in , ,

Munich High-End 2015: Is MQA the new DSD?

  • munich_high_end_2015_logoMaster Quality Authenticated. Back in December, Stereophile’s John Atkinson reported on how he’d the heard the future of streaming. It was called MQA and it had come from the mind of Meridian’s Bob Stuart. Essentially a zipping mechanism for hi-res audio, MQA folds the hi-res portion of the file under the noise floor of a Redbook-sized container. Any “MQA-ready” DAC could then decode the complete hi-res data stream whilst standard DACs were still able to play back the file in CD quality.

    The implications were substantial. Hi-res audio streaming was no longer a congested pipe dream – at least in theory. As one might expect, Meridian and Tidal are good friends.

    “Sounds. Too good. But true”. This 3 x 1”-punch phraseology announced Meridian’s MQA at Munich High-End 2015. I had to read it twice as I sat down for one of their scheduled demos. I was eager to hear its benefits first hand.

    Meridian’s “JJ” demos MQA at Munich High-End 2015.

    Our demo host was Peter Kay soundalike “Jay-Jay”, an affable chap from the north of England. His employer is from the southeast Midlands – Cambridge to be exact. The hardware behind the software demo was Meridian’s own high-end actives: their DSP8000 floostanders roll up D/A conversion, amplification and air movement into an all-in-one loudspeaker system.

    MQA is apparently based upon three principles, the first of which is timing. Nyquist-Shannon tells us that when deocding a Redbook feed a digital filter is required at just above the 22kHz marker. Unfortunately, these filters ring both before (‘ghosts’) and after (‘echoes’) the transient, thus smearing temporal information and confusing the listener’s ear-brain.

    Meridian believes humans are incredibly sensitive to time data and MQA permits A-to-D conversion using processes that don’t ring and therefore don’t smear proximate events.

    Which brings us to MQA’s second principle: ‘bad origami’. Data up 100kHz is folded down under the noise floor of MQA’s Redbook-sized FLAC container. DACs sans-MQA will still see the Redbook portion of the file and decode it as per usual.
    That much we already knew.

    For MQA to propagate at the consumer level then Meridian will license their software solution to hardware manufacturers who can then roll it out via firmware updates or new models. We’re told that this is well underway with some ‘big names’ already signed on.

    MQA’s third principle is ‘authentication’. The mastering engineer can digitally sign his work as MQA authorized. The implication being that for a file to be MQA encoded, one must go back to the studio’s master tapes.

    That was certainly the case with the three pieces of music heard at Meridian’s M.O.C. demo: Frank Sinatra in mono, a classical piece from Norway’s 2L and a fresh entry into the book of cliché demo music: a cut from Daft Punk’s Random Access Memories.

    Meridian’s Bob Stuart at Munich High-End 2015.

    As Bob Stuart himself enters the room I ask, “What about music for which the studio master isn’t available? What if all you have is a standard 24bit/192kHz PCM file?” Here, Stuart himself speaks up: MQA can repair the damage exacted by a non-MQA conversion process, especially if the studio specifics are already known. Sometimes the A/D convertor and its digital filter characteristics are known, sometimes Meridian must measure the A/D converter used and other times a best guess must be made.

    Without a back-to-back comparison with the original file it’s impossible to judge the size and nature of MQA’s bettering of existing studio encoding methods.

    Moreover, Meridian’s supplemental claim – that MQA’s non-ringing encoding process will sound slightly superior to a standard CD – remains unverified.

    Audiophiles like you and I need to hear first hand the sonic benefits of MQA material – from studio encoding to consumer DAC decoding – and compare them to the files which we already own. In Munich, that didn’t happen. On the plus side, we thank our lucky stars that no-one uttered the words “Just as the artist intended”. That thought then triggered the question: is MQA why Neil Young walked away from Meridian’s partnership with Pono?

    Hi-res streaming applications aside, the danger here is that Meridian’s encapsulation process could become just another audiophile-centric thing. Its success is predicated on mastering studio take-up resulting in a software catalogue large enough to sway the listening public into seeking out an appropriately equipped D/A converter.

    The pivotal issue here is studioland’s willingness to take on MQA as a de facto encoding standard. Will it see broad acceptance or will it remain relegated to a niche of ‘audiophile’ labels like 2L? Much will no doubt fall to Meridian’s MQA encoding licensing fee structure but with that currently an unknown, are we instead about to witness marketing bods and journalists stir up talk of the next big thing such that 1) consumers ask if their next DAC purchase is MQA ready and 2) DAC manufacturers load in MQA for fear of missing out on a sale, both long before the arrival of a decent-sized library? Just like DSD.

    Further information: Meridian MQA

    Munich High-End 2015 coverage sponsored by LH Labs:


    John H. Darko

    Written by John H. Darko

    John is the editor of Darko.Audio, from whose ad revenues he derives an income. He is an occasional contributor to 6moons but has previously written pieces for TONEAudio, AudioStream and Stereophile.

    Darko.Audio is a member of EISA.

    Follow John on YouTube or Twitter


    1. I think the temporal thing is the most important part of MQA and I imagine it’s a requirement in the A-D process more than any data reduction scheme. The way it’s described, it seems better than current hi-res. Not sure why there wasn’t something like this before now with hi-res. Are they going to remaster everything again?

      • I don’t think it’s a matter of remastering but one of re-encoding. Still, either way, that’s a lot of work. And the bigger questions remains: will we be asked to purchase titles again?

        • According to Meridian, the answer is yes. MQA enabled DACs will do nothing for non MQA content. Much like HDCD in the 90’s. You will need to get an MQA files and an MQA DAC for the full MQA experience…

    2. I’m always open to new standards in recording, mastering and delivering music, as long as they are genuinely superior to what came before. I am willing to spend a bit more on those superior techniques. But – you are absolutely right – we need more proof of superiority, more music licensed through MQA and more DAC manufacturers that license the MQA technology.
      BTW, it is now clear to me, why Meridian backed off from the deal with Neil…

    3. I attended the MQA demo at the Chicago show last April. The demo was filled with many buzzwords and marketing language. At the end of the day, Meridian found a way to hide additional data in 16/44.1 files. All the talk about timing, tracing back to the source and improving the original are sales talk. I see this as a good solution for streaming data from the web, although owning a file locally on my music server, I could not care less about the size. Oh, and with regards to DSD, MQA works only with PCM encoded data, so no love for the DSD folks out there.

      • I agree: great for streaming but files hosted locally won’t benefit as much unless there’s a substantial SQ difference between what I have already and the MQA-encoded version.

        • Basically, the big question is does MQA sound better than 192khz PCM? If so then they’ll might have something there. Otherwise, I don’t really care. DSD seems to have faster transients. I guess from the higher sample rate used. If MQA could give the transient speed of DSD without the noise problem of DSD64, I’d think it will be somewhat successful. At least among audiophiles. What I fear is if it’s almost as good as hi-res PCM and major labels use only MQA to distribute everything.

          • I went to a demo in NYC. One of the tracks was a 192/24 file encoded with MQA and played back. The MQA’d file sounded better than the original file – and by a wide margin. This simply means that:
            1- MQA does not keep the data intact – unlike what the presenter argued
            2- There’s some euphonic processing of the file
            Now, I am all for good euphonic filters – consider HQPlayer for example – but I am deeply disappointed by marketing hype that is so blatanty untrue.

        • MQA should simply be used as a method to stream music – ie encode-transmit-decode. If it is successful at that it will be great. Other claims like ‘re-encoding master tapes’ etc are lofty dreams that have zero probability of materializing – in my opinion that is.

          • Yes, its application for streaming hi-res looks solid…but I have my doubts about anything that necessitates cultural change at the studio level.

          • It could be used for streaming only. But why would major labels go through the trouble of re-encoding with MQA just for streaming? Unless I’m not understanding the correctly, I would think they want to do it roll the tape and do a digital transfer only one time for all distribution.

            • I don’t see the motivation for labels and studios to re-encode to ANY new format. I suspect the majority are quite happy with the way they do it now.

    4. Firstly, I’m all for “euphonic enhancements”, if proper. However, I’m completely against BS like “as the artist intended”. I went to an MQA audition in Manhattan – it sounded great. Even a 192/24 file processed by MQA sounded better than the 192/24 file itself! Hence my comment on “euphonic enhancements”. But all the blabber about “master tapes”, “audio chain”, is simply disingenuous. I asked whether MQA encoding could be tuned based on the audio chain used – I was told “no”. I think Meridian created an interesting music folding algo – that’s great. Trying to pass it for more than it is not quite honest. This is the HDCD of 2015. If not licensed widely it will die quickly.

    5. Hats of to meridian who have always been a forward thinking company but I feel they will fail on this one, this is another SACD waiting to happen. We should applaud companies who push the boundaries but sometimes the will to succeed is not always enough as you need the public swell to carry you.

    Antipodes Audio DX music server review

    LH Labs show prototype Geek Out v2 at Munich High-End 2015