+1
Been lurking in this thread. So far have not heard MQA; I share the doubts of many that the industry will adopt it, but am certainly trying to keep an open mind. The degree of hype in the press, however, has IMHO been so over-the-top that my suspected bullshit-detector has long since kicked in. I suppose that will be resolved one way or the other once I get to hear it. If however they do go with DRM that keeps me from making as many copies as I need for my own use or requires me to “authorize” different servers, that would be an instant and total dealkiller for me. And IMHO if the capability exists in the data structures then that result is a foregone conclusion once the corporate lawyers get involved. Let’s hope for the best.
I wish I could be as enthusiastic as you - because I truly love new tech and new ways of doing things. But let’s imagine for a moment Tidal decides to go forward with MQA. That would likely mean a couple of things. They would probably increase their offering from two resolution plans to three. Currently, you can sign up for $10 a month for lossy compression, or pay $20 a month for lossless CD quality that any DAC and computer setup can play. Once Tidal is installed, they’ll add a third tier - let’s call it $30 a month - and you get high resolution. So far, nothing wrong with that!
Only, the middle level, the current lossless CD quality level, will become MQA encoded across the board. But wait! That’s supposed to be CD quality - maybe even better. Right?
Wrong. It’s actually a lossy version of the higher resolution recording. Here’s a quote from MQA.
“Preferably, the first lossy representation is an accurate representation of the input audio signal other than the effects of time-invariant filtering, sample rate reduction and requantisation that imposes a time-invariant noise floor. If all quantisations, including those within the sample rate reduction, are performed to a constant bit depth and with appropriate dither, the “lossy” representation can be of a standard equivalent to CD quality and would have been considered “audiophile” reproduction only a few years ago. This is in contrast to traditional “lossy codecs” which dynamically adapt the spectral noise floor and sometimes the bandwidth in response to the input signal.”
Now Tidal will have a compressed lossy cheap level, a mid lossy higher bit rate compressed level, and finally and very expensive high resolution level that we pay for.
Right now, Tidal streams the CD quality in lossless FLAC (if I remember correctly) and we know that once decoded FLAC is truly lossless.
This is just something to think about - because sometimes you have to be careful what you wish for… it just might come true and then be hard to unravel, if not impossible.
Elk said I certainly hope they are otherwise the same file with MQA being the only difference. Otherwise, not only are the comparisons invalid it is basic fraud.Which reminds me of another issue with claiming the vinyl version is more coherent than the CD. (““The CD of the recording has an unfocused, diffuse image of a piano hanging in space, with the room reverb mixed in and confusing the picture.” But the vinyl version sounds coherent.) Most LPS are made from digital recordings. For the above to be true, vinyl would have to fix problems inherent in the digital source file as the digital master would also be incoherent.
But I like the idea of compensating for any imperfections in the original ADC when the recording is made. If MQA can do this alone, it will be an advance and this would improve all sound downstream.
I am troubled a bit that they claim to be able to improve on the performance of every DAC and ADC. Does that not indicate they believe all DACs and ADCs are flawed? Certainly nothing is perfect, but it takes some cajones to suggest they know better than others what should be done with signals. It just sits poorly with me, especially when I know what someone like Ted has done to keep our DAC’s internals edge of the art - and they can wave a wand and make it better?
What am I missing?
I interpret what they are the same way. The appear to be claiming that all ADCs and DACs possess identifiable flaws which MQA can fix.
I have difficulty accepting both that these flaws exist to this extent, and that MQA can somehow detect these various flaws in all the available interfaces. Certainly, each piece of equipment can be bettered, but I doubt the typical recording chain and audiophile DAC has such major issues a single bit of software can improve them all. Especially not in a way which is specific and unique to each bit of kit. Color me skeptical.
I am less skeptical of the claim they can device and encoding scheme which allows transmission at lower bit rates. FLAC does this brilliantly, but perhaps this can be bettered. I get nervous however when it appears it is lossy. This is too reminiscent of eMPty3.
Maybe the DAC certification process also is about control and making sure everything works (aka Apple store approval, sort of).
I also think that since MQA perform audio origami, which (supposedly) has to do with human auditory perception/psychoacoustical relationship, it could be that we consider it better, but in fact it is only different. It’s like favouring second order distortion from a tube amp - it might get you closer to the music from a ‘comfort zone’ standpoint but it isn’t necessarily a measure of true high fidelity if this is what you relentlessly strive for…(?)
Frode said . . . it could be that we consider it better, but in fact it is only different.Exceedingly possible, especially when hearing it only for a short demonstration.
Thanks, Paul, Elk and Frode! You all made my daily swim far more interesting than it might have been today. Much to ponder here. It was great exercise for the little gray cells and the time sped by.
So, unlike many audiophiles, the term “lossy” doesn’t make me apoplectic. In fact, I regularly fool audiophiles by playing an eMPty3 (nice one, Elk) file through the DS and then a Redbook file and ask them which was the higher fidelity. Most, including me, flunk the test as often as not. That doesn’t mean that I don’t prefer Redbook or am unable to ever tell the difference. My point is merely that I don’t believe there is a one-to-one relationship between bit rate and what sounds good. God knows I have some 24/176 files I wouldn’t pay 2-cents for. If the “lossy” version Paul refers to sounds better than CDs to my ear, knowing that it’s lossy will cause me no cognitive dissonance whatsoever. I am no fundamentalist. All religions are welcome! 
I favor the second order distortion of tubes even if it is heresy to the measurement crowd, so it is entirely possible that such a phenomenon explains what I heard at CES. Pleasing distortion. I like it and humbly confess my sin.
That said, I do not favor anything that will force folks to use or pay for something they don’t like. So, I hope that is not what is on the table from Mr. Stuart.
Elk saidDifferent is definitely seductive and our own beta testing with firmware proves that initial reactions can prove to be misleading. What's odd about CES is the uniformity of opinion among those who heard separate demonstrations and that multiple recordings were used and reacted to by a disparate group of individuals. The CES experience was not a science experiment so I don't propose that it proves anything. But I also don't think that what we all heard was just* a matter of it being "different." Trust me on this: I've heard plenty of "different" that was NOT BETTER.Frode said . . . it could be that we consider it better, but in fact it is only different.Exceedingly possible, especially when hearing it only for a short demonstration.
vhiner1 said I don't believe there is a one-to-one relationship between bit rate and what sounds good. God knows I have some 24/176 files I wouldn't pay 2-cents for.An excellent point and I fully agree. Sound quality is primarily a function of a good room and a well-placed mic. Equipment is a secondary consideration. Bit-rate is a very distant third. An MP3 of a wonderful recording can sound extremely good.
I favor the second order distortion of tubes even if it is heresy to the measurement crowd, so it is *entirely* possible that such a phenomenon explains what I heard at CES. Pleasing distortion.I hear you. I like tubes also, although those with no significant distortion, yet they still sound different. :)
But I do not think MQA adds distortion or anything euphonic. Rather, I bet it takes something out. As what is likely a poor analogy, I can remove the sound of a ventilation system from one of my recordings by sampling the unwanted sound and then removing just that sound. It is brilliant. The resulting recording is not only blacker, but also seems to be richer, more dynamic. You hear just the good stuff.
As I understand it, MQA is claimed to address filter pre-ringing and other anomalies. If it truly can do so, like removing jitter, the recording should sound better. It will seem cleaner, crisper (in a good way). Perhaps you experienced something like this.
Good points, Elk. I’d never thought of the removal of things. Damn this is fun to ponder!
This is just to share my experience with MQA (hopefully i’m not fully off-topic here 
)…
I had Mytek Brooklyn with MQA enabled firmware for a week in my system.
In general to make a base-line - the Mytek playing standard files have less energy in deepest bass and more energy in bass/mid-bass in comparison to DS. Also soundstage is much less defined than DS and completely missing holographic image of the DS.
With MQA files played via the Brooklyn the sound was very interesting - much more dynamic (I would dare to say even overly dynamic, just imagine the DS sounded dull/flat when compared the NON-MQA track via DS to MQA track via the Mytek).
Also MQA on Mytek made strings much better defined (like much faster/precise impulse response) - this will be probably this “temporal-blur removal” which is so much described as MQA advantage. This interested me the most from whole MQA benefits as it sounded really nice.
However even with MQA Mytek was still far far away from soundstage definition of the DS, however I think this would be probably the “lacking feature” of the Mytek rather the MQA property.
What I found suspicious was that on VU-meters of the Mytek the MQA files has been visibly louder than non-MQA version of the same file. So this can be very tricky as we know we usually prefer louder sound (in this case MQA). So it is still questionable if perceived better SQ of MQA was not just because of louder sound - I did not go so far to level the volume by using measurement microphone etc.
I have used 2L records MQA demo files in my comparisons.
DS was fed via bridge (minimserver transcoding to WAV).
Mytek was driven by Auralic Aries + JB + Regen with J-Cat refference USB cable (USB cable power branch not connected anywhere).
This is very informative and thanks. I have long suspected that what’s going on with MQA, in some cases, is that a relatively OK sounding DAC is made better by MQA. That makes sense. It may also answer why MQA has consistently made the DirectStream sound worse.
Thank you, Maniac. Your observations on dynamics and definition are of particular interest.
Your observation that MQA enabled tracks are louder than the reference tracks is disturbing. Many SACDs were produced this way with the DSD layer louder than the Redbook layer.
If you can easily see the difference on the VU meters this difference is pronounced (it is hard to see a couple of dB RMS difference on dynamic material as the needle bounces around so much).
Thanks, Maniac. Very interesting. I’m always interested in first-hand experience.
Munich, May 6, 2016 — Warner Music Group (WMG) today announced a long-term licensing deal with music technology specialists MQA.
Could this change things ?
For whom and in what way?
I think he means that soon the train is about to leave the station and some hesitant passengers are left behind on the platform.
Aurender has decided to not support RoonSpeaker and quite a few customers and prospects are unhappy with that decision to put it mildly.
Warner is not Apple, but anytime a significant distributor of music signs on to a process, it will have an impact on the market…if customers respond positively. The folks at Tidal should be happy because this will mean that they won’t have to encode Warner titles. If Sony were to sign on as well, then I would predict widespread adoption of MQA. Just some armchair observations.
Consider this: Warner, which represents 15 percent of the world market, just announced today that it is making more revenue via streaming than any other distribution method. (NPR). If that’s true of Warner, then it will soon be true of Sony. If streaming is the future, MQA is currently the only serious way to offer high quality recordings to people without huge broadband pipelines.
Here’s something else worth noting: Auralic and Aurender have demonstrated MQA-enabled products. MQA has announced that Berkeley Audio, dCS, Ixion and Kripton were actively planning MQA-enabled products.
Sony’s position on MQA will be critical.