp, when I enjoy +15db bass boost on my headphone, when imagine dragons win music awards with an horribly distorted and saturated song....
the mistake is to assume that if I prefer something, then it's the one with better sound fidelity. this has never ever been a rule.

 

[Gringo]

  there is no need to go so far as to dress up as goths to remove emotions from the listening test. we have plenty of examples of people who prefer the lesser sound.
when I enjoy vinyls more, when I enjoy a colored tube amp, when I enjoy +15db bass boost on my headphone, when imagine dragons win music awards with an horribly distorted and saturated song....
the mistake is to assume that if I prefer something, then it's the one with better sound fidelity. this has never ever been a rule.

I can whole heartedly agree with this.

 

[sonitus mirus]

  there is no need to go so far as to dress up as goths to remove emotions from the listening test. we have plenty of examples of people who prefer the lesser sound.
when I enjoy vinyls more, when I enjoy a colored tube amp, when I enjoy +15db bass boost on my headphone, when imagine dragons win music awards with an horribly distorted and saturated song....
the mistake is to assume that if I prefer something, then it's the one with better sound fidelity. this has never ever been a rule.

 
I chose the word "emotion" poorly.  It was meant to be in a context of something like an unconscious bias.

 

[Joe Bloggs]

I chose the word "emotion" poorly.  It was meant to be in a context of something like an unconscious bias.

Too late, we all know you're a pointy-eared Vulcan now

 

[Gringo]

They put forward that image, yet their own image shows that the TPDF dithering of CD is below the noise floor of both the MQA AND CD master throughout the spectrum.

Since there are no higher-res versions of 2L-120 track 01 available for comparison, I downloaded the file set for 2L-111 track 15 instead (another track they produced graphs for).

I compared the MQA file noise floor with both that of the CD-quality download provided and that of the 352kHz master manually remastered to CD quality 16/44.1 format via TPDF dither. The results are as follows:



As you can see the TPDF encoded CD format is lower in noise than the MQA file entirely throughout the range.

IN OTHER NEWS, the same 352kHz master, converted to 24/88.2 and saved as FLAC level 8, yielded a 13.1MB file (compared to tihe MQA FLAC's 15.9MB). The noise spectrum of the 24/88.2 file is included for comparison...

Maybe I am being completely inept in reading the 1st graph, (which is quite possible), that you refer to but it seems to me that you are drawing inaccurate conclusions.

The noise floor of the track being examined, (not chosen by MQA but rather one used to originally criticise them), shows the undecoded MQA noise floor almost exactly the same as the CD file up to just below 15khz and from there on it is progressively lower. The undecoded MQA channel noise floor signal between 0khz and about 14 kHz is below the reference CD signal noise floor apart from briefly at 9khz.

The decoded MQA channel noise floor signal is at all stages indistinguishable with that of the 24 bit master signal. The CD signal of -120 dB TPDF you refer to is a reference signal and not the CD signal that is being compared to MQA. Note that the MQA noise floor signal is virtually identical to that of the 24 bit TPDF reference signal between 5khz and 20khz

As I think you have misinterpreted that first graph I am offering no comment whatever on your second example. I note that it does not go into similar detailed comparison and does not even offer a legible key of which signal is which.

In the other news comment there is no indication of what level FLAC the MQA signal was so the indicator of file size might not be representative of like with like

 

[Joe Bloggs]

Maybe I am being completely inept in reading the 1st graph, (which is quite possible), that you refer to but it seems to me that you are drawing inaccurate conclusions.

The noise floor of the track being examined, (not chosen by MQA but rather one used to originally criticise them), shows the undecoded MQA noise floor almost exactly the same as the CD file up to just below 15khz and from there on it is progressively lower. The undecoded MQA channel noise floor signal between 0khz and about 14 kHz is below the reference CD signal noise floor apart from briefly at 9khz.

The decoded MQA channel noise floor signal is at all stages indistinguishable with that of the 24 bit master signal. The CD signal of -120 dB TPDF you refer to is a reference signal and not the CD signal that is being compared to MQA. Note that the MQA noise floor signal is virtually identical to that of the 24 bit TPDF reference signal between 5khz and 20khz

The TPDF is the actual digital noise floor of the CD if you use TPDF dither. The reason why the actual CD has higher high frequency noise is probably because it uses noise-shaped dither, which trades higher high freuqency noise for lower low-frequency noise.

But as the TPDF noise floor was shown by their own graph to be lower than the actual noise floor throughout the range, I thought about what would happen if you used TPDF dither and the result is shown in my second graph--in which purple is the CD track downloaded from 2L, green is MQA, red is the CD track I synthesized from the 352kHz master using TPDF dither, and blue is the 24bit/88.2kHz track. (You can see the legend if you click on the picture then click on the link to the original size pic)

 

[sonitus mirus]

Too late, we all know you're a pointy-eared Vulcan now

 
I never completed my Kolinahr discipline training. 

 

[gregorio]

Maybe I am being completely inept in reading the 1st graph...

 
Don't worry about it, that graph is completely all over the place and makes no sense. I didn't even bother addressing it earlier.
 
For example:
 
1. The original recording was on a DAT, which is limited to 16bit. How did it suddenly become 24bit?
2. I have no idea where/how they get those figures for the noise floor on the master. They look bizarre though.
3. The CD noise floor doesn't make any sense either.
 
I can't comment on any of the MQA items on the graph.
 
G