MQA from Meridian

Much is written on MQA with various levels of lucidity .. I think it trries to be too many things to too diverse a target audience and ends up being confused. 

I like the matched construction and reconstruction filter idea... the Authenticated part in MQA if you will... these is a weakness in modern digital audio production and reproduction in my opinion.

But the lossy 'hidef' elements leave my unconvinced.. I am not entirely sure the increasing awareness and understanding of what hidef actually means by many aligns with Meridians... and this is a weakness.. whenever you have made something lossy you have made a concious decision on what to discard... now if some of that info turns out to be useful after all you are stumped...

much, arguably all, of the confusion can be laid at the door of meridian themselves

I run across MQA when Tidal went public last year announcing they will transmit MQA hi res files in the near future. MQA is a brilliant idea as to replace the current obsolete red book cd format with one that will have audible improvement. The guys behind the idea are very smart indeed as they tried to solve the problem of sound quality looking from the average consumer point of view, taking account of the size of the files, transfer rates etc.

So what does MQA offer? First we have to have an understanding on digital filtering process that goes on the music production business due to the limitations of the red book format. These filters have side effects that cause unnatural sound and time smearing in certain frequencies. So through the process of reconstruction of these filtering MQA results to a clearer natural sound. Studies have shown that time smearing effects are far more noticeable from a the human brain even at older age than high frequency response.

Note here that hi end companies like Naim or Chord have their own digital filtering process to address these matters outside of the dac chip used in their application. Most of the times with good results. But for the average consumer without that luxury the results with MQA will be dramatic! 

As far as the lossyness of the format, it has been done in such a way that affects only information above 48K that is inaudible and mainly noise. There is no microphone currently in studios that captures more than 40K frequency response.

vinylrocks posted:

As far as the lossyness of the format, it has been done in such a way that affects only information above 48K that is inaudible and mainly noise. There is no microphone currently in studios that captures more than 40K frequency response. 

But as said several times, and there several research papers published on it - and some not even that new, hi definition is not just about frequency pitch response.... this is one area where I think Meridian have got them selves into a trap - perhaps we will see something called 'MQA Pro' in the future that doesn't rely on lossy hidef info

S

I see your point. Nevertheless, if you come to think about it cd quality that is today's standard is lossy as well since it's sampling rate is rather low and the gaps between the samples have to be bridged. So what would be the new standard now that technology has gone so far? We have to make clear that all new recordings should be made with 24/192 resolution and that it should be available for anyone to buy. But straight uncompressed hi def 24/192 results in such huge files that take a lot of space and couldn't be transmitted in real time through the internet by today's standards. There's a huge market out there for the music industry that doesn't include audiophiles and could be interested in a compromise between convenience and better sound quality. And to whoever says storage space is cheap nowadays I say not in my iPhone it isn't!

For us the most important factor is sound quality and in that respect I hear people in sound demos think they hear better results from MQA against normal 24/192 files due to the filtering reconstruction and the AD to DA authentication process. Of course only time will tell if that's true or just a first impression...

vinylrocks posted:

We have to make clear that all new recordings should be made with 24/192 resolution and that it should be available for anyone to buy.

<calls across lab to Tony Faulkner, one of the leading classical recording engineers in the world>

"Tony, when did you start doing 24/96?"

"1993"

"And 24/192?"

"1998 for the Colin Davis New World Symphony recording -- 176.4 actually (4xFS)"

We've been doing hires/hidef for a *long* time now. 

:-)

vinylrocks posted

For us the most important factor is sound quality and in that respect I hear people in sound demos think they hear better results from MQA against normal 24/192 files due to the filtering reconstruction and the AD to DA authentication process...

Of course it's not reconstruction filtering per se, as all DACs and even technically so do turntables have reconstruction filters.. they key thing is matching the ADC to DAC construction/reconstruction filtering more precisely.

I've certainly read a lot of polarised opinion, that's for sure. There's a hard core out there who seem as determined to be as outraged about it as many of the 'remainer' camp in the 'Brexit' argument. I'm a bit more pragmatic. It is what is and I will either like it or not when it all shakes out. As a long-term Meridian owner I regard it with interest. However, to say that MQA originated in their stable, current Meridian owners are still rather in the dark about M's product development.

From all I've read and to the best of my understanding MQA it's sort of a hybrid encoded format that's not strictly lossless neither lossy as it preserves the dynamics of the master. What seems to be different is actually what Meridian removes from the audio and any processing applied. And that's where I will disagree with Simon - there'll never be a 'MQA Pro' because Meridian have optimised it already, using the highest resolution master as the origin.

Overall it really seems like a more respectable and listenable successor to MP3. Btw I think AAC is really bad, and WMA really good (when encoding with WMP as shipped with Windows).

Other than you can only either be lossless or lossy, there is no inbetween state. My 'pro' variant would remove the lossy elements so there is no loss of information such that all phase and timing info outside of the frequency pitch pass band is maintained. True hidef should not be a function of a physical mastering and reproduction process, it should be, according to growing opinion by some at the AES  be where we don't differentiate from reality irrespective of the mastering or encoding format and process. My point is Meridian have taken a subjective view in their current MQA of what to pragmatically discard  (i.e. Lossy), however continuing research is showing that some of this extended  info might form part of the content our brains use to reconstruct sounds and their meanings in our heads.

Simon-in-Suffolk posted:

Other than you can only either be lossless or lossy, there is no inbetween state. My 'pro' variant would remove the lossy elements so there is no loss of information such that all phase and timing info outside of the frequency pitch pass band is maintained. True hidef should not be a function of a physical mastering and reproduction process, it should be, according to growing opinion by some at the AES  be where we don't differentiate from reality irrespective of the mastering or encoding format and process. My point is Meridian have taken a subjective view in their current MQA of what to pragmatically discard  (i.e. Lossy), however continuing research is showing that some of this extended  info might form part of the content our brains use to reconstruct sounds and their meanings in our heads.

But that defeats the whole purpose of MQA, the raison d'être of MQA i.e. Meridian is promising hi-res quality audio at the file size of around a 16/44 WAV file. Their reduction technology - note I avoid the term compression - is admittedly a black box, but they're offering MQA has a true hi-res experience.

That is the crux of MQA, and to offer a 'Pro' version will defeat its whole purpose completely, no matter how subjective their technology may/may not be. From what I understand they start from the highest resolution master, and then encode it. I repeat, it's a newer, better form of MP3, to put it very simplistically.

If one were to argue that they're simply being subjective or purely reductionist they could have easily reduced the size of the file, I just think the technology has more merit than you're prepare to give it credit for, apart from the prevalent 'authentication' concerns.

Phil Harris has confirmed that the new Uniti range would just need a firmware update for MQA compatibility - so that tells me Naim is open to the format. I'm neither here not there about the format but would certainly like to give it a try before dismissing it completely.

What's the point?

Storage costs are reducing, transport costs are reducing, FLAC gives 2:1 compression, can handle HiDef losslessly and is publicly licensed not proprietary.

Huge posted:

What's the point?

Storage costs are reducing, transport costs are reducing, FLAC gives 2:1 compression, can handle HiDef losslessly and is publicly licensed not proprietary.

It's not just storage, which is considerably less than hi-res FLAC and makes internet streaming feasible, but Meridian are promising outstanding sound quality as well.

I agree that it doesn't look hopeful as in 2 years since it was announced it hasn't gained much traction, but still...

manicm posted:

Huge posted:

What's the point?

Storage costs are reducing, transport costs are reducing, FLAC gives 2:1 compression, can handle HiDef losslessly and is publicly licensed not proprietary.

It's not just storage, which is considerably less than hi-res FLAC and makes internet streaming feasible, but Meridian are promising outstanding sound quality as well.

I agree that it doesn't look hopeful as in 2 years since it was announced it hasn't gained much traction, but still...

I did point out the reduction in transport costs, and that includes internet (VDSL or cable anyone?).

According to the original proponents of MP3, it's virtually indistinguishable from CD.     Yeah, right.  

MQA might find a use for portable devices in a few years time as a replacement for MP3, IF anyone's willing to take on the risk that licence fees may suddenly be introduced at any time.

manicm posted:

Huge posted:

What's the point?

Storage costs are reducing, transport costs are reducing, FLAC gives 2:1 compression, can handle HiDef losslessly and is publicly licensed not proprietary.

It's not just storage, which is considerably less than hi-res FLAC and makes internet streaming feasible, but Meridian are promising outstanding sound quality as well.

I agree that it doesn't look hopeful as in 2 years since it was announced it hasn't gained much traction, but still...

If it would have been that great, it would have gained more traction....agreed....