MQA is Vaporware

[PeterV]

For some (non-believers..) MQA seems to be 'evil' but for me, this algorithm is a blessing. I am listening now almost 1 year to MQA remasters and compare them with the original FLAC or wave files. In almost all comparisons MQA wins. Only some DCC remasters sound better. With the Tidal stream, it's a no-brainer. Instead of 16 bit 44 kHz now I receive 24 bit 44, 48, 88.2, 96, 176, 192 and even 382 kHz.  So no vaporware at all..! It works, it upgrades the original files, since the time-smear is reduced and this effect is audible! Last week Donald Fagan Nightfly' A/B comparison is very convinvincing. 

[PeterV]

I am the 'fanboy' as soxr, the 'hateboy' mentions in this thread. Yes, I own a NAD C390DD MQA certified Direct Digital Feedback Amplifier, including the BluOS MQA certified streamer module. I am aware of the architecture of this integrated DAC/amplifier and although it operates in 35 bit and 108 MHz masterclock, it is for sure converting the digital PWM signal into an analogue signal, since loudspeakers do not swallow digital signals Frederic :-)  I never claimed that the MQA filters are analog, but explained that MQA is an end-to-end process because temporal and other artifacts can only be minimized if the two ends cooperate in real-time in MQA the decoder/DAC is controlled by the encoder. 

 

The question was asked in January to Bluesound:

 

One of the features of MQA is that it can fix errors in D/A conversion and to do so, it must be tuned to each make/model of a downstream DAC chip.

Bob Stuart quote: 'Full software decode is not possible because the DAC must be known and characterized. MQA is an analog to analog process."

According to the company, the more MQA knows about the gear responsible for A/D and D/A conversion, the better it can eliminate and/or correct for unwanted digital artifacts introduced by the conversion process. MQA does this by applying specific filters and processing based on the actual gear in use on the A/D encode side and on the DAC decode side. MQA is, ideally, an end-to-end technology. One of the most sonically important corrections these filters make, according to the company, is in the time domain. The company refers to this as "de-blurring". MQA is an end-to-end process because temporal and other artifacts can only be minimized if the two ends cooperate in real-time; in MQA the decoder/DAC is controlled by the encoder.

But what happens if there is no traditional D/A conversion with a DAC? A bluesound powernode 1 for example converts the digital audio signal (PCM) through a digital sound processor (DSP) into a pulse width modulation (PWM) signal that directly drives the speakers. In this concept there's no DAC involved, so is there a) nothing to be fixed because this system has no design flaws like jitter or b) MQA doesn't come to its full potential because the DAC stage is missing or cannot be corrected?

 

Answer by Bluesound:

The DAC in the Powernode (and POWERNODE 2) is fully MQA certified. There still has to be a DAC of some sort to drive your analog speakers  That being said C is much closer to A than B (at least in our alphabet...)

[PeterV]

By the way soxr, please read this article, it will enlighten you.. MQA is not cheating with the Bluesound functionality to offer EQ in the digital domain. The end-to-end data integrity is maintained. They offer even more extensive EQ in a portable MQA player like the Pioneer XDP-100R which I purchased for very reasonable price a few months ago.  unfolding to 24/384 and no problems with EQ...

 

http://www.soundonsound.com/techniques/mqa-time-domain-accuracy-digital-audio-quality

[PeterV]

OK, no problem, there are reasons for hiding someone's identity and I am totally not surprised with this guy that he requires this...

[PeterV]

Well.. I was triggered by the fact that soxr was using my post with my picture on a post at the Blusound helpdesk page and that he promoted me as a 'fanboy' with no knowledge how my DDFA amplifier works..   For someone with a background in engineering not very smart from soxr not to know that even a Direct Digital Feedback Amplifier requires an analoque output signal to steer the loudspeakers ...  I can imagine that his boss will not appreciate such stupidity and that it would jeopardize his employment.. So my apologies for revealing soxr's name..

[PeterV]

Well, that is not the first time that someone assumes that I am being paid by MQA.. Soxr also did and also at other internet fora. My Name is Peter Veth and ii live in The Netherlands. I have no background in electrical engineering, but a chemical background. Music and audio are my passion and I was one of the lucky guys to own a relatively affordable audio set with the 2 x 150 Watt DDFA Dac/amp and Bluesound streamer with which I am now listening to music in Wave, Flac and MQA format. Loudspeakers are Wilson Audio Watt Puppy 5.0 (modified) and the combination seems to be able to reveal the difference between MQA and non-MQA encoded music files. I like what I hear and it is convincing. But I am not being paid by anyone for just being interested about the developments and discussion around MQA. I was and still am very surprised by the hostility when it comes to this format-war and am actually also very interested what is really behind all this.. What is there to lose, what is there to win..? With regard to the claims of MQA, I am very interested to get more confirmation if the algorithm is capable to reduce time-smear or not. All discussion here and elsewhere are not focussed enough to this aspect. Is anyone capable to measure the difference in time resolution between a MQA and non-MQA music file or not..?  I only am here to find confirmation to what I hear and what I seem to like.  So if this is disrupting someone's beliefs or backward engineering tactics in which the magic of MQA is nothing but úpsampling' I believe that is very naive.  Other arguments pro- and against are required to convince me that MQA is 'vaporware' to me that is a pre-occupied opinion and nothing more than that. Let's focus top the time-domain claims.. much more interesting!  if there are other algorithms ou there which are capable to repair past- and present time-smear.. well, let's listen, test and measure but do it correct please..

[PeterV]

Sad to notice that no one out here seems to have the interest nor the tools to verify the claims by MQA that with their algorithm both the time-smear  and impuls response are drastically improved. Upsampling is a also something fundamentally different to defolding.. It is positive though that this thread is keeping the name recognition and mystery around MQA much alive..?

[PeterV]

Well Fokus..let's ask MQA themselves, but also find a way to do proper A/B comparisons in the analogue domain. Is this possible or impossible? Even though loudspeakers and microphones have their own intrinsic (time-smear) flaws, why not test, measure, compare? 

[PeterV]

I am not interested in measurements or graphs or reversed engineered software programs. All I care about is that MQA is contributing positively to the sound and that is just my personal observation. What I want to know is if this is due to the improvement in impuls response and as a result also to the reduction of time-smear in air. If my ears are capable to notice this en MQA claims are very clear about the reason and call this a 'paradigm shift' then all I care about is to get professional and objective confirmation. 

[PeterV]

And off course, reverb and other natural or artificial sound is present in many recordings. With MQA we are talking about the existence of unwanted time-smear in recordings due to (cumulative)  aliasing effects. This is what they claim to be tackling, not the artist's signature of the sound. I just want to understand their philosophy and get confirmation how it works ( or doesn't...) Let's just  take a step back and look at MQA from another perspective to understand better if and how time-smear reduction works.

[PeterV]

Elsdude..apologies, a typo..I am not interested in measurements in the digital domain which are not related to any effect in the time domain...Off course I am interested in measurements, but those in the domain where my ears operate as well.. I am very interested in the time-smear claims. This is the core of MQA's claim of being capable to regenerate old recordings into a more analogue, natural sound. This week one of my favorite 80's albums by Donald Fagan - The Nightfly' was released on TIDAL in MQA. Perfect material for me to audition and compare. But how much better is the impuls response ans time Smear of this record in MQA compared to the original (digital) master? Is this measurable or not?

[PeterV]

I am only trying to combine what I am capable to hear and what I read about how MQA might work. For example this part in the excellent sound on sound review interests me a lot: 

 

" a single apodizing filter can be designed to compensate for the time-domain responses of any number of linear-phase brick-wall filters in the signal path, and without needing to know their exact number or precise types. This means that apodization is equally effective whether applied to a current conventional digital signal chain, or to an archive of historical digital recordings." 

 

The complete article here: http://www.soundonsound.com/techniques/mqa-time-domain-accuracy-digital-audio-quality

[PeterV]

I will take the challenge immediately to do A/B comparison just by listening..even with ears like you describe esldude. But some measurements in air will be helpful to get confirmation, that's an objective parameter I would say.

[PeterV]

I clarified earlier that I do have interest in objective measurements, but focussed towards impulse response and time-smear improvement, that's all. 

[PeterV]

7288 albums and counting.....:-)  not bad for 'vaporware'.....!...http://www.meridianunplugged.com/downloads/MQA_List.csv

[PeterV]

6 hours ago, Ran said:

 

Thanks for the list. I will now know which albums to avoid on Tidal.

Do not worry Ran...both MQA and the 16/44 versions are and will stay available on Tidal :-;

[PeterV]

Hi realhifi and indierock. More important than just this list are some albums which I know for years are now released and are making a profound impression. The 24/192 MQA unfold of Crime Of The Century is the best sounding version I have ever heard and I own several vinyl, thr MoFi CD and the HDtracks download. Also Yello's Toy sounds much better, which surprised me, since it is mostly electronic music.

[PeterV]

What is even smarter than I thought  is the way MQA is able to use all these discussions regarding their technology totally in their advantage. This amount of free publicity by nay- and yes- sayers is priceless. If MQA would be an inferior format like MP3, no audiophile listener or criticaster would care to discuss it. Now, the fuzz and buzz is a strong marketing tool :-) just count the amount of times the 3 letters are being mentioned every day.. just listen and you will understand that it's all about about the MUSIC

[PeterV]

My audio set and room acoustics has its limitations, but is sufficient enough to distinguish increased dynamics, improved soundstage and presentation of detail like the trainstation voice on 'Rudy' the cuccoo and harmonica etc. Supertramp's COTC MQA version is for me a marvelous example what can be avhieved with the technology. The noisefloor has decreased and overall sense of presence and energy has increased. But all this is a relative and personal observation, I can only advise to test this for yourself but with certified MQA streamer and DAC  to ensure end-to-end optimised analog sound. 

[PeterV]

Again: the only way to 'debunk' anything is to listen to it. If you like MP3 that's fine.

 

I am perfectly informed how MQA works and that it's lossless in the audible band up to 48 kHz, above it can be regarded as noise. Compared to the cut-off frequency of a CD this is allready a large improvement. But you need to understand that the end-to-end encoding and decoding process is all about improving the total impulse response and decreasing time-smear. 

 

That's what results in a very convincing audible improvement of these albums ans all others I am listening to dor just 20,- euro per Month. No need to download music anymore and the soundquality is better than ever before.

 

 

[PeterV]

What is your point jabbr..? Youtube has no MQA encoded albums available (yet) and if they would, you will need to decode it properly. Only then you will hear what it has to offer

[PeterV]

You really have no clue Frederic. I seem to have made a lucky choice with NAD 390DD. It is relatively affordable and it is build around the fully digital architecture a the 35 bit 844 kHz PWM DAC. Plenty of headspace for full 24 (or more!) resolution. Check the specs and reviews on the web. As mentioned earlier A/B testing shows MQA is a clear winner here at home and I am very satisfied with those albums I know by heart.

[PeterV]

17 hours ago, The Computer Audiophile said:

 

You're supposed to sit in front of the loudspeakers :~)

Haha, no, this strange position of the chair wa caused due to a recent upgrade to cat&m gigabit ethernet router switch and 75 cm of Audioquest Diamond ethernet cable (2nd hand lucky purchase) for the last bit towards the streamer/DAC.

 

Usually, the audio gear is located in the centre and my chair right before it for non-listeneing usage. But when I dive into the sweetspot, I know where to sit :-)  But interestingly enough, most of the times I just lie on the bench nowadays and especially when I relax and not tease my brain pinpointing too much, I really enjoy the music especially when the MQA versions are played.  But again: this is my personal situation. I cannot compare with audiogear costing 10 times of what I have collected over the last years.  The NAD DDFA technology captured me when I read the review of the M2 at the time. But it was too expensive for me, so the 390DD was a perfect replacement for the Krell 300i. The Watt Puppies are second hand and I paid 4.500,- euro for them. The woofers needed to be repaired and this was done professionally with the required rubber rings. Step by step the sound improved, and I now appreciate these loudspeakers a lot, it might be part of the reason why I am able to distinguish the Time Coherency improvement by MQA, since Wilson Audio teached us years ago the importance of Time- and Phase Coherent design of loudspeakers. This pays of now with MQA I suppose.. 

[PeterV]

18 hours ago, jabbr said:

I’m just tryin to learn, so would MQA (if it did YouTube) convert:

Into: 

? 

Or vis versa

??

Hi Jabbr,

 

Please understand that it is not up to MQA to decide which albums by which artists will be remastered using MQA, but it is up to the record companies Warner, Sony and all others who meanwhile incorporated MQA for current and future streaming services.

 

The process of remastering albums to MQA is described here:  http://bobtalks.co.uk/blog/provenance/provenance-series-introduction/

[PeterV]

9 hours ago, Don Hills said:

 

This. Very much this. How much time do you think has been spent doing this per album converted so far? For the vast majority, I'm betting on some semi-automatic profiling of the frequency response of what they are given, to pick a set of filters that provide the maximum aliasing ("time domain optimisation") with acceptable distortion. And as Mans has shown, most of them end up with the same filter.

 

Well, that might be true, I do not know. And yes, it would be an enourmous job to do it like this. But that is the responsibility of the record companies I assume. Even if the MQA remastering proces is an automated form of reprocessing which includes fingerprint settings from the studio with regard to the sum of all applied A/D D/D or D/A processing used during the production of these albums and at the end this information is decyphered including compensation for that DAC performance at home, it is an excellent, indeed total time-smear reducing tool. Fact is that I hear the improvement on the spot, not only at home, but also during live A/B comparisons. It is very nice to notice that most classic albums from the label Deutsche Grammophon are being released on Tidal in both 16/44 and in MQA 24/96. Perfect comparison material. It will be interesting to compare Qobuz 24/96 flac with Tidal MQA. The audible difference will not nu huge I expect. So again: MQA is a no-brainer with regard to upgrading your streaming solution. What it does with a classic album like Crime Of The Century made a huge impact on me. It delivers the best sounding remaster with higher resolution than all other formats and knowing that the original album was recorded analog im 1974, it shows me that there is very much to gain with the technology  especially with older analog recordings, but also with transporting and presenting nowadays DSD or even DXD recordings at the highest (24/384) level and that via a streaming service...So what is there to complain about..? What is there to loose?  NOTHING!  There are only winners for those who enjoy it and those who do not: just proceed listening with what you have, WAV,  FLAC 16/44 sounds still excellent! Especially when the recording was done with great care. DSD and DXD sound superb. Maybe soon streaming of such formats will become available. But altogether it is just a niche market. I assume over 80% of streaming is still in AAC or MP3.. so..let's wait and see what will happen.