MQA discussion

[awsmone]

I am not sure of your point

Any high sample rate file will have superior time domain information to a lower sample rate file

This is not MQA specific

My understanding of MQA deblurring was an understanding of the original ADC process in particular antialiasing filter ringing

In order to improve frequency response/time domain you would need to recover information
It's possible they could recover information buried above the noise floor, or using predictive or sparse sampling to do this

However the error would increase with the residual noise in the file

Interesting stuff though for older, but high quality recordings

 

[microstrip]

Spirit

This could be a case of mastering. Not of intrinsic format sound. The same apply to all the MQA kudos. Alow me skepticism

It is why having both versions and performing spectral analysts of them is needed before commenting on formats, unless we can be sure that, for example, both come from the same DXD master.

JA was very clear in his Stereophile article:

"My analyses comparing the spectrum of the original WAV file with that of the decoded MQA version did indeed prove that MQA's "music origami" worked, the spectra of the original WAV file and the decoded MQA version overlaying one another exactly up to the 44.1kHz Nyquist frequency of the original recording."
Read more at https://www.stereophile.com/content/listening-mqa#6KG78IjAoWXik7x5.99

 

[KeithR]

My biggest objection, and what should be a problem for anyone, is simple. There is no technical reason why the MQA process needs to be lossy, and no technical reason why all the filtering and DSP mumbo-jumbo needs to be proprietary (except for Meridian to make money). All the audio goodness could be put into a lossless container with appropriate software (not hardware or firmware) at the other (user's) end. But then that would end up being hacked or open-source, and Meridian can't have that...

Rob, I think its because streaming hi res file sizes is a non-starter for anyone. And if hi res can't be streamed from Tidal, Spotify, etc. then the format is doomed. We already have hi res FLAC and its a dead market. Streaming has increased from 15M to 100M subs in two years.

I think Peter McGrath's (and JA's) demo playing hi res FLAC vs. MQA is promising despite the lossy concerns but I share your skepticism on the "A/D secret sauce." I'm also encouraged that Ludwig and Massenburg are onboard.

 

[rbbert]

Rob, I think its because streaming hi res file sizes is a non-starter for anyone. And if hi res can't be streamed from Tidal, Spotify, etc. then the format is doomed. We already have hi res FLAC and its a dead market. Streaming has increased from 15M to 100M subs in two years.

I think Peter McGrath's (and JA's) demo playing hi res FLAC vs. MQA is promising despite the lossy concerns but I share your skepticism on the "A/D secret sauce." I'm also encouraged that Ludwig and Massenburg are onboard.

I understand about the potential importance of streaming, but I am concerned about markedly poor artist income from streaming. Regardless, the fact that the lossy compression and the MQA "secret sauce" are inextricably linked is very disturbing in terms of benefit to the consumer

 

[rbbert]

IMHO the most interesting aspect of MQA for audiophiles is what they call "correcting for time domain errors", not the compression. However, does this mean we are paying for extremely expensive clocks for our digital systems and then a Meridian calculated "clock distortion" is being superposed over them to improve subjective sound quality?

As mentioned, Meridian has claimed both correction for time domain errors (which almost certainly means some phase error correction as well) and correction for (any) errors introduced by the ADC. However, these "corrections" are supposedly made by ear by the Meridian engineers, no actual measurements involved (per Bob Stuart at at least one of the many audio shows where he has had a Q & A). In any case, this is why I made the statements in my original post about why this should not be proprietary; if prominent audio reviewers and other industry manufacturers had been more emphatic about this from the start, maybe by now it wouldn't be proprietary.

 

[ack]

It is why having both versions and performing spectral analysts of them is needed before commenting on formats, unless we can be sure that, for example, both come from the same DXD master.

JA was very clear in his Stereophile article:

"My analyses comparing the spectrum of the original WAV file with that of the decoded MQA version did indeed prove that MQA's "music origami" worked, the spectra of the original WAV file and the decoded MQA version overlaying one another exactly up to the 44.1kHz Nyquist frequency of the original recording."
Read more at https://www.stereophile.com/content/listening-mqa#6KG78IjAoWXik7x5.99

I personally never believed the claims - and what's written on wikipedia - that MQA is a lossy algorithm; instead, I continue to be skeptical about it "fixing" the ADC and DAC; JA more eloquently says the following in the link you posted, which is what TAS, other sources and Meridian have also said:

MQA is claimed to be able to compensate for the time-domain errors of both the original A/D converter used to make a recording and the D/A converter used to play it back. This results in the complete recording/playback chain having an impulse response equivalent to a few feet of air, and temporal resolution of the same form and order as that of the temporal sensitivity of the ear-brain.

I just can't see how it can possibly and *accurately* undo any errors in the original recording A/D, never mind the reproducing D/A, especially errors in the time domain. It reads nothing less than magic.

 

[KeithR]

I understand about the potential importance of streaming, but I am concerned about markedly poor artist income from streaming. Regardless, the fact that the lossy compression and the MQA "secret sauce" are inextricably linked is very disturbing in terms of benefit to the consumer

I think artist income is more related to how listening habits have changed then streaming itself.

 

[still-one]

I personally never believed the claims - and what's written on wikipedia - that MQA is a lossy algorithm; instead, I continue to be skeptical about it "fixing" the ADC and DAC; JA more eloquently says the following in the link you posted, which is what TAS, other sources and Meridian have also said:



I just can't see how it can possibly and *accurately* undo any errors in the original recording A/D, never mind the reproducing D/A, especially errors in the time domain. It reads nothing less than magic.

It must be really difficult for you to enjoy a hobby when you don't trust your ears. Now back to listening to music.

 

[Al M.]

It must be really difficult for you to enjoy a hobby when you don't trust your ears. Now back to listening to music.

What does wanting to have a technical explanation have to do with trusting your ears or not? I don't see a connection here.

I just can't see how it can possibly and *accurately* undo any errors in the original recording A/D, never mind the reproducing D/A, especially errors in the time domain. It reads nothing less than magic.

+ 1

 

[rbbert]

I think artist income is more related to how listening habits have changed then streaming itself.

Getting a little off topic here, but any number of artists have said they now make singles/albums to get people to come to their concerts, which is where they make money (and one reason for ticket price creep). This is the opposite of what was true for decades, where artists would tour to support albums rather than the reverse.

 

[ack]

What does wanting to have a technical explanation have to do with trusting your ears or not? I don't see a connection here.

Ignore him - he's clueless about what's being discussed. Now back to "magic" - how can an algorithm possibly and *accurately* fix time domain issues in the recording D/A, starting with possible time-domain errors in its analog section, never mind possible time-domain errors in the digital section of the D/A.

 

[KeithR]

Getting a little off topic here, but any number of artists have said they now make singles/albums to get people to come to their concerts, which is where they make money (and one reason for ticket price creep). This is the opposite of what was true for decades, where artists would tour to support albums rather than the reverse.

of course- and its because people download (or stream) a song or two, not an album. that's what i meant by listening habits. ironically this is also why MQA can succeed alongside 16/44 as it doesn't require people to go drop $25 on a download.

it seems that artists are now using EPs more and that albums in general are shorter than ever. they also are feeding singles out way, way earlier than prior decades. how this weaves with music distribution is going to be interesting to watch.

 

[KeithR]

Ignore him - he's clueless about what's being discussed. Now back to "magic" - how can an algorithm possibly and *accurately* fix time domain issues in the recording D/A, starting with possible time-domain errors in its analog section, never mind possible time-domain errors in the digital section of the D/A.

if George Massenburg and Bob Ludwig are on board technically-speaking, doesn't this say something about the idea?

 

[ack]

if George Massenburg and Bob Ludwig are on board technically-speaking, doesn't this say something about the idea?

It says something, but nothing concrete about the technical aspects I (and similar ones by others) raised - specifically, it says it's "different", not necessarily "better" than the original. I am going to make an effort to listen to this alleged "magic" at Goodwin's with the Berkeley Ref2+MQA soon, but I must admit, as a technical guy, skepticism is all over me.

 

[Steve Williams]

To get back on topic re MQA........

It might be the next best thing to sliced bread but here is where I have a personal problem

At one of the audio shows last year when MQA was getting a bit of a foot hold and some rooms were demoing it, I happened to get into a room early in the morning where there was no traffic and I was the only one there.

I sat in the sweet spot and was given the MQA demo which I liked, however I immediately asked the man who was giving the demo that I'd like to hear the same songs played simple 16/44.1 PCM nd he refused. I was stunned. I asked how else can I compare it to simple redbook and know what the differences are. His comment to me is that MQA forbids it

That's when I personally lost all interest in mQA

 

[Al M.]

To get back on topic re MQA........

It might be the next best thing to sliced bread but here is where I have a personal problem

At one of the audio shows last year when MQA was getting a bit of a foot hold and some rooms were demoing it, I happened to get into a room early in the morning where there was no traffic and I was the only one there.

I sat in the sweet spot and was given the MQA demo which I liked, however I immediately asked the man who was giving the demo that I'd like to hear the same songs played simple 16/44.1 PCM nd he refused. I was stunned. I asked how else can I compare it to simple redbook and know what the differences are. His comment to me is that MQA forbids it

That's when I personally lost all interest in MQA

Precisely also because of that somewhat 'fishy' aura about MQA some of us want to know more about certain technical aspects, apart from these aspects being interesting in themselves.

In this context, the following article seems to be interesting:

http://www.digitalaudioreview.net/2017/07/kih-46-mqas-missing-link/

 

[still-one]

Ignore him - he's clueless about what's being discussed. Now back to "magic" - how can an algorithm possibly and *accurately* fix time domain issues in the recording D/A, starting with possible time-domain errors in its analog section, never mind possible time-domain errors in the digital section of the D/A.

Admit it. you just like hearing your self talk. Another typical WPF thread for you.

 

[still-one]

To get back on topic re MQA........

It might be the next best thing to sliced bread but here is where I have a personal problem

At one of the audio shows last year when MQA was getting a bit of a foot hold and some rooms were demoing it, I happened to get into a room early in the morning where there was no traffic and I was the only one there.

I sat in the sweet spot and was given the MQA demo which I liked, however I immediately asked the man who was giving the demo that I'd like to hear the same songs played simple 16/44.1 PCM nd he refused. I was stunned. I asked how else can I compare it to simple redbook and know what the differences are. His comment to me is that MQA forbids it

That's when I personally lost all interest in MQA

Thats funny Steve. At last years RMAF they did perform exactly those comparisons.

 

[Steve Williams]

Go figure

 

[KeithR]

Thats funny Steve. At last years RMAF they did perform exactly those comparisons.

and at LAAS this year. i do think the first show MQA was introduced they only played MQA files though.

better question - have we seen a well-respected source or reviewer say after comparisons were done, at any show or home, that he preferred the redbook to the MQA versions?