MQA

[DimitryZ]

I’d agree that the “MQA sound effect” is quite minor, akin to changing a DAC’s upsampling filter.

If it was just a software button you could turn on or off, or an effects box like the m-scalar, I’d have no problem with it.

Instead they replace the original file with their mangled version, which they have the cheek to call a “master”. Then they require you to use licensed hardware to unmangle the file and process it in a certain way that they have deemed to subjectively sound the best. In case you have the temerity to disagree with their interpretation, you are referred to “psychoacoustic research” and/or “what the artist heard/intended”. None of which you can ever verify (of course).

MQA is a genius invention for all the wrong reasons.

You have a lot to learn. You may never, and that's OK - godspeed.

 

[marshanp]

Whereas you have a lot of time to waste...

Never mind. Your grammer (sic) is fine, although your spelling still needs work.

 

[Heckyman]

You have a lot to learn. You may never, and that's OK - godspeed.

Then why don’t you enlighten us by explaining what you believe are the advantages of this new closed system vs the current open one?

 

[DimitryZ]

Whereas you have a lot of time to waste...

Never mind. Your grammer (sic) is fine, although your spelling still needs work.

Thank you for your attention. Please forward your kind corrections. I am sure your kind help will aid diversity on this English-centric forum.

Those few of us who aren't native English speakers will surely and happily accept your certain corrections to our inarticulate mistakes.

 

[DimitryZ]

Then why don’t you enlighten us by explaining what you believe are the advantages of this new closed system vs the current open one?

I have. Wether you hear is up to you.

 

[Southeastern]

I've been reading through lots of historic posts on the forum lately and whenever I read threads which include @DimitryZ and all he ever does is descend into a passive aggressive abusive rhetoric to make his point heard. Its funny like but a bit annoying.

 

[Heckyman]

All I could glean from the last 49 pages @DimitryZ was that you think it sounds ok and that you don’t mind their business model. Anything else?

 

[DimitryZ]

I've been reading through lots of historic posts on the forum lately and whenever I read threads which include @DimitryZ and all he ever does is descend into a passive aggressive abusive rhetoric to make his point heard. Its funny like but a bit annoying.

Well, I guess you have passed your final judgement on me.. I am diminished by your ejudication.

 

[DimitryZ]

All I could glean from the last 49 pages @DimitryZ was that you think it sounds ok and that you don’t mind their business model. Anything else?

Sure. Give it an actual listen. What a crazy audiophile idea, right?

 

[Heckyman]

Sure. Give it an actual listen. What a crazy audiophile idea, right?

I did give it a listen, didn't you read my post?

 

[Timcat]

No, they are mostly correct. There is multi-year churn over the status of MQA as "lossless" or "lossy." The argument revolves around two controversies.

One is wether MQA, purely as a data handling system, can reproduce entirely bit-perfect LPCM file, say 24/96 accros the entire frequency bandwidth. Many audiophiles have a strong belief that any alterations to the original bit-perfect file is unacceptable, since it is the "true source." They are not bothered by the facts that an actual studio tracks undergo multiple non-bit perfect alterations on the way to become "the master." Since MQA is a closed, proprietary, analogue-to-analogue system, it has proven very difficult to ascertain it's exact "purity" status and MQA has been cagey on the subject The current, consensus is that MQA is not bit perfect data handling system in at least ultrasonic part of the spectrum.

The other, equally burning controversy, is the proprietary digital manipulation of the original master to reduce pre-ringing artefacts that are produced in the A/D process. Many or most ADCs used in the studio today, introduce these into the music on transients due to the popularity of linear and minimum phase filters in both recording and playback equipment. MQA claims to have analyzed popular recording chains and to have calculated the amount of pre-ringing that would be present in the "lossless master." They say they apply a reversing process to "null out" this recording artefacts from the original file. In practice, this appears to consist of applying a digital filter from the so called "apodizing" filter family, likely with varying coefficients to "null out" the pre-ringing present in the original file. This filter is also available in most modern DACs, though in a single form. Many audiophiles consider different digital filters used in digital audio to be inaudible (I don't), which suggests that whatever digital reconstruction filtration MQA is going is at best subtle. However, again they object is "doing something" to the "lossless master" which in our hobby is often considered wrong to the point of sacrilege.

Since MQA clearly says they are subtly changing the studio master, in their explanation to reduce the digital artefacts of the recording chain, they are not bit-perfect to the original by definition. They claim to produce most distortion-free analogue reproduction of the original analogue input into the digital recording chain.

This concept has been embraced by some audiophiles who have actually critically listened to it (by no means all), as it can sound remarkably not digital on much material. Others, who mostly never heard it, declared to be horrible and dangerous.

I am slowly beginning to understand (I am not the sharpest knife in the draw), that for many audiophiles, MQA approach runs counter to their belief in the perfection of digital audio and a near holly status of the studio master. They are convinced that once a mastering engineer sealed the two-track downsample for CD or hires release (after it has gone through dozens of DSP engines and effects boxes), there is nothing that can be done to improve the sound to bring it closer to original analogue.

I am listening to Melody Gardot's latest release (she is a very exacting musician) in MQA. I have listened to Qobuz version yesterday. On this type of material (on which MQA excells) it superceeds LPCM.

If MQA are correct in suggesting that their process reduces or reverses digital artefacts introduced during the recording chain, why is it that musicians who care about these things, the prime example being Neil Young, say that they can hear and dislike the distortion to their original work that MQA introduces?

I believe that he said: MQA is the company supplying technology to TIDAL. In their own official descriptions they go into what they did to my original files. They altered them and charge a royalty. I feel that my master files are in no way improved. They are degraded and manipulated. I made them. I know the difference. I can hear it.

 

[Rosewind]

EDITED

All I could glean from the last 49 pages @DimitryZ was that you think it sounds ok and that you don’t mind their business model. Anything else?

Exactly.

I would invite DimitryZ to ask MQA to be allowed to redo Golden One's testing of MQA. You have the dedication and the technical insight, it transpires, to be able to do it correctly according to your own - and MQA - specifications through your playback chain.

Full disclosure to prove that MQA is more than a filter inserted after the music file has been already once authenticated as a clone of the original recording.

If MQA is nothing more than a filter then simply offer it as a filter.

Don't pretend to make MQA out to be more than it really is.

Lossy MQA is a filter process that changes, cuts out, adds, folds and unfolds again. It changes the original non-MQA file to something new. Ask Neil Young.

You think it sounds ok and more than that. Ok. Fine.

You think that because a given lossy MQA file sounds better to you than the same source master in lossless FLAC, we all should leave MQA well alone. Ok. Fine

You approve of MQA's stated aim to be the only format available in the future. Ok. Fine.

You approve of the MQA effort to try to corner music into a locked format that needs licensed software and hardware treatment to be played back as MQA intended. Ok. Fine.

You only care about subjective perceived sound quality. The fact that the original master file is manipulated and altered means nothing to you. Ok. Fine.

This is all. Nothing more, nothing less.

That will be all, I hope.




Post script:

Pleade note that I don't need to be border-line abusive in order to answer you.

 

[Jim Audiomisc]

Really?

You demand peer-reviewed papers from MQA as proof of their goodness.

You routinely accept "webpages" and YouTube as "proof" of their perfidy.

How would you describe this obvious dichotomy?

To me, this is an explicit bias.

Erm. Can you give the number of the posting in this thread where I "demanded peer-reviewed papers from MQA"? I can't actually recall demanding that.

You seem to confuse "proof" with "evidence". This is, admittedly, a common muddle that people make.

I wrote the webpages to try and explain things for people in general. I have also repeatedly since said they should also read the patents and form their own views. You might care to consider also the meaning of the term "peers". Again, it is a term where the actual use in science isn't what people may assume from its use in many cases wrt the House of Lords". Although when one of the Lords uses it to refer to the other Lords it gets closer.

Thus in the context of this forum my 'peers' are all the others reading and contributing to this forum.

Note also that I don't think anything I wrote is 'new' in terms of the science. Methods like undersampling, filters that choose to alias, etc, are well established in other fields. It is an idea uses in varous kinds of receivers, etc. But in many cases they employ other methods to deal with the aliasing so it can be dealt with.

e.g. the use of IQ sampling so you can distinguish +f and -f components, but allow undersampling because you filtered the input to the IQ 'mixer+sampler' to limit the range *either side* of the sample rate. (OK, some RWRs, etc, deliberately fold as they want to take in a wide bandwidth and make detection the priority. Ditto for some hopping-source detectors and other applications in ECM.)

So I simply made my own analysis, argument and conclusions available for others to consider, and expect them to make up their own minds about it. If you think it is twaddle, fine, you can - and have been - making your own views clear. However as I indicated earlier, I don't think you advance your case by adopting "go for the man, not the ball" tactics.

 

[Jim Audiomisc]

(De)blurring is a meaningless made-up thing.

Depends what they/you mean. In principle tweaking the response slightly is feasible. However the snag is along the lines of the old Nasrudin story.

A man walking home one night saw someone on the pavement under a street-light. He stopped and picked him up, thinking he had tripped and fallen over. The man explained that he was looking for the key to his front door which he had dropped.

So he got down with the man and helped look around the pavement. When they still could not find the key he asked the man "What were you doing when you dropped it?"

The man pointed across the road to a house and said, "I was trying to open my front door".

"So why aren't you looking over there?"

"Well, its dark over there. Here I can see."

Thus back in audio: AIUI The MQA system says it aims to tweak the data to deblur the behaviour of the ADCs and digitial chain in the studios. But in reality the microphones used will nearly all have 'ring and die' HF with response that fall away at lower frequencies - and a response that varies with distance to the mic and angle of incidence. Loudspeakers used in the home will also have these sorts of HF responses. etc.

Thus optimising the ADCs data effects does seem to overlook bigger, and more varied, effects elsewhere in the chain from the orginal *sound*.

Of course, you can put that another way: That the sounds of mics and speakers dominate *and this may be what the creators and listeners wanted*. Chances are, though, that except for early digital recordings - which often show many other, more serious, digital defects - tweaking the 'blur' of the ADC may not be that important. However, fair enough if it does help, although it would be hard to know given the other effects.

Added: Occurs to me to suspect that nothing did before, or does now, prevent a studio from doing their own reprocessing to 'deblur' old recordings. To me it seems that they probably may have done this anyway in the past without needing MQA to do it for them.

 

[nostromo]

The more I learn about MQA and what it was initially trying to do, the more perplexed I get.

To me it seems MQA initially wanted a better way to compress music files to aid streaming on limited bandwidth but bandwidth outgrew the need very quickly leaving MQA in the lurch.

Then MQA looked at the ADC issues in studios and DAC issues and decided some DSP could be beneficial.

Somewhere along the line they came up with some grandiose marketing words and a cult was born with the aim of world domination.

 

[Tony L]

The more I learn about MQA and what it was initially trying to do, the more perplexed I get.

Agreed. It strikes me as a product with no current context other than providing a marketing opportunity to lock multiple stages of the recording and music distribution chain into entirely unnecessary licensing fees. As a business model it has similarities to certain lines of clothing for emperors.

 

[mansr]

Depends what they/you mean.

Nobody knows what they mean. Through all the years since they first introduced the format, they have failed to define what this "blurring" is supposed to be, much less explain why their "solution" is needed. The only logical conclusion is that it's something they made up to impress the less technically knowledgable, e.g. the hifi press.

 

[mansr]

The more I learn about MQA and what it was initially trying to do, the more perplexed I get.

To me it seems MQA initially wanted a better way to compress music files to aid streaming on limited bandwidth but bandwidth outgrew the need very quickly leaving MQA in the lurch.

Then MQA looked at the ADC issues in studios and DAC issues and decided some DSP could be beneficial.

Somewhere along the line they came up with some grandiose marketing words and a cult was born with the aim of world domination.

(Here's something I started writing after a few beers a while back.)

The story of MQA begins with Bob waking up one morning and thinking, "how can I make some money?" His hardware business, Meridian, was losing money, and his earlier foray into digital formats, MLP, had had limited success. Clearly, a new approach was needed. The labels control the music, Bob thought, and thus the flow of money. Something to tap into, but how?

What do the labels desire the most? "Control," Bob said to himself, "and that's what I'll sell them." In another word, DRM. An end to the scourge of piracy. Of course, the music-buying public had long ago rejected DRM, so something clever was needed.

DRM is based on cryptography, and besides secrecy, cryptography can also be used to verify authenticity. Discerning music lovers care about provenance, and what better assurance could there be than an authentic signature from the label itself? Bob had found his Trojan horse.

With a plan to conquer both the music labels and the consumers, one market player still remained unexploited, the hardware vendors. How could they be persuaded to contribute to Bob's fortune? The answer, he decided, was to insist that his new format be decoded only within the DAC. This would also be a further incentive for the labels in that DRM coverage would extend all the way to the analogue stage, elegantly preventing copying without losses, just like in the good old days.

 

[Heckyman]

The more I learn about MQA and what it was initially trying to do, the more perplexed I get.

To me it seems MQA initially wanted a better way to compress music files to aid streaming on limited bandwidth but bandwidth outgrew the need very quickly leaving MQA in the lurch.

Then MQA looked at the ADC issues in studios and DAC issues and decided some DSP could be beneficial.

Somewhere along the line they came up with some grandiose marketing words and a cult was born with the aim of world domination.

Well I think the idea was to "right the wrongs" of rippable CDs and DRM-free WAVs (from the music industry perspective) by piggybacking on the "hi-res" / "better than CD quality" streaming bandwagon.

Most people's ears and equipment cannot differentiate between high bit rate lossy, 16/44 WAV and 24/44 or 24/96 (192) so the obvious solution is to do some DSP to make an alternative format more obviously different sounding.

Add some clever but meaningless words such as "masters" "quality" "authenticated" "unfolding" "deblurring" etc. and convince people that different = better. Funny lights on DACs. Get the hifi press on board to provide the “apparatus of justification”.

Now of course, since even lossy streaming is both good enough and affordable enough for most people, there is no longer an incentive to pirate music. Pirates would probably have trouble shifting £2 CDs at car boot sales as everyone already has all the music on their phone / Alexa / in their car etc. Unless there are huge numbers of hidden audiophiles who also pirate music like crazy....

 

[Jim Audiomisc]

FWIW I continue to buy CDs. Just had a set of 10 Duke Ellinington 'Private recordings' CDs arrive from the bloke who runs a good Jazz shop. Second-hand because people can own a CD, not rent access to the content. Before that I bought the 100+ CD set of all the 'Warner' (sic) recordings he made. Real bargain if you like his conducting. Relatively new transfers which fix some of the idiotic flaws on the initial CDs releases, and also some discs if items never before released on CD. Sound good to me.