[jagwap]

Huh. Imagine that. Wonder why that would be,
But this is actually the problem: no master sounds anything like the original performance because it isn't possible, or even intended. That's not what recording/reproducing audio does. The recording/reproduction process strives to achieve a representation of the original that enables the listener to suspend disbelief well enough for the recording to be entertaining, because the reproducing system lacks any ability to reconstruct the original event (or even one similar). Often that representation is of something that never existed as a performance event at all. So it's not possible for any master to sound like the original performance, and that's not what is being attempted. Stating that a post-process can recreate something that goes beyond the abilities of any sound reproducing system is clearly bollox.

Sure, but that wasn't my point. The statement was the master should sound like the master. That does not say anything about quality.

Well, I don't know what MQA "thinks", but that's what they are selling. Perhaps they can do that to a small extent under certain very specific conditions. We still are asking if what they are doing is audible...at all...beyond just the claims...and then it would be nice to know the limitations.

Me too. I'm asking. Some here have already decided.

Too bad they didn't include analog recording in the graph. Or speakers...any speakers. Or rooms (with air in them). All of which show far more of this "blurring" than any digital system, yet aren't being considered. Sort of like saying how much better your vision is when you clean your glasses, but you're looking through a dirty window out into dense fog.

So it is not worth improving a mastering process because analogue tape form 35 years ago wasn't as good as the previous system? Or because speakers have imperfect crossovers, and the sound passes through air? Never mind that headphones exist, shouldn't the mastering process be as innocuous as possible compared to all the other effects? This stuff is usually cumulative.

That one made me smile. Agreed!

 

[bigshot]

I haven't made my mind up about MQA. But I have made my mind up not to assume it is just marketing hype.

Do you have any reason for that beyond subjective impressions?

If a regular DAC can perform this un-smearing function, why do we need a new file format? Why do we need a new file format to "authenticate" masters? Now we're back to the super audible frequencies that The Nightfly doesn't contain. Why do we need MQA for The Nightfly?

 

[bigshot]

One more quick question. Since the specs of a typical 24 track studio recorder don't go beyond 25kHz, what is the advantage to preserving super audible frequencies in an analogue recording?

Now it's been whittled down to the advantage (assuming it's audible) and justification for the MQA format only applies to digital recordings with sampling rates above 96.

 

[Strangelove424]

I agreed the marketing is over the top. I've been in meetings with incompetent marketing people and it makes my blood boil. I have now finally worked with competent marketing people but it has been the minority unfortunately. The problem usually is they end up believing their own BS.

Me too. I wasn’t one of them, but had to work with them. There’s a detachment from reality there, or from the effect of their ideas. My biggest criticism of advertising is the fear aspect. Making people question and fear their audio systems, mp3 files, face wrinkles, thigh thickness, wealth, health, etc. It’s a diseased seed to plant for the sake of profit, and usually afflicts the one who planted it first.

Try the graph I was talking about. I had difficulty linking it as I was on my phone:
https://www.stereophile.com/images/816mqafeature.MQAfig10.jpg
https://www.stereophile.com/images/816mqafeature.MQAfig11.jpg
From https://www.stereophile.com/content/mqa-questions-and-answers-some-real-world-comparisons

They are trying to show that 48kHz and 192kHz is not as good as MQA in this respect, or 5m of air. Is this necessary? I can't say. They think so. You are entitled to believe not. I will wait and see. Until the proof comes, either way, fine.

I won't hold my breath for an ear.

No, no, you cannot wave some Escherian links in my face and sit back and expect my ear. It’s a fine ear, and it won’t be won so half handily.

Let’s make them visible on the thread so that more people can offer their insight, and give everyone a better chance of winning my ear. Let's also make clear that the Stereophile article and all containing words, graphics, and data were contributed by Bob Stuart, owner of MQA.

Graphs compare measurements of like things. The axis show correlations between measurements, and the plotted points indicate precise comparisons of those measurements. A length of one thing can be compared to the length of another thing, one format can be compared against another format. But you cannot compare a codec or frequency rate to a length of space. I’m not trying to be sarcastic or mean, but I can’t believe I even have to write this. You cannot compare a codec or frequency rate to a form of matter, or a length of that form of matter, they are two completely unlike things. And what does MQA or 192kHz sound like separate of air? Can I hear MQA in space? MQA makes sound that travels through air. This is becoming Pythonesque.

Also, another very important thing for graphs… units! Y-Axis is missing them. Neural response in what? How was is measured? Neural response is a biological aspect, was this conducted with an EKG or EEG?

I’m feeling very secure about my ear right now.

It feels like some here are being dismissive for sport. I came here with good intentions of putting the other side, but then I met this lot and it rubbed off. Hence my comment about Luddite juice. I have also spent my career and spare time working in audio. But so have the guys behind MQA. I don't work for, with or near them, but I have watch these guys at Meridian and I've never seen them produce anything that wasn't genuinely intended to advance the art of audio. Also Peter Craven the same, from his work at B&W with Micheal Gerzon (sad loss) and since. Look those guys up. They are audio heavyweights.

I just ask that you have an open and rational mind. I know we can get a little uncouth in here, some more than others, but take note: You were the first to use the word “troll” and not a single person fired it back at you. And I don’t see you as a troll either, so let’s talk about ideas, not people. (Except for the guys, you mentioned, I’ll look them up.)

 

[bigshot]

How does "neural response" relate to me sitting my couch in my living room listening to music?

 

[Strangelove424]

Another important question. Beyond all the faultiness in the data, what is the context of it? What is its meaning and relevance to me as a person? The wildest theories in science, like aspects of relativity theory, made sense long before they were even proven because they had context, and gave things meaning. They explained data, even if the explanation wasn't yet proven. They made sense. None of this MQA stuff has context or makes sense.

 

[castleofargh]

ok I want to play too!!!!!!!

let's do 10 different stuff in 3 different places without telling the consumer what or when for a third of it. including complete remaster but not always to keep the people on edge. because nothing says clear technological superiority and care for fidelity, like deliberate confusion, and making fair side by side comparisons almost impossible.
let's trade everything for time stuff and then talk as if audio was unidimensional. then show how time really is improved. confused as they are, expect people to miss those trades and assume actual fidelity increase.
let's use Dirac pulses in graphs to show an impact that is completely disproportionate compared to actual music or human hearing, so that people will imagine a problem that doesn't exist.
let's offer to improve ADCs with a method DOA that is just as impractical as full DSD recording/mixing/mastering, but before PCM came to the rescue with DXD. and if you like games, look how DSD did it first for pretty much all of this post and how MQA is really just diet DSD.

 

[Strangelove424]

Just noticed the x-axis of the graph has negative units of time. Very psychedelic. I am aware of no measurement that can be made in the past. This is some seriously theoretical territory. Flux capacitor territory. And if you assume "neural response" is in time domain, then both x and y axis represent time, and it gets even more psychedelic. Like a scientific equivalent of an Escher drawing, I am unable to yank my gaze.

Far out.

 

[bigshot]

Hooray for Stereophile and their editorial board!

 

[Sterling2]

Someone posted this earlier on this thread, "But this is actually the problem: no master sounds anything like the original performance because it isn't possible, or even intended. That's not what recording/reproducing audio does."

This statement is opinion, not fact. I have been a producer of Radio Advertising Commercials since 1981. I have indeed produced commercials which could not be distinguished from the live performance, with or without sound reinforcement. With digital recording being perfected in the mid to late 1990's, a recording engineer would need to really screw up to make a master which did not sound like the live performance.

 

[bigshot]

When I'm working sound, I'm always trying to create something that sounds *better* than the live performance. Most rock and pop music is tracked with overdubbing. Recording live is primarily for classical and jazz.

 

[pinnahertz]

Someone posted this earlier on this thread, "But this is actually the problem: no master sounds anything like the original performance because it isn't possible, or even intended. That's not what recording/reproducing audio does."

This statement is opinion, not fact. I have been a producer of Radio Advertising Commercials since 1981. I have indeed produced commercials which could not be distinguished from the live performance, with or without sound reinforcement. With digital recording being perfected in the mid to late 1990's, a recording engineer would need to really screw up to make a master which did not sound like the live performance.

The first 3 chapters of "Sound Reproduction" by Floyd Toole deal with this.

As a producer of radio commercials, are you seriously asking me to believe that your spots comprised of a voiceover with music under and possibly FX sound like the original orchestra, FX and voice over artist in person? Lets just take the VO guy. So, your record him with a mic with non-flat response, possibly proximity effect, and...um... just a tiny bit of compression, right? Sure you do! l Every squash a voice track with compression and boost the LF so he sounds like a movie trailer? Sure you have! Never EQ these tracks either? You have, if they're going to please your client. What about the music bed. Every dip the frequency range of the VO to help it float on top of the music? Never compress the music to keep it stable in the mix? And, the FX...synthetic? Real? If real, were the the actual sound or something that sounds more like it that the actual sound? Like soft wind the the forest sounds nearly identical to distant city traffic if it were held back in the mix.

You're working in a created media world that does not exist in reality. None of it. So, your master doesn't sound like the original acoustic event. Probably even less than a good orchestral recording.

edit: the Toole book deals with the impossibilities in precisely replicating the original acoustic event because of flaws in the entire process, and the rather radical differences between human hearing and the recording process. The best you can do is to create something that is an acceptable illusionary version of the original, adequate to convey the original idea. Recording a voice over artist in a booth with a single microphone, the reproducing that recording in the control room with that same voice panned center between two monitor speakers should pretty much illustrate a very basic version of the problem.

 

[jagwap]

Do you have any reason for that beyond subjective impressions?

Yes. I've stated all of this before earlier in this thread:

The people behind this technology are known for their work in striving to improve audio: Bob Stuart and Peter Craven. I've been to their AES lectures, met them both. I have several colleagues who have worked with then at Meridian and elsewhere, and even disgruntled ex employees cannot say anything against Bob Stuart's technical and practical knowledge. They are both highly thought of in the industry, and as far as I can see only this subject has brought out people without this connection to assume they are charlatans. Sure, there are two camps on apodizing filters, but never with this vitriol.

If a regular DAC can perform this un-smearing function, why do we need a new file format? Why do we need a new file format to "authenticate" masters? Now we're back to the super audible frequencies that The Nightfly doesn't contain. Why do we need MQA for The Nightfly?

Why dilo you think a regular DAC can do this "unsmearing"

My guess is the DAC needs to be an MQA DAC and the file need to be MQA encoded.

 

[bigshot]

I had dinner tonight with a fellow who has been involved with audio for many years working for one of the biggest audio technology companies in the world. He knows the main guys at Meridian personally, so I asked him about MQA. He spoke very highly of Bob Stuart. He said that he's a genius and a straight shooter, but apparently he was burned when his theories about apodizing filters were "appropriated" by other DAC manufacturers. Since then, he has played his cards very close to the vest and won't say exactly what he's selling. He also knows the marketing guy behind MQA too. He described him as a real "character" and said he is probably the source of the more dubious MQA claims, not Stuart.This fella hasn't heard MQA himself, but he believes it probably sounds good just on the basis of Stuart's reputation. I got the impression from him that the main mojo behind MQA is a refinement on the apodizing filter that Stuart has wrapped in secrecy and marketing to prevent other manufacturers from competing with him. He said that Stuart doesn't have any particular interest in streaming. That is just the current thing. He has plans for developing MQA for lossless, and in particular multichannel movie soundtracks. But apparently not necessarily for physical media marketed to consumers. Interesting conversation.

 

[jagwap]

I had dinner tonight with a fellow who has been involved with audio for many years working for one of the biggest audio technology companies in the world. He knows the main guys at Meridian personally, so I asked him about MQA. He spoke very highly of Bob Stuart. He said that he's a genius and a straight shooter, but apparently he was burned when his theories about apodizing filters were "appropriated" by other DAC manufacturers. Since then, he has played his cards very close to the vest and won't say exactly what he's selling. He also knows the marketing guy behind MQA too. He described him as a real "character" and said he is probably the source of the more dubious MQA claims, not Stuart.This fella hasn't heard MQA himself, but he believes it probably sounds good just on the basis of Stuart's reputation. I got the impression from him that the main mojo behind MQA is a refinement on the apodizing filter that Stuart has wrapped in secrecy and marketing to prevent other manufacturers from competing with him. He said that Stuart doesn't have any particular interest in streaming. That is just the current thing. He has plans for developing MQA for lossless, and in particular multichannel movie soundtracks. But apparently not necessarily for physical media marketed to consumers. Interesting conversation.

Sounds about right from my experience and was my original point when I joined this thread: to counter all the comments that Bob Stuart, and by inference Peter Craven were con men.

Peter Craven has said, for the full and best expression of apodizing, the audio should be pre encoded. So this is why I suspect this is what is going on. If you are doing this and you know the previous ADC, you can include its behaviour too. Being secretive should not be entirely surprising.