You would have to remove most of the posts here - particularly those that point to Bob’s marketing claims of “end to end”, ADC fix in MQA, etc. etc.
NOW, starting now, Roon/this forum wants to be “intolerant” of “misinformation” and the circus that is MQA!?
NOW Roon/this forum wants to be “objective” with MQA, and @john is the man to make these decisions!?
Heck I would support this effort as long as it was given due consideration for the history of the MQA/Roon partnership and the 3600 posts in this thread…
John, before the pandemic, I taught logic and argument for a living. I know an ad hominem when I see one. I did not engage in that informal fallacy.
I will state again for emphasis that Roon management participating in the MQA debate presents bad optics. I am all for correcting misinformation. And Wim should be faulted for posting without attribution or context spectrum analysis that had been revised later.
However, Roon staff should stay above the fray. In this case, negative reinforcement – negation of a technical criticism of MQA – makes MQA look better (or less bad) and Roon staff (or staff member) look pro MQA. Again, let independent third parties handle that.
I agree Roon directly should remain impartial or neutral to the thread, remember earlier variations of this company is from the heart of Meridian so they need to tread carefully or they may be seen as influencers of MQA
As an independent third party who understands the MQA algorithm (no NDA of course), you are off on your two points.
MQA is end-to-end in every part of its design and intent. This isn’t possible to the same full extent in all implementations, for example 44.1/16. But it is true throughout the original high resolution encoding algorithm.
And the ADC correction is indeed part of the MQA encoder.
How? What is the engineering/math behind this claim? Objectively, which is to say truthfully, all that can be substantiated by those who are “intolerant of misinformation” is that MQA is a super MP3 - a clever compression algorithm, that most of the time is encoding without any knowledge of orignial ADC at all, then upon decompression applies a slow roll off and leaky out of phase (commonly refered to as “minimum phase”) reconstruction filter that is claimed somehow/someway, against all math/knowledge, corrects for…what, exactly?
You see, I don’t think you understand at all - you just assert the marketing claims of Bob Stuart and MQA Ltd. If you have some knowledge of “MQA algorithm” then please enlighten us how it works, and in what way exactly it is “post-Shannon”. Truly, the Militarly Industrial Complex wants to know!..and a few audiophiles.
Substantiate this claim. Show us the math. As near as anyone can tell it is misinformation
The more I think about it, the more I support this turn of at least one Roon employee towards an objective “intolerance of misinformation” about MQA. @john , I support you! What say you to @robbi_burdeck claims to “understand” the math behind the substantive and extraordinary claims of MQA?
Could you provide some specific examples that rise to the level of posting spectrograms claiming to show the performance of the MQA Decoder when the MQA Decoder wasn’t actually engaged?
You (and others) are invited to share your opinions on MQA’s marketing claims and poke holes where your stud-finder says there’s nothing but sheetrock. I’m not going to take responsibility for fact-checking everything in this thread, I’m not involved in the daily moderation of these forums and I don’t intend to be.
I decided to post in the thread with the intentions of:
A) Clear up any potential confusion about Roon’s performance using the MQA Core Decoder.
B) Make what I felt was a helpful contribution to the discussion happening in this thread.
C) Ensure that these forums remain a place for substantiative and good-faith conversation because it’s in the interest of everyone involved.
I agree that the best “look” for us is to remain neutral, and I think that we do a good job of maintaining that ideal. We want everyone to feel comfortable sharing their opinions and analysis on the forums because it fosters substantiative conversation and gives people a compelling reason to keep visiting. If people felt that Roon staff were cracking down on conversation in an attempt to silence reasonable criticism I think you’d see traffic start to decline and conversations spouting up on other sites regarding this Orwellian stance.
At this point in time I don’t see either of those two things happening. There will always to be people whom disagree decisions we make and how to interpret them, it comes with the territory. At the end of the day I think it’s the users who will hold us accountable, vote with their wallets and feet etc.
I feel that my contributions to this thread were not a violation of what most here would consider “neutrality”, but you are free to disagree and to express that opinion openly. I hope you believe when I say that my actions were done not with the intent to stifle any viewpoints but to clear up a potential misconception about Roon’s performance and to advance the conversation in a meaningful manner.
Crenca,
Since your post shows no real knowledge of MQA that I can see, I’m not sure anything I say will help, or make a difference.
I’ve posted references to the relevant papers and patents, and discussed parts of the math, several times previously in this and other threads here, which you can find if you look. There are a couple of primary, and about 8 secondary, published journal papers by Bob Stuart and Peter Craven that are relevant to MQA, and then two main patents. If you understand those, you can read the blogs at the MQA website to see the overall implementation (without proprietary detail). For those of us who are audio engineers with sufficient background in DSP, MQA is not a mystery. It uses principles known from lossless compression, lossy compression, sampling theory, filtering, and wavelets. It incorporates several new approaches to filtering. It uses the Craven/Gerzon buried data technique. It is a well engineered and sophisticated algorithm.
MQA is not a super MP3, its design has nothing in common with the psychoacoustic ideas in MP3.
Post-Shannon: There is an extensive area in modern sampling theory referred to as finite rate of innovation sampling. It is post-Shannon and developed mostly over the last 15 years. It’s used in other areas but not known well in audio. The MQA filters are based on that. They are short linear phase spline filters and have a second stage flattening filter. The flattening filter is minimum phase and gives a small phase roll-off, but the filters don’t have the phase characteristic or long post ringing that conventional minimum phase filters do.
After nearly a decade down the line no studio is recording mixing mastering or archiving in this format, if it was “god’s gift” to digital audio nirvana everyone would be using it
The only ADC I am aware of to come out was by Mytek this has now been reduced to the digital graveyard since Mytek walked away from the manufactures of their product range this year
Spotify going lossless with PCM (FLAC) will also water down the Hi Res fortunes and the mainstream will either take up 16/44.1 PCM or stay lossy.
Which leaves Tidal the only jewel in the crown left for MQA and this has now been sold so who knows what “Mr Twitter” will do with it
MQA wants to have one file for consumer distribution that’s it ultimate goal, so you have to buy into the MQA Eco system or you are left with second best sound
I will stick with PCM and create a better sound using my own kit and not what someone thinks it should sound like as it cannot be “better” than the studio because if that what the case the studio would archive it that way
The mass conversion of Warners files recently at 16/44.1 proved to me it was just a business venture, I am aware over 7,000 items that were archived at Warners using state of the art technology which also covered for “deblurring” so why would you do it again
jesus roosevelt christ what a tedious thread. listen, just listen, and if you don’t like it, you have options. use them, and move on.
I will say that unless you’ve heard a full hardware decoding, you haven’t really experienced MQA. I applaud Roon’s implementation, but it’s half the solution, and maybe not the most important half.
The problem in this ‘tedious’ thread is trying to separate objective facts from PR gobbledygook.
Also your post, be it unintended, only adds to the confusion and troubles the waters even more. Are you objective? Or are you, as Sr. Director from GraceNote, just talking PR?
Nobody knows, and you do not add a lot of factual clarity either, except for saying ‘unless you’ve heard a full hardware decoding, you haven’t really experienced MQA’.
I believe he is referring to that as further confirmation of MQA’s “proprietary” aspect, which is one of the main reasons for criticism against that format around here. Other than the usual cheaper/costlier DAC options, there is no other format that requires specific equipment to show its intended benefits.
I don’t know. When I Google it I find ‘Entertainment Data technology’. You tell me. Why asking counter questions instead of being up front and clear about it in the first place? You could have started with a disclaimer.
ANYONE having a commercial relationship with whatever audio company, or professionally representing/associated with that company, should post a disclaimer whenever that company is being discussed; this is already required practice in other forums I have been a member of.