Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[-3m-]

http://www.musicischanging.com/

Anyone hear about Meridian MQA? It seem decode by software. Is it available in MC?

[JimH]

I think it may be just released.

[Vincent Kars]

As far as I know, you can't by any recording is this format yet.

[birgitta]

the best sounding format, size like 16-bits FLAC file. Foobar can handle it right now. Its free.  You can read about it in www.hifi-musik.se translate w google. a revolution

br
Birgitta

[birgitta]

it sounds better than 24/192 

[Arindelle]

Thanks OP never heard of this until today.

the best sounding format, size like 16-bits FLAC file. Foobar can handle it right now. Its free.  You can read about it in www.hifi-musik.se translate w google. a revolution

br
Birgitta

where have you had a chance to listen to this? If you have a link I'd appreciate it.


This smells of big money, though at least FLAC is supported -- just hope this is not a cartel push by the remaining 3 majors ... I'm sure this will all be open source and Meridien will not be licensing this to block independant musicians and producers


That being said if the "industry" can agree to define what they are selling like the following quote, smaller file sizes for  equivalent audio quality could be interesting ... it reads like it would be better though

The Digital Entertainment Group, the Consumer Electronics Association and The Recording Academy have teamed up with Sony, Universal and Warner Music Group to come up with a standard definition for high resolution audio: "Lossless audio that is capable of reproducing the full range of sound from recordings that have been mastered from better than CD-quality music sources.

"Four different file formats have also been agreed: MQ-P; MQ-A; MQ-C; and MQ-D. "MQ" stands for "Master Quality", meaning that the file is produced from a digital or analogue master recording. The file types are as follows:

MQ-P: from a PCM master source 48kHz/20-bit or higher (typically 96/24 or 192/24 content).
MQ-A: from an analogue master source.
MQ-C: from a CD master source (44.1kHz/16-bit).
MQ-D: from a DSD/DSF master source (typically 2.8MHz or 5.6MHz content).
Although the Master Quality file types and definition of hi-res audio are voluntary, the hope is that with major players backing them, they'll come to be a de facto standard.

might be less snake oil hi-rez recordings floating about than now.

quoted from a simplified but interesting article here http://www.stuff.tv/music/why-it-s-time-get-pumped-about-hi-res-audio/feature which was linked from a blurb about MQA here http://www.stuff.tv/meridian/meridian-s-mqa-format-allows-streaming-studio-quality-music/news

PS- from the above suggested definitions what is the rationale for an MQ-A and an MQ-C? (nothing to do with Meridiens "MQA" encoding decoding BTW so off topic) to insure authenticity? if thats what they mean ok, if they mean sound improvement euh I'm sceptical.

[mwillems]

PS- from the above suggested definitions what is the rationale for an MQ-A and an MQ-C? (nothing to do with Meridiens "MQA" encoding decoding BTW so off topic) to insure authenticity? if thats what they mean ok, if they mean sound improvement euh I'm sceptical.

I think it's a "truth in advertising" type issue.  When you're starting from an analog master, bitdepth and sampling rate are kind of artificial/unhelpful measurements.  You can use 192/24 bit equipment when transferring an analog master, but if the original tapes had a 60 dB SNR (an effective 10-bit bitdepth, not uncommon with magnetic tape) it would be deceptive to claim 24-bit bitdepth.  And the reality is that most available analog sources have (comparatively) low effective bitdepths due to hiss/static, and those bitdepths are not fixed (the same tape may have hiss at -60dBFS in some parts and at -70dBFS at others, etc.).  My understanding is that the best available analog tape has an effective SNR in the 90's of dB with acceptable distortion, so about like a redbook CD (i.e. 16 bit)

In my view any attempt to assert the actual resolution of a transfer from an analog source (As opposed to the resolution of the equipment used in the transfer) will be at best an approximation and (at worst) a trick.  

Some more info:
https://en.wikipedia.org/wiki/Comparison_of_analog_and_digital_recording#Dynamic_range

And see the section titled "The Noise Floor, and How It Effects Dynamic Range" in this article:
http://www.sonicscoop.com/2013/08/29/why-almost-everything-you-thought-you-knew-about-bit-depth-is-probably-wrong/

[Arindelle]

Some more info:
https://en.wikipedia.org/wiki/Comparison_of_analog_and_digital_recording#Dynamic_range

And see the section titled "The Noise Floor, and How It Effects Dynamic Range" in this article:
http://www.sonicscoop.com/2013/08/29/why-almost-everything-you-thought-you-knew-about-bit-depth-is-probably-wrong/

hey thanks excellent articles and the first a fun read .. I like that 160-180db dynamic range = death 

Now if upsampling vs nos dacs had a simple explanation of both pros and cons I'd be up for it

[flac.rules]

I am skeptical, the website has no technical info, the video there is full of what i would consider to at the very best, dubious claims (cow dung was the first thing that sprung to mind).

The only thing that will considerably improve music from a format point of view is multichannel vs stereo.

[InflatableMouse]

There is absolutely nothing wrong with digital in its current form. We don't even need hires. What we do need are properly mastered and produced albums instead of the overly compressed crap that's been coming out since the early 90's or so.

That graph, it made me cry and laugh at the same time, but I guess it shows the audience they are targetting; Analogue fanboys. LP's offer better quality than CD, DVD, DVD-A and even SACD ... Seriously now? Who would agree with that?

[Arindelle]

There is absolutely nothing wrong with digital in its current form. We don't even need hires. What we do need are properly mastered and produced albums instead of the overly compressed crap that's been coming out since the early 90's or so.

I agree with you. Its like the bit perfect syndrome - if its a bit perfect crap remaster.

That graph, it made me cry and laugh at the same time, but I guess it shows the audience they are targetting; Analogue fanboys. LP's offer better quality than CD, DVD, DVD-A and even SACD ... Seriously now? Who would agree with that?

well frankly I'm no expert but I'm learning and probably know a bit more than the common joe. There are az couple of surprises -- Dave Grohl who bought this famous SoundCity analogue deck and made an album two years ago (forgot the name, I have it but anyways ). I have heard the cd and the (very heavy and thick^^) vinyl and surprise the LP was much better.  euhhh ...  BUT and this floored me .... they were made from 2 different masters!! The CD was over compressed to the max, the LP had tons of dynamics. I don't understand why he did that as he is somebody who can control that kind of thing but it proves your point -- its the master which is finally used on the end format itself.

But then again, this site really devulges nothing. If they found a way to make a file small without any artifacts being removed, that could be interesting technology unless Warner, Sony and Universal are using this as some new drm copy protection scheme.

They do mention the archives though (eg Ella Fitzgerald), so they are implying quality improvements too??  I(m confused. Anyways I signed up for their newsletter (as DR. Arin Delle   they give Dr. as a choice! ) -- see what comes of it

[InflatableMouse]

well frankly I'm no expert but I'm learning and probably know a bit more than the common joe. There are az couple of surprises -- Dave Grohl who bought this famous SoundCity analogue deck and made an album two years ago (forgot the name, I have it but anyways ). I have heard the cd and the (very heavy and thick^^) vinyl and surprise the LP was much better.  euhhh ...  BUT and this floored me .... they were made from 2 different masters!! The CD was over compressed to the max, the LP had tons of dynamics. I don't understand why he did that as he is somebody who can control that kind of thing but it proves your point -- its the master which is finally used on the end format itself.

That is the only reason why vinyl sounds better IMO, when they come from a different master. That tends to be the case these days. I have many examples in my ever-growing vinyl collection; I cherry-pick them for this fact.

But there are examples the other way around, too. You think you're buying a nice remaster on vinyl that's getting good reviews stating dynamics as a plus over the CD, but it turns out the reviewers aren't that cricial (or, are simply deaf) and the album is just as compressed as the CD and actually sounds worse.

Either way, vinyl is much more fun . I just picked up a collectors box of Neil Young, albums 5 thru 8 and its a master piece in every respect. The choice of materials, the quality of the pressings and of course sound quality (these remasters are really very good!). Album covers are made like their originals from the 70's, from the same materials, complete with printed inner sleeves and lyric sheets - all like the originals. It's things like this that keep me buying vinyl, CD's never appealed to me with their cheap plastic cases, downsized booklets. Even collectors items in nice boxes were more often cheaply made with little attention to detail.

[gvanbrunt]

That is the only reason why vinyl sounds better IMO, when they come from a different master.

This is likely true. I suspect that music industry knows that those that buy vinyl want more dynamic range and master it differently. Is a sad but true reality that most people like compressed masters better. Just like they also pick the TV in the store with most saturated and processed image. Which is why it is done in the first place. Sad but true. What would be nice is if the music industry offered a choice of 2 masters all the time no matter what format. Well, I suspect there would be 0 buyers of compressed vinyl, so maybe not that one. Lowers the bar to becoming a "connoisseur of music".

One other thing with Analog/Vinyl: It is often highly distorted by reproduction equipment. If you look at many rigs of "audiophiles" you'll notice a butt load of tubes. From pre amps, to poweramps and everything in between. The reason tubes are preferred by them and musicians (guitarists, bassists, etc) is that when they first "clip" they produce mostly even order harmonics. As opposed to solid state which is mostly odd. The coloring of even harmonics adds a "warmth" to sound. Tubes amps can be designed to be quite linear, so if you do it properly you can produce a linear amp that sounds better if you "accidentally" clip. The problem is most of the rigs I have seen are dialed up to push it into distortion. So the guys that want "pure" sound actually like the distorted version and dial it up in thier settings. There are also some devices that add "tube warmth" to a signal path. They just add "plain old" distortion. It does sound better most of the time though. Basically it adds a slight degree of compression as well. Yes that evil, vile thing everyone seems to want to avoid. My guess is that was the start of where the music industry got it's "idea" to compress masters more, the problem is they went too far.

One of the most common pieces of equipment in a recording studio is a tube preamp. There are even digital equivalents that you can plug into recording software. Almost every mic is run through them. All of this is an attempt to "fatten up" the sound. So my take is that Vinyl doesn't offer any difference in sound by itself. Its the different masters and rigs that make all the difference.

[flac.rules]

Two masters seems a bit excessive imho, just add a "compress" button on the stereo, and people who like it can turn it on.

[gvanbrunt]

Two masters seems a bit excessive imho, just add a "compress" button on the stereo, and people who like it can turn it on.

That would be about the same as saying just add an expander for those that don't like the compressed version.

[InflatableMouse]

Actually I think compression can be done realtime. Aren't broadcast stations using devices like that? I think some DSP plugins can do something similar called an exciter or something.

[flac.rules]

That would be about the same as saying just add an expander for those that don't like the compressed version.

No it wouldn't as compressing realtime with good result is a reasonable thing to achieve, an "expander" in this sense is in practice impossible.

[mwillems]

Real-time compression is very common, and you can still buy devices that will do it, but most folks use software now.  Inf act, dynamic range compression is one of the things that the "Night Mode" flavor of adaptive volume does on-the-fly so you can test out how a "compression" button might sound by playing music in "Night Mode."

Back in the days of reel-to-reel production you did compression with a box in more or less real-time.  You played the first master tape and ran the output into a compressor, and then recorded the output of the compressor and that's your new master.  Nowadays I think most professionals use software to do the compression which is "faster than realtime."  While devices called "expanders" do exist, they can't effectively recreate the original from a compressed master.

Although, reading it again, I'm not sure gvanbrunt's point was that it's technically infeasible to do compression on the fly, maybe he meant that an "on/off" type switch wouldn't necessarily be a good idea. A one size fits all approach to compression wouldn't produce good results.  A "compression button" is a little like the "loudness" button on old receivers; it's unlikely to lead to good results or make many people happy. 

[gvanbrunt]

My point was that compressing a whole mix at once would not sound the same same as mastering a mix that used high compression. Not even close. There is whole lot more going on during mastering than just slapping some compression top of the whole thing. So from the perspective of a record executive, if you had to choose between a low compressed "audiophile" version or a compressed one that 99.9% of the population would prefer, what one are you going to pick? You are going to pick the compressed one. Saying that 2 masters is excessive isn't realistic, as there is no way the recording industry would switch to uncompressed versions and tell most users: "just press the compression button on your music player". I would say it's about as realistic as me suggesting audiophiles just slap an ex-pander on the compressed tracks.

I'm not saying I agree that compressed is the way to go, I'm just saying most people haven't got a clue when it comes to audio and video. And I'm saying the music industry complains of poor sales and pirating could easily add value by giving consumers a choice.

[theoctavist]

more BS from audiophiles. LOL

[AndrewFG]

Did they release a technical specification yet? Up front they say that the format is "lossless". But later in their blurb they mumble something about "selecting frequency bands" which is exactly what MP3 does (and that is not lossless at all). So my guess is that it is not "lossless" at all. I guess they have a lossless layer at CD quality, plus another lossy layer that provides additional higher resolution cues on top of the basic CD layer. They also mumble something about "encapsulation" which might imply that the CD layer and the lossy layer are encapsulated in separate streams (a bit like video files). They talk about it being "backward compatible" but its not clear how. Perhaps they encapsulate the lossy stream as separate "boxes" in RIFF, AIFF or MPEG4 streams. In which case the data may be "backwards compatible" but you would still need a custom demuxer to get at the lossless PCM part. Of course the above is all speculation, so I would love to see the technical specifications.

[dtc]

I read  speculation somewhere that they are analyzing the content above some threshold (like 60 Khz in a 192 KHz sample rate file) and if it is below some level (like -100 dB) then they would stuff other content in that section. That would technically be lossy, but for all practical purposes, nothing would be lost. Not sure if that is actually what they are doing, but it is an interesting idea as to how to compact large hi rez files.

[AndyU]

There's a decent attempt at an explanation from Stereophile here.  Loosely it's a way of packaging 24/192 files in the same filespace as 16/44.1 by exploiting redundancy in the high-frequency content. Technically it is lossy, but Meridian's argument is that what is "lost" is noise or inaudible so who cares.

[BartMan01]

Technically it is lossy, but Meridian's argument is that what is "lost" is noise or inaudible so who cares.

That is exactly what MP3 claims it can do at higher bit rates.  Calling it 'lossless' when it is actually throwing away bits is very misleading.

Edit - I see what they are doing there:  They are lossy compressing the data first, then they are losslessly compressing the results of the first compression...

[contium]

It would be interesting to listen to the additional data alone.

[AndrewFG]

There's a decent attempt at an explanation from Stereophile here.  Loosely it's a way of packaging 24/192 files in the same filespace as 16/44.1 by exploiting redundancy in the high-frequency content. Technically it is lossy, but Meridian's argument is that what is "lost" is noise or inaudible so who cares.

Thanks. The Stereophile article gives a good explanation. The "origami folding" part is where the smoke and mirrors is done. This "folding" requires two concepts: 1) coding a second signal into the unused bits below the noise floor of another (main) signal (vertical stacking), and 2) coding a higher frequency signal into samples that have a lower sample rate..

There is absolutely no problem with 1) -- it just requires some math processing to do it. The problem is with 2) -- with conventional sampling it is impossible (you cannot convey a frequency that is higher than the sample rate). So this is the area where they are doing "magic"...

I can think of several ways of doing such "magic": a) halving or quartering the signal frequency before sampling, and then doubling or multiplying it by four before playback, b) describing the signal as commands to a regular audio synthesiser e.g. "play 76kHz at -90dB for 5mSec" (a bit like a MIDI player), or c) describing the signal as parameter values for a parametric "audio floor synthesiser".

In all cases this "magic" is (must be) lossy. The physics limitation on coding high frequencies at lower frequencies requires this to be so.

[AndyU]

What do you make of the claims in the "Correcting the Source" section of the Stereophile article? I read that as claiming that MQA somehow (magic? mathematics?) "improves" a 24/192 master.

And if at the MQA-encoding stage, the temporal effect of the A/D converter can be compensated for, the complete system offers a transparent window into the original musical event. Meridian describes this as "taking an original master further, toward the original performance, in an analogous way to the processes expert antique picture restorers use to clean the grime and discolored varnish from an Old Master to reveal the original color and vibrancy of the work."

This claim of "improving" a hi-res master is borne out by the experience of Jason Serinus in the listening test:

In Audio High's exceedingly dry listening room, we began with an 24/88.2k file of Hilary Hahn playing what I believe was a movement from J. S. Bach's Violin Concerto No.2 in E, BWV1042. As compelling as the untreated hi-res file sounded, I literally laughed at the difference when the MQA version began. Not only did it feel as though a veil had been lifted, with far more color to the sound, but instruments also possessed more body.

This now looks like an incredible claim. Take a 24/88.2 hi-res master, magic it into the same space as a 16/44.1 cd and make it sound better all at the same time.

Water into wine next mebbe?

Anyone know whether the technology will be built into DACs - Meridian have already released it in their Explorer DAC - or will software players like MC and foobar be able to decode it on the fly?

[mark_h]

or will software players like MC and foobar be able to decode it on the fly?

Undoubtedly, as long as a license is paid

Still waiting for more technical details, but anytime Meridian come out with something it usually has merit.  They are masters of digital audio.

[AndrewFG]

What do you make of the claims in the "Correcting the Source" section of the Stereophile article? I read that as claiming that MQA somehow (magic? mathematics?) "improves" a 24/192 master.

This claim of "improving" a hi-res master is borne out by the experience of Jason Serinus in the listening test:

This now looks like an incredible claim. Take a 24/88.2 hi-res master, magic it into the same space as a 16/44.1 cd and make it sound better all at the same time.

Water into wine next mebbe?

Anyone know whether the technology will be built into DACs - Meridian have already released it in their Explorer DAC - or will software players like MC and foobar be able to decode it on the fly?

I guess that if it can be proven that all A/D converters in the past have been consistently adding a predictable systematic error, then you could retroactively subtract that error and thus "improve the original".

Or one could even imagine doing "big data analysis" on huge numbers of recordings to detect if there was a statistical "shadow" of a consistent systematic error..

However you would not specifically need MQA for this since such a correction could just as easily be applied to recordings made using conventional hi-res sampling techniques.

Furthermore I don't know how you would argue with the recording engineers who released a track because it sounded just like they wanted it, and who might get pissed if somebody messes with their masterworks; but perhaps you could make this "improvement" switchable..

[sirkus]

The only thing that will considerably improve music from a format point of view is multichannel vs stereo.

That's the point! Just listen to Tacet's "Real Surround Sound" recordings, this must be the future! It can be surprising the first time, but after that, when you just listen to the music, suddenly you realize that you are "inside", where each instrument has a place. After that, stereo sounds flat.
Unfortunately, this requires new recodings and it's more costly than the almost costless "better than CD" oversold BS. Added to that, it will be hard for a lot of manufacturers  to completly re-work their catalogue for surround. The industry goes to simplicity, despite their rethoric about quality.

PS: the convenience graph on MQA's site is a marketing joke

[ken-tajalli]

Yet another digital format!
But if Bob Stuart is behind it, we must take it seriously. (I have been a big fan of his work)
I believe it is primarily aimed at high quality streaming music, or possibly portable music.
But as the internet speeds are increasing, and mobile storage's getting larger, I wonder how much of a life span it will have.
Now adays I have over 128GB on my phone, 4G and an external DAC!
Not to mention 150Mb/s internet on my PC.
I switched to uncompressed music over two years ago!
We will see - any super format is still at the mercy of playback hardwares, I think we need to address that first.

[sfspirit]

Hi,
Certainly would be interesting to see how MQA compression compares with MP3 and AAC for the same bitrates.  Can AAC and MP3 be used for 192/24??  How would the filesize compare?  How would the sound compare?

[flac.rules]

Why is the filesize so large? Using only the dynamic range needed is the main way almost all lossy codecs work, and having lossles up to 22 khz shouldn take more than maybe 700 kbits, then adding lossy on top of that for 48-96khz should only be around 200 kbits.

The descirption of how the sound changes doesn't seem like anything that will hold up to scrutiny.