, ,

HighResAudio.com calls for a deeper technical analysis of MQA

Gendarmenmarkt, Berlin. The afternoon was apparently warm for mid-February but not quite warm enough for an outdoor sit-down. Joining me for coffee inside Newton Bar was one Lothar Kerestedjian. Kerestedjian had once been personal assistant to a member of INXS and had the stories to prove it.

Having since relocated to Hamburg and with two decades in audio/video technology development under his belt, Kerestedjian now runs HighResAudio.com – a music download service for the cash-rich/time-poor consumer.

Over a cappuccino, Kerestedjian explained how every title sold via his website is first analysed to verify its high-res content; that it isn’t up-sampled from Redbook or even “Mastered for iTunes”, which, according to Kerestedjian happens more often than he’d like. Such files are sent back to the record label with a ‘please explain’ attached. If one thing was clear from our conversation, our man from Hamburg is BIG on quality control.

I then asked Kerestedjian if he was excited to be offering MQA-encoded titles for download, especially those of greater interest to the average DAR reader? He paused before taking our conversation off the record.

A happy camper Kerestedjian was not. On the day after Valentine’s Day, Kerestedjian was no longer in love with Bob Stuart and Peter Craven’s music technology. His primary concern was one of quality control. The MQA incoming files couldn’t be analysed by a third party – their hi-res content remained unverifiable. The labels had to be trusted not to provide up-sampled content as hi-res masters. Kerestedjian’s pre-MQA experience said otherwise. (Was he aware that the MQA encoder is designed to automatically detect upsampled content?). Kerestedjian’s secondary concern: a perceived lack of promotional support from the MQA mothership.

Kerestedjian’s dissatisfaction manifested officially a week later via a Facebook post (since deleted):

“Breaking News: HIGHRESAUDIO to stop offering MQA. Proprietary system solutions and licensing models aren’t what customers want. MQA is NOT lossless, the original signal is never recovered, estimate to recover at most 17bits (reduces the sampling rate), reduces the frequency range, SNR reduced by 3bit, aliasing with artifacts at 18kHz. MQA encoding filters manipulate drastically the original source. No analysis tools are available to verify the encoded MQA content. Therefore no quality control is possible. highresaudio.com stands for offering purity, original mastering source, none manipulated, tweaked or up-sampled content and codecs that are widely supported and offer use of freedom. We hope that MQA will adjust all the above issues. We are truly disappointed, the way MQA has progressed in the past year. We have been mislead and blinded by trust and promises.”

As the conspiracy theories and accusations of ‘Fake News’ gained steam over at Computer Audiophile, an email exchange with Kerestedjian brought some sober clarification to the still unfolding situation: only some MQA titles would come off the site.

“I [want to] sell only genuine authenticated MQAs. Meaning, only if I know the origin of the master and we can verify and encode the content in-house.” That was 23rd February.

Today, an email from Kerestedjian outlines further his dissatisfaction with MQA, this time on the record and with technical paper in tow.

“I hope my email finds you well. Please find attached [link below – Ed] our technical document which analyzes MQA deeply and reveals many facts (at least what was possible) that MQA is a “lossy” format and procedure. Dipl.-Ing. Stephan Hotto of XiVero GmbH is responsible for the evaluation and is, of course, available for questions and answers.”

“I have asked MQA weeks ago to correct the marketing communication towards the end user and media. As long as MQA is not prepared to straighten the facts, we will not offer MQA any more. The customer needs to know what he pays for, and we have to be able to check technically what we offer and sell to our customers. We are in a very sensible and delicate niche music market. Over the past seven years we have established a very good market position, created a new business for the music industry and artists and customers that cherish the best audible sound reproduction. We moved the music and HiFi-industry into a new business domain, with very little support from anyone. Our USP is that we guarantee (and this is not just said and done) our customers, nothing but the true, native and original source. We can analysis and verify any other audio codec (with MusicScope even DSD and DXD). For MQA is nothing available to assure that the customer is getting our „promise“.”

“Selling HighRes Audio files requires so much dedicated and detailed work prior in selling (download / streaming) the music. This time needs to be invested by qualified audio engineers and a team that understands the total reproduction path. An extensive quality control is therefore a „must-have“ and needs to be in place to fulfill the „promise“.”

“If these parameters are right, then we can provide our customers with honest facts to purchase MQA, and then we will continue to only offer “Authenticated MQA” again. Since the beginning of 2016 we have offered native Studio Masters with MQA encoding, where we can trace and verify the origin of the source. Meaning, the Mastering Studios that use MQA in their production workflow and process are personally signed off.”

“No matter whether downloading or streaming. The perception and expectations of the customer is different. We can not sell and promote HighRes unless HighRes is supplied! For our customers, quality, trust and reliability are the top priorities.”

“Am I now the bad guy? No, I would like that we continue to offer in the niche, customer native and original high-resolution music. There are plenty of new and established customers that are looking for high-resolution albums every day. The HighResAudio market would have grown rapidly and successfully, if the awareness among the responsible people in the music industry were familiar with our target group. Since 2010, we have been a single player on a broad front. Our mission: to offer music lovers, artists, hi-fi enthusiasts and manufacturers of audio devices to offer a new perspective in the digital age for the perfect music reproduction for a unique listening experience.”

“Last but not least: we offer MQA since April 2016 with approx. 250 selected attractive audiophile recordings. But MQA does not reach more than 1% of the total turnover – also not with the back-catalogue of 300 albums from Warner Music in January and February 2017 on board.”

Stephan Hotto’s “Hypothesis Paper to support a deeper Technical Analysis of MQA (Master Quality Authenticated) by MQA Limited” can be found here.

Via email Kerestedjian adds, “I’m not against MQA. I will support the format, as long as we can communicate facts and figures.”

Further information: HighResAudio.com

Written by John H. Darko

John H. Darko

John is the editor of DAR, from whose ad revenues he derives an income. He is also an occasional contributor to 6moons and AudioStream and currently resides in Berlin, Germany.

Twitter: DarkoAudio
Instagram: DarkoAudio
Facebook: DAR

86 Comments

  1. Thx for this! It’s good that digitalaudioreview helps in fostering a decent debate about MQA.

    Instead of reporting on the concerns raised by several audio-companies and by many well-informed participants online (Archimago Blog, CA-Forum) the big audiophile-press outlets espc. in the US continue to hype MQA and offer little insight and information beyond that.

    With Shiit, Benchmark, HighResAudio & Linn four major companies in digital audio already have raised critical questions on the merits and goals of MQA. The piece written by the Linn employee is especially read-worthy and so far has been ignored by the larger audiophile press.

    Beyond these statements it would be easy to raise the quality of the discussion by doing interviews with engineers, musicians, with crypto-experts or lawyers and company representatives. This would help to shed light on the merits and problems of MQA.

    Looking at the facts know about MQA it’s actually astonishing that they got so far with their story. Anybody who has basic knowledge about music production will notice that their “like-the-mastering-engineer-hears-it” claim would necessitate to get a bit-perfect copy of his file – which is actually what a 96/24 download gives you.

    What you get instead with MQA is actually an “unstudioed” master: a file that lacks information present in the original file but has an added DRM component to authenticate that the manufacturer of the consumers DAC has paid MQA a license fee.

    They also absolutely failed to give any substance concerning their claims on temporal-blur and continue to prevent proper third-party reviews of this. Consequently they get “peer-reviewed” by people on the internet & the results are not favorable.

    It more and more looks like MQA oversold a proprietary, lossy product with a questionable DRM-component to an audience that is technologically more savvy than they expected. But given enough label support it might just work 😉

  2. I wonder why no other high res audio site like HDTracks has gotten into this degree of quality control – or maybe they have, and I am not aware of it?

    HighResAudio is not available for download from USA

    • Not on HDTracks… Because then they cant sell 1/2 to 3/4 “hi-res” albums they offer. I read once they left the responsibility to record labels and we all can imagine (and Mr. Kerestedjian have some evidence to it) how that ends.

  3. Very interesting! I respect his position (and I am a ‘fan’ of what MQA is trying to achieve).. PS the download link for the tech seems to be requiring software installs.

    Regards

    Simon

  4. Very well said. Until know I see MQA is like smashing a pumpkin and then ‘cook’ it back to a pumpkin. Perhaps very tasty and a pleasure to eat but I guess not the same as the original pumpkin. Right now, if I can avoid putting money towards MQA, I will. Let’s see if MQA can change my stand point in the future.

    • Agree . . . and just to highlight some inconsistency in the HiResAudio argument about validation and quality: If you go to their site right now, they are advertising the U2 album ‘Achtung Baby’ as ‘remastered’. I don’t believe that album has been remastered (and I would be joyously happy if I am wrong) . . . so how does that align to Kerestedjian’s chest beating?

      Good article DAR_KO . . . thanks for highlighting these issues in the digital audio marketplace.

      Peace

      • Achtung Baby was remastered in 2011. I have it on CD and it sounded fabulous. Not hyper compressed at all. I believe they redid their whole catalog as I also have a remastered Joshua Tree.

        • Thanks, that link validates what I understood. I believe my point is valid about the hypocracy of HRA. They clearly did not validate the remastered claim of Achtung Baby album so why all the fuss about MQA? I think you understand my point.

          • David, you picked one dubious example and claimed overall hypocrisy. That’s a stretch, to say the least.
            If you are an MQA supporter, that’s fine. No need to denigrate its critics just because you don’t like what they’re saying.

    • I agree with Boboso, MQA is a threat to some companies’ business models. Quite frankly, if I can rent my music on a monthly basis with the quality that I have already heard from MQA material, I will never again buy a download. I just hope that people will not put unnecessary roadblocks in front of the MQA folks. Not saying that Mr. Kerestedjian is in this camp. Just making a point overall.

  5. MQA, not interested. It’s a cash cow for one company, or a few, and locks in the future growth of technology to 2016.
    BS is not the right word for it, but it’s close to the truth.
    What we know instinctually, we should pay attention.
    What we know is, if it doesn’t smell good, it likely won’t taste good.
    For the Fan Boys, enjoy it while it lasts.

    • Interesting how many folks on the Internet just know “instinctually” that MQA must be a scam. Because emotions. Reminds me of how many of my fellow Americans knew “in their hearts” that Obama was a Muslim and that Hillary was operating a child-sex ring out of pizza restaurant. There’s no arguing with things people “just know.”

      • Have you read the patents by MQA? I did, and they agree with the observations by HRA.

        Neither the instincts of people who think MQA is a scam, or yours who think those people believe everything they read are really relevant.
        Before Trumpism became popular, fact checking was actually a thing and in this case it is very easy.

        The patents are there and MQA does not solve any problems for the listeners to music.
        The 13 bit streaming format is comparable to 320 MP3 and the 17 bit download format is only relevant in direct comparison with 16/44.1 WAV. It is as large, or larger, when compared to 16 bit FLAC and it cannot compete in quality against 24 bit FLAC.

        So, please do not start an argument about your opinion about other peoples opinions. Simple fact checking will show that HRA is correct in their observation.

      • Well, we also “just know” that Donald Trump claimed that the previous president wiretapped him at Trump Towers. What you don’t know is that Hillary & Bill Clinton claimed the very same thing against HW Bush and then some! And the Clintons vandalized GW Bush’s White House: http://tinyurl.com/z7m5ytt & http://tinyurl.com/h6d5hd8
        What we are told is that Trump’s almost-national security adviser Michael Flynn worked as an agent of a foreign government. http://tinyurl.com/h6d5hd8 But he registered as. What we don’t know, and it’s true, that Hillary Clinton did likewise. But she did not register as a foreign agent. http://tinyurl.com/z9g5hc2 In other words, we do know Hillary is a criminal and Trump is a successful global businessman who had just added $3.2 trillion to the market cap.
        What we do know is, don’t throw stones if you live in a glass house.
        MQA is a stone. It appears attractive, and it is. So can be a hooker with several VD variants. Now I am not alluding any nefarious purpose to MQA – I make that clear. I am saying that it is an attempt to establish a standard to be locked in for the future, and because it is not open code, we assume at this point it is locked in, not susceptible to improvement.
        What we do know is that pre-and-post- ringing are filter issues we can deal with w/o MQA and there are a number of options for so doing – including software and firmware.
        We do know some of these have significant audible improvements that may be comparable to MQA, and may even surpass it w/o locking in a standard.
        We also know that an greedy and corrupt old lady, a chronic alcoholic who had frequent fainting spells and falls – and head injuries because of, and needed help to stand on her two feet to prevent her from fatally injuring herself, would not have made a good leader of the free world. I know many will not agree with that assessment, but then I say that’s why they make horse races, and that’s why I’m right about what ‘i had just stated.

  6. For Anirudh: I’m in the USA and downloading two classical albums from HighResAudio as I write this.

    • How are you doing this? Are you going through a VPN? I’ve wanted to download from HighResAudio but I constantly get a message saying “Dear HIGHRESAUDIO Visitor,

      due to territorial constraints and also different releases dates in each country you currently can`t purchase this album. We are updating our release dates twice a week. So, please feel free to check from time-to-time, if the album is available for your country.
      We suggest, that you bookmark the album and use our Short List function.
      Thank you for your understanding and patience.
      Yours sincerely, HIGHRESAUDIO”

      • No VPN or any other workaround, just a straightforward, plain-vanilla connection . I’ve run into the same issue you’ve encountered, though, but only with a very limited number of titles where it appears that distribution agreements preclude access to products of a specific label because they’re available for download from North American sources–as an example, the Fritz Reiner RCA iteration of Richard Strauss’ “Also Sprach Zarathustra”, which can be purchased at Acoustic Sounds. In other words, it’s not a technical issue, but rather one of licensing and distribution rights.

  7. I don’t see the reason for all the negativity, if MQA is actually a significant improvement over CD quality. I have stayed clear of high resolution downloads so far because of the cost involved of buying albums in high res quality.

    It is just too expensive to continuously buy music that way. If something gets reasonably close to high res quality for a fraction of the cost, it seems like a win (especially via tidal without the need for a MQA DAC. Having the first unfold already appears to be a healthy step up). Audiophiles that demand the highest possible high res level of quality, can still pay the additional premium at highresaudio.com or similar vendors.

    As mentioned above a pre-requisite would be that MQA actually offers a real improvement over CD quality, which I still have to test for myself.

    • “If something gets reasonably close to high res quality for a fraction of the cost…”

      That would be 16/44.1 then. Actually it gets far closer than ‘reasonably’; in fact, some folks simply can’t tell the difference.

  8. DVD-A, SACD and now MQA.

    Funny how less than 1% of the worlds population cares about digital high rez audio.

    Funny that these special formats have died….any bets for MQA not dying soon?

    • Lots of people have predicted the death knell for MQA. I understand, there have been plenty of formats promising audio nirvana in the past that have not delivered to scale. But you now have two major studios (Warners and Universal) adopting the format and various hardware and software folks coming aboard, too. I have heard Tidal files in MQA in their full glory via Bluesound Node analog out, and I really, really like what I hear. In addition, once the Dragonfly Red/Black get their MQA firmware updates, I will be able to listen in my car from my iPhone 6 to offline Tidal tracks in MQA. Now we’re talking!

  9. Good for HighResAudio.com. I’m glad someone is questioning them. Just consumers like us complaining will accomplish nothing.

    To those who say “Give MQA a chance.” Now is the time to question it. Not after it takes over the market.

    One company controlling everything is not good for everyone else. I guarantee MQA’s goal is to be the exclusive hi-res format. They probably wouldn’t survive if they only did streaming.

  10. This is clearly a serious piece of work. It would be interesting to see what MQA have to say about it. My own experience with MQA suggests I can’t tell the difference between regular 24/96 material and the MQA equivalent streamed from Tidal, so I’m happy to keep using it.

    I have one query so far for the authors of the paper. They say that MQA claim lossless compression of 24/192 material into 24/48, but as far as I can see, MQA have never claimed this, in fact Bob Stuart has explicitly stated the opposite. But this may down to interpretation of ‘lossless’ which is always difficult in this debate.

  11. Here’s my take. Stream the album on tidal listen to it on CD. If you actually like the sound of the streamed MQA file, you have your answer. I have given up on this conversation on fidelity and lost bits. To my ears MQA sounds nicer and I don’t pay for it like I do for every Hi Res down load without knowing if its is genuine ‘hires’ or not ( I have downloads from HD tracks that sound unforgivably bad when compared to their CDs). So, Whats not to like. I think, MQA streaming will shut down these hi res download sites soonish. I think they are worried. And yes, highresaudio.com does not allow downloads in Hong Kong.

    • MQA streaming won’t close high res download sites because they do not operate on the same market. People buying High res download either are not interested in streaming or if they are want to have some albums on files, in which case if buying MQA files they would also have to have an MQA capable DAC. So for downloads there is very little incentive to go on the MQA road, unless people are completely confident that MQA provides a “better” sound.
      I’m trying Tidal at the moment and listened to some MQA files, when compared to same the high res files in my library I either, couldn’t tell the difference, or MQA files sounded worse, sometimes even worse than the CD version.
      To me, MQA might be useful for Tidal if it reduces the bandwidth needed to stream High res files, but I don’t see the point for the consumer. MQA base their whole marketing on their “deblurring” process, so their target audience needs to be convinced that MQA brings better-souding music compared to CD version and usual high res versions as well. I don’t think that will happen in number large enough to insure a steady income for MQA, and I don’t think high res downloads retailers are worried about MQA.

  12. Well…I think the problems which Lothar had with MQA are solved. At this vey moment highresaudio are selling MQA downloads again , amongst them HOTEL CALIFIRNIA from THE EAGLES.. 🙂

    500+ MQA downloads can easily be found using the Advanced Search function in the menu.

    • HRA.com have yet to take their downloads offline. Doubtful that the issues raised could be resolved so swiftly, Peter.

      • They have been offline a week or 2 ago, Lothar told me he had to solve some issues with MQA. I assumed he had problems with them because of the recent start of MQA streams with TIDAL, but that seems not to be the case. I was positively surprised that highresaudio is now selling MQA downloads again. Maybe you can check with Lothar to verify?

        • This story stemmed from yesterday’s email exchange with Lothar – I’d call that as recent as one could get.

          • Thanks John, IT puzzles me..

            I hope there are plausible and audible reasons to doubt on MQA, but it seems that progress is not always appreciated. I do remember your personal listening impressions with MQA are very positive, correct?

  13. The discussion, as some are already starting, should not be whether or not MQA sounds ‘nice’. It should be if the quality is verifyable (no), if it has any advantage over other already available formats (no), if it can sound as good as non-lossy formats (no) and if the added DRM poses risks (yes).

    The original idea behind MQA was that it would offer near-CD and near-HQ quality, depending on the masters using less bandwidth. There are actually 2 different formats: de 17 bit-ish download format and a 13 bit-ish streaming format. The problem for MQA is that bandwidth is not an issue anymore: People are streaming 4K video to their phone, surely streaming a 24 bit FLAC file through a cable is easy.
    So, they tried to re market their development and that is where things went wrong.
    MQA is not a product for HRA as their customers do not want near-HQ quality, so HRA made the right decision.

    Other companies, like Linn, they do a lot of smart filtering and even volume adjustment in the digital domain. This is where MQA is still scrambled, so even if they wanted, they would have to compromise their design choices for MQA.

    Lastly, and imho the biggest issue is that Bob Stuart always claimed that (pre-) ringing is the biggest issue with music these days and that MQA was designed to battle that. The problem is that MQA, because it is a lossy format, introduces (pre-) ringing by itself.

    And lastly-lastly (promise) : Warner cannot hand-ear-remaster their complete library into MQA, that would be an enormous task and would take ages. So, it has to be automated and that means there will be dithering issues.

    • Agree. Bandwidth issues will be solved faster than updating huge catalogs. I support Schiit Audio’s motto: ‘made for the music you have not the music you have to buy’ And as it always was and will be, a fantastic recording/performance will proof itself in playback. Recently I discovered Hollie Cook (and band). Really enjoyed some youtube playback from them in the office and at home. Today I heard their stuff for the first time via Spotify in the car. Not the same of course as at home but very enjoyable!

      • Sadly I could only find one song of her on Qobuz (and a couple of remixes of the same song) so I have no idea if I like it or not. I’m not a huge fan of raggae though, but that one song I liked.

        Can I counter with the album Upsweep by Hannah Epperson? It really is very, very different from anything you have heard before and I would suggest you listen to the 1st song first, and than to the 6th, which is a different take on the same thing.
        Do not listen to it in the car, that won’t make any sense at all. 🙂

  14. @Martin Boer,

    Your post has so many inaccuracies in it! Please actually read the descriptions that are available of MQA.

    • I have read the patents, the information of their DRM provider and have had a very interesting discussion with Meridians marketing director.

      What do you suggest I read next?

      • LISTEN an you will understand… All hypothesis why DSD or WAV or MQA is better on paper is per definition providing an incomplete picture. Check the video interview of Bob Ludwig, a legendary recording and mastering engineer what his statement is on MQA. Analog releases of MQA remasters will be available soon as well, exciting!

        • That is actually not as easy as you might think. If I would make the comparison I would be listening to the difference in equipment rather than the difference in format. This is also true for most of the PCM versus DSD debates, but in this case even more so.
          My Linn streamer cannot decode MQA, or DSD for that matter, and it is so stunningly good at PCM (*) that even if I tried to be objective, PCM would always win.
          I could of course try to use a DAC designed by MQA/Meridian but I’m sure you’ll appreciate that those might have a preference for MQA over the other formats.
          I’m not even being paranoid here: every DAC has a preference.

          *) And I like its sound

  15. I also want to thank Darko for his balanced coverage of the MQA issue missing everywhere else. Balance in all things is important. It’s fair to say that most of us had swallowed, at one time or another, a bill of goods that seemed too good to be true.

  16. Viewed through the prism of the HE audio press, I find all this MQA discussion fascinating … when mp3s came out they were poopood to no end; AAC, arguably a far superior encoding algorithm, still gets poopood to no end. Fast forward through the non-standard-standard DSD (in all its silly iterations) and hires 24bit PCM format “wars”, and we ended up with MQA (2 years ago). Not one person of the press at the time (or really any other time), has ever picked up on the fact (and made the usual bone about it, see mp3, AAC, et all) that MQA is in fact a lossy format and that no matter what you do, or how you play it back, it will always remain a lossy format. Fact.

    This is only confusing the audience, confusing the consumer and adding headaches to manufacturers for the soul purpose of fixing a problem that doesn’t even exist.

    Fascinating indeed.

    Danny

  17. PS: I forgot to add that in on the manufacturing side, two highly respected audio designers did in fact call this out for what it is… Andreas Koch and Arian Jansen. They deserve to get credit here somewhere.

      • Sure… when I first approached Arian Jansen with MQA 2 years ago, he told me right off the bat – having looked at the available technical summaries – that this is compressed data and that while clever in its implementation, it would serve Meridian to replenish their now essentially defunct revenue stream from their DVD licensing days (MLP – Meridian Lossless Packaging is baked into the DVD format), similar to what Sony/Philips did with SACD when CD patents started to run dry. He did say that it would be good for music in general because even truncated and lossy, it would get the consumer to “understand”
        24/96 hires files and content quality, proper mastering permitted.

        Andreas Koch pretty much said the same thing – as a big, major proponent of DSD, he’s obviously had a vested interest in the format’s success. Alas, I think much the same: MQA is cool for getting folks into “mainstream” hires; the available stuff on Tidal sounds good to great, alas, who knows if that is because of the mastering involved or MQAs magic.

    • Because I have a high regard for his opinions, I’ve wondered what Andreas Koch thought about MQA but could never find anything. I thought of emailing him because he’s approachable and very smart. And I’ve corresponded with him in the past. But never got around to it. After reading you comment I searched and found an interview on Youtube from a few days ago.

      https://youtu.be/JPempOn1QgY?t=6m22s

      For those who don’t know Andreas Koch is. He worked for Studer Revox and Dolby Labs among others, but more importantly he worked with Sony to develop the SACD standard and designed the Sonoma DSD workstation which is probably still in use Very smart man. You won’t hear BS from him. A few years ago I questioned why his product didn’t decode quad DSD. He told me current technology at the time didn’t make it sound any better than DSD128 and maybe sounded worse. So he was holding off. I guess he’s working on it now.

      • Good to see that more and more people from the field are now calling MQA for what it is.
        I am just an amateur with an interest in technology so it is nice to know that my reasoning doesn’t deviate from someone who actually knows what he is talking about.

        The (pre) ringing thing is actually very interesting. All microphone recordings suffer from it somehow and a studio may or may not use the same kind of filtering that MQA does.
        If it does, and MQA does the same filtering again, it will actually decrease the quality.
        If it doesn’t, than MQA actually may have a slight head start.
        Things become interesting when you combine voice with electronic instruments.
        The former will have pre-ringing, the latter will not.
        So when transcoding the Warner library it will really be a gamble how things work out.
        Older music from the analogue era might actually benefit, anything from 1985 and forward probably not.

        But then comes the really, really silly design choice: lossy formats introduce their own (pre) ringing. This is just mathematics.
        And because of their choice of 13 and 17 bit formats you get dithering as well, again, applied math at work.

        I really do not understand what they were thinking there.
        The only explanation I can think of is that because of their age they made choices that would have been valid 25 years ago.
        It is like they designed a wooden wheel that has extra traction on paved roads.

    • I realized after I submitted my comment that a Quad DSD DAC and ADC is already available from Andreas Koch’s company which is Playback Designs. They use FPGA chipc instead of off-the-shelf converters.

  18. I stick to 96/24 when ever I can higher up does not make any sense. We a group of guys that use various formats DSD etc.download same albums with different Resoloution and get sometimes together for some hearing sessions and frankly most of the time we cannot hear any difference ? CD quality if proberly done can sound great up grate to 96/24 can make a difference but not all the time . MQA -let see no need to jump on every new formats

  19. Just a question: Is there a technical conflict with regard to DSP software for music, like Xivero develops and the closed environment which MQA applies for its algorithm?
    If this is the case, it is quite logical to say that these DSP software solution companies are directly threatened by the end-to-end MQA architecture. The integrity of the mathematical MQA process will not allow potentially disturbing algorithms to take place in parallel. This places the hypothesis lecture in a somewhat different perspective I would say. Disruptive technology always encounters lots of criticism and suspicious reactions in general. This seems to be a necessary proces which many inventions and new products have to undergo, meanwhile contributing a lot to broader exposure to potential customers as well. I am quite optimistic that the MQA team will respond with mathematical proof on top of the currently widely accepted audible proof sooner or later. Interesting times for audiophiles..Nervous times for some manufacturers with conflicting interests

    • In what way is MQA disruptive technology?
      This is the first time I think that disruptive is being used to describe a method to keep the old industry in power. You are really horribly misinformed.

      As for your question: if the expensive DRM that MQA has attached is worth its money, and I have no reason to doubt that, it shouldn’t be a problem to put it into software. In fact, there are already software libraries that allow DRM decoding.
      MQA, or the DRM for that matter, are really no rocket science, and yes, I would know.

      As for the mathematical proof, I can’t wait to dissect that one. It will be loads of fun.

      I cannot imagine that other manufactorers are still nervous. MQA is getting enough flak already and I’m sure they’re already looking for a suitable exit strategy.

      • As soon as third part DSP processing will interfere the MQA encoded signal, the decoder will be unable to deliver Master Authenticated Quality to the end user.

        If hardware manufacturers have technical limitations to apply MQA certification onto their DAC, their consumers will not be able to enjoy the benefits of listening to deblurred music from both old and new albums.

        Since the audible improvement of MQA has been confirmed by many normal consumers and audiophiles worldwide, the popularity will continue to grow.

        It is therefore quite obvious that forthcoming standardisation to MQA as the new and widely accepted reference method for streaming HD music, will have a large impact to various soft & hardware manufacturers and can therefore be regarded as disruptive.

        The fuzz & buzz by persons and companies who clearly are afraid of loosing market share or technology is confirming this. Lucky for everyone: Original Mastertapes will stay what they are: Original, so future technology with even better restoration capabilities will still get access to ‘The Original’

        MQA is just a very smart technology which wraps and filters out past- and current A/D, D/D, en D/A errors and with that it is intrinsically ‘lossy’ BECAUSE it gets rid of artefacts which we are not interested in. It is simply improving the sound convincingly, offering best of both digital and analog which no other technology is able to offer at the moment.

        So let’s just wait and see, but especially LISTEN how this will evolve. All controversy and hypothesis papers are actually just contributing to broader interest, so that’s fine for MQA and consumers who have convinced themselves just by listening to it.

        • It would have been nice if you had written somewhere that you’re a sales representative for Onkyo, a company that sells equipment with MQA.

          Care to continue this discussion?

          • Peter – is this the case? That you rep for Onkyo or are affiliated with them somehow?

          • Well..Thanks for finding this second Peter Veth for me guys..! This is a stunning voincidance, although I have heard many years ago that someone else with the same name is indeed working in the consumer electronics business.. I recently read an article where funny matches between jobs and names are mentioned. Somehow there must be some relationship..I work in the chemical industry selling analytical equipment for a German company. But audio and music is my passion like most probably my counterpart will be as well. I will tryto get in touch with him the next audio show

          • Hi John H. No, I am not affiliated with any audio or consumer electronics company in the world. Just an enthusiastic early adopter of MQA

          • I can assure you that I am not in any way related to Onkyo or MQA or Pioneer, Mytek, DCS technologies, MSB , Bluesound, NAD, or any other company who adopted MQA in their productrange.

            I am just an early adopter of MQA ever since I auditioned it during high-end demonstrations over here in The Netherlands. In my home situation, I am able to enjoy the TIDAL MQA feed and some purchased MQA albums on a NAD C390DD with BlueOS module and vintage Wilson Audio Watt Puppy 5.0 I am aware thisis not Top-End gear like you own with your Linn setup, but I am satisfied and still impressed by the excellent match I achieved in my livingroom. Once and a while I ‘recalibrate’ my mind and ears during audio shows, where I hope to become flabbergasted by something much better than at home. So far, the difference has never been that huge that big changes need to be made.
            But the MQA effect is audible, it is there and I like it and cannot wait to hear more and am much looking forward to my all-time favorites albums from Pink Floyd and Supertramp and others.
            MQA is no marketing scam or fake news. Bob Stuart and Peter Craven did a great job and the IP of their invention is protected by many patents and this closed encoding – decoding container which causes lots of annoyance and suspicion. But that is part of the business. Vinyl lovers have their own convictions and believes..Nothing new in audio..

        • Peter – being an industry representative you better than anybody know that the “authenticated” part of the MQA story is the biggest BS of it all.

          What does authenticated here mean? In the MQA-business-chain it means that a mastering engineer sends the file to another site where it gets encoded & encrypted in MQA format.

          The MQA encoding process is lossy and in most cases does not happen in the studio where the mastering engineer made his sonic adjustments.

          So the MQA file is actually “unstudioed”: it is not authentic anymore as it is a.) lossy, b.) not encoded in MQA by the original engineer and c.) not in the same room and on the same equipment where the original engineer worked.

          The only thing that gets authenticated is the fact that somebody paid MQA-fees and the file can be “unfolded” on devices build by manufacturers that paid MQA-fees too.

          MQA is a rent seeking business. And Onkyo obviously has decided to pay rent and hence needs to defend MQA. Attacking competitors who don’t is really bad form.

          MQA marketing begins to crumble and you are a very nice example of it.

          • Dear Mcgillroy,

            Interesting to notice that you and others in this thread respond in the same way I did when I read the articles by highresaudio, Xivero and Mr Koch. I also checked their backgrounds and credentials and discovered that there are obviously business related reasons for their negative arguments against MQA.

            But let’s just take a step backwards, so I can explain first of all that I am in no way related to any audio or consumer electronics company at all. Funny enough, someone carying the same (rare..) Dutch name is working for Onkyo, so I will be looking forward to meet this person somewhere in the future.

            Secondly, yes..I am potentially overreacting a bit towards the aligations against MQA and the doubts towards the integrity of the developers Mr. Bob Stuart and Peter Craven. Besides the very convincing sound improvement MQA is showing, I also have high respect for these persons and their love for music and their unique invention.
            Ever since CD came in my home, I noticed harshness, digititus, less ‘being there’s than is achievable with the best vinyl or analogue recordings. From many years, the problem was related to jitter as being the nr1 enemy. So DAC manufacturers started a rat race, focussed towards better clocks, interfaces, less jitter.
            MQA is, however, much more fundamentally solving the time-smear problem which PCM still has even at high sample rates. This is a fundamentally different approach and it works flawless.
            As a consumer an hobbyist, I am only interested in this breakthrough, which offers me enlarged listening pleasure. Especially if such a technical solution is provided free of extra costs with my Tidal streaming service, it is a no-brainer to embrace and enjoy it.

            For purists who only can and will accept ‘The Original’ PCM or DSD files in full glory there is no real problem. Streaming services not offering MQA are there and these original formats will still continue to exist.

            So why all these concerns against MQA..? Since it is a business model..? Since it cannot sound better than ‘The Original’..? Well, just do not consume it than I would say. Most audiophiles also stay away from MP3, so just stay away from MQA if other solutions appear more to you.

            For me and many others, there is no debate, just increased listening pleasure.

        • Your opinion of MQA seems a bit over the top.

          I think audiophiles have heard and read more than enough to have serious doubt about MQA and their motive for the it’s existence.

          That leaves average consumer who, in my opinion, is the true target audience of MQA. In his article Andreas makes a case that the better doesn’t always win. And I think MQA knows their format isn’t better. But I think think marketing MQA to the average consumer is a tough sell, because they don’t care about sound quality and don’t want to pay a premium for it. Thus I think the only way MQA can succeed is if there’s no alternative.

          • Hi Larry,
            Just started the weekend with a cup of coffee, listening to Pink Floyd ‘Cluster One’ As soon as the cymbals start playing subtely, MQA proves it’s magic to me once again. Natural, less grain, improved articulation, just easier to listen to.
            It’s not that I cannot listen anymore to non MQA encoded music anymore, but it’s for sure addictive.

  20. Isn’t MQA really dolby digital with some secret sauce on top that makes record labels feel more secure about releasing high quality master tapes without the fear of exact studio master copying. Its DRM handholding for record labels while meridian pockets the cash…

    • Hi Gzubeck,
      MQA is indeed a ‘secret sauce’ as you name it, since the IP is patented and no one ( expect for the MQA team) knows how the algorithm is capable to upgrade the original PCM Master files by extracting ‘time-smear pollution’ out of it. This proces is intrinsically changing the original music file and for many purists this is a ‘No Go’ since the original is sacred and always the best for reference. Maybe we should re-think the debate and compare MQA what is being done allready for many years with photography, where digital enhancement and deblurring technology is allrady widely accepted and even available on our GSM cameras. It is able to improve the sound of the original PCM file, but this can only be judged and conformed by yourself by listening to it and do an extensive A/B comparison over a longer period of time.

    • Hello Peter,
      I have no problem with opposing opinions, lack of knowledge and trust in marketing blurbs however are another matter. I have really read the MQA patents and it really is troublesome.

      Video deblurring is a method you use when the source material is in a bad shape.
      If you have a good source, be it 24/48 or higher or even 16/44.1 there is no reason, and I’d say excuse, to mess with it.
      Downgrading the source to save bandwidth in 2017 and using devlurring afterwards is just plain stupid.

      The method that MQA uses to ‘improve’ the source by compansating for microphones will never work for a combination of voice and electronics. The voice might gain a little, the keyboard will lose a little. It is best left untouched; the studio engineer already figured out how it should sound.

      I have deblurred satellite data during my astronomy study, and have written software to do so, so this is all close to home.

      • IMO the reasons for the lossy compression are selling points. The record industry equates compression with not giving up the crown jewels. And asking the industry to adapt a standard which consists of deblurring alone would be a tough sell.

        How can you degrade the signal one way and claim you’re making it better with another? How can it be better while not giving up the crown jewels?

      • Hi Martin, the point is that many PCM recordings and Mastertapes, especially those from the early days of digital recording, are actualy blurred by pre- and post ringing effects. How MQA is capable to filter these errors out and compensate for time-smear effects can be found to some extend in their patents, but they do not disclose exactly how it is done. Probably the algorithm uses a new type of apodizing filter, including inverted modes to correct these errors. If even very experienced recording engineers confirm that the MQA version sounds better than their own mastertapes, it is proof to me that there is much more behind the technology than a marketing scam to sell new compression technology for streaming purposes. But the most important and convincing proof is my personal experience with MQA, one which I am able to distinguish flawlessly in A/B comparisons.

        Also experienced audio journalists like John Darko himself know what MQA is capable of. Quote from his review on MQA in june 2016 http://www.digitalaudioreview.net/2016/06/an-inconvenient-truth-mqa-sounds-better/

        ” In my digital audio library sits Gaucho as original CD rip, remastered CD rip and as (remastered) hi-res download. Why then would I want to keep hold of Gaucho’s first cut in MQA?
        Put simply: because it sounds convincingly better than the normal, non-MQA’d 24bit/96kHz file. Lest you thought MQA was just a way to pack, transmit and then unpack hi-res audio via what Bob Stuart calls “audio origami” or “encapsulation”, it isn’t. It’s that. But also more”

        I have been asking several journalists around the world to do some comparative measurements of impuls responses of MQA dac’s. This would be interesting material I suppose. The only problem is that we will need MQA encoded testtones for this. Maybe one day they will become available, it will clarify some of the mystery around the deblurring capabilities of MQA.

        • As MQA does not release those testtones you should ask yourself -why- that is.
          I am as gullible as the next guy, but the fact that MQA is not willing to produce proof says a lot.

          I prefer my music to sound as real as possible, the fact that you mention it improves the quality of early digital recordings sends shivers down my spine. Audio equipment should playback as faithful as possible. What you are describing is horrid.

  21. Well Martin, opinions and appetites vary, also with regard to the perception of sound quality and technological advancements.
    Many audiophiles did not appreciate the sound of digital audio at all when the Compact Disc was introduced. They stuck to the analogue LP and the sound of vinyl even untill today. Ever since digital recording was introduced, analogue music was chopped up into bits by 14 or 16 bits A/D converters and PCM encoding. The digital ‘glare’ decreased over the years when higher resolutions and sample rates were introduced. Then followed by DSD and the format wars proceeded. DAC manufactures nowadays apply femto clocks is upsampling technology etc. In order to decrease jitter artefacts and improve the last step of the conversion into an analogue sinewave. This evolution was incapable to improve the temporal resolution of the polluted original, untill MQA came. You might not believe it works or even cannot hear a positive improvement. As mentioned several times, others like me have a very positive experience. With that, I am not promoting MQA as being the one and only technology for archiving new recordings. I agree that highest resolution DSD or DXD or even 24/96 PCM is still best available technology for the moment. But maybe a non-folded MQA recording at 24/192 or 24/384 or higher is a future proof archiving method as well. Let’s just see what will happen. As long as there are many beautiful non MQA DACs on the market and reproduction level is excellent, why change? What fascinates me though is why a MSB technology DAC of 90K Euro or or DCS DAC in the same price range are adopting MQA. One of these days, I hope to be able to listen to the differences. Cheers!

    • Hi Peter
      As you mention 24/96 is at least for me more then enough any more resolution adds more noise (as stated by many experts in the field )beside that some audio friends of mine including my self did some music .hearing sessions >same music with different resolutions from normal CD up the ladder until DSD .(In some cases we had 4 to 5 various resolution )As we had decent equipment switched from Tube Amplifier to Dual Mono Amplifier all of us could not hear a difference .Only one person switched the Music and he did not tell us the resolution meaning we had an open mind therefore it was a blind hearing session as it should be ( Even with top of the line HP we could not detect a noticable difference ) except the switch from 44.1 to 24/96 did produced sometimes a better result etc. I understand that some people have a trained ear to hear the slides variation particular the audio writers but just selling more resolution to making more money is doubtful as most existing music material is not recorded in a higher resolution therefore it is beefed up without much improvement .
      Even Mr Waldrep from AIX studio that produces really excellent high resolution recordings admits that 24/96 is more then enough .Sometimes people mix up better sound with different sound if you get my drift .I keep an open mind in regards of MQA but aim getting tired to switch formats and have to invest again for a new DAC .Aim sure after while the industry will come up with new gadget on how you can improve even further the MQA which means more money .After over 50 years listen to music and went through all various formats I came to the conclusion that most formats can sound great if the recording engineer does a great job .

      • Hi Roland,
        Please understand that MQA is not a new format. It is a sophisticated DSP algortithm which corrects for pre-and post-ringing effects which are intrinsically present in all digital filters. Even when recording at 24/96 or 24/192 PCM, this unnatural audible blur will be captured onto the final Master of digital records. CD quality has a temporal blur of 5 ms, 24/192 1/10th of CD (500 micrseconds), but MQA has found a method to reduce this to 10 microseconds throughout the entire digital encoding and decoding chain.
        MQA is ‘transported’ via a normal .flac or .wav file and fully compatible with non-MQA DAC’s

        • Aim aware of the technical aspect as I did some reading on the topic all what aim saying aim happy with 24/96 not need for me to go higher or to any other format or MQA . Aim at a stage of my life were aim happy with what I have in regards of music and equipment otherwise it is a never ending story which I followed the last 50 years .

        • Peter, I think it’s super that you’re having a great experience with MQA, I really do. But before you write to much more about the technicalities regarding de-blurring and MQA patents you might pay a visit to computeraudiophile, where quite knowledgeable people from both camps have chipped in. The delivery chain within the MQA container have been clarified and “reverse-engineered”. The conclusions is largely consistent with Stephan Hotto’s.

          Now, that shouldn’t take away from your enjoyment of MQA. But a lot of grand claims are being thrown around by Bob that simply is at best alternative facts.

          I think that’s why some people are having just a bit of doubt with regards to this “revolution” in music streaming technology.

          Cheers.

          Tyge

    • The differences heard probably have more to do with the recording and mastering than MQA itself. I’ve mentioned this before. I remember reading an article in Stereophile where. John Atkinson couldn’t reliably tell the difference between MQA vs. non-MQA versions of his own recordings.

      I doubt there would be any archiving using “unfolded MQA” because it would no be longer MQA, would it? Just apodizing filters which anyone could do. And probably already do in some cases.

      MQA is more or less is a patented collection of processes or operations which are applied to the files and licensed for revenue.

  22. The mystery of how our brains perceive sound has deepened, now that musicians have smashed a limit on sound perception imposed by a famous algorithm. On the upside this means it should be possible to improve upon today’s gold-standard methods for audio perception.

    Devised over 200 years ago, the Fourier transform is a mathematical process that splits a sound wave into its individual frequencies. It is the most common method for digitising analogue signals and some had thought that brains make use of the same algorithm when turning the cacophony of noise around us into individual sounds and voices.

    To investigate, Jacob Oppenheim and Marcelo Magnasco of Rockefeller University in New York turned to the Gabor limit, a part of the Fourier transform’s mathematics that makes the determination of pitch and timing a trade-off. Rather like the uncertainty principle of quantum mechanics, the Gabor limit states you can’t accurately determine a sound’s frequency and its duration at the same time.

    13 times better
    The pair reasoned that if people’s hearing obeyed the Gabor limit, this would be a sign that they were using the Fourier transform. But when 12 musicians, some instrumentalists, some conductors, took a series of tests, such as judging slight changes in the pitch and duration of sounds at the same time, they beat the limit by up to a factor of 13.

    This shows that the Fourier transform is not the whole story, says Magnasco. “The actual algorithm employed by our brains is still shrouded in mystery.”

    Brian Moore of the University of Cambridge says he is not surprised that the musicians beat the limit: he already assumed that other mechanisms were at work.

    Understanding human sound perception could inspire better systems for sound recordings, speech recognition and sonar.

    https://www.newscientist.com/article/dn23143-musical-brains-smash-audio-algorithm-limits/

MQA comes to Audirvana Plus

Pause for the jets