Tidal adds hi-res audio streams with 'Tidal Masters' (with MQA)

Discussion in 'General Audio Discussion' started by weldp, Jan 5, 2017.

Tags:
  1. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    Don't take this as a defense of MQA ... it's not ... far too early for me to say anything one way or another about it from a realized sound quality (or potential lack-thereof) perspective, but one thing to note is that while MQA is generally described as a "lossy" compression system it is not "lossy" in the same way that MP3, AAC or Vorbis are (perceptual encodings that play around with, and do all sorts of nasty things to, FR and time-domain information).

    Simplified, drastically, if you MQA encode a file, regardless of all the temporal correction stuff, you wind up losing three effective bits (the least significant bits) of dynamic range. If the content in question has 13 or fewer bits of dynamic range actually utilized, the encoding is lossless even if played back on a non-MQA enabled player (software or hardware). With an MQA enabled player, either software or hardware (in your DAC), then you lose nothing and all 16-bits of dynamic range are recovered.

    On non-MQA playback chains using MQA-encoded files (e.g. running current TIDAL MQA output with "Passthrough MQA" enabled into, say, an Yggdrasil or Modi MB) that will, in some cases (i.e. those where the source has more than 13-bits of dynamic range utilized), result in dynamic-range compression. You all know about this ... it's a common thing ... the root of the "loudness wars" and it is important to be aware of because, in general, initial perception of music that is dynamically compressed, especially to an unfamiliar ear, is that it sounds "better" (same way that "louder" is generally perceived as "better").

    If the player/DAC is MQA enabled, dynamic range should not be affected.

    I'll refrain from comments as to whether this is good or bad, from an audio-end-result point of view. And there are myriad other issues (good or bad) from the business-model on down, but I wanted to try and provide a bit of clarity as to what is/isn't going on with MQA with more clarity than most of the "press" nonsense has managed to date.
     
  2. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    In another direction ...

    The concept of "DRM" as it relates to MQA is also not directly comparable to traditional "DRM" schemes. Back when all music downloads were copy-protected it was not, without "cracking" the encrypted files, possible to play those files outside the player/eco-system they were intended for. If you bought a protected song on iTunes, then you were stuck with iPod or iTunes playback (or cracking/hacking the file).

    With MQA the encoded files will always play on a non-MQA chain.

    They will do so subject to the dynamic range "issues" described in my previous post, but they WILL play.

    You don't need anything special to playback an MQA file. They're all in native lossless containers (FLAC, ALAC, WMA-Lossless). Anything that can play those formats will play an MQA encoded file without additional limitations.

    The only thing MQA can "hold hostage" from a "DRM" perspective is the ability to unfold the additional encoded data, source authentication, and "benefits" (real, imagined or made-up) of the MQA processing. And, again, this is only even a potential issue if your source material was using more than 13-bits of actual dynamic range in the first place.

    It's surprising how often modern releases don't get NEAR to using those 13-bits.

    ...

    The practical upshot of this is that, whether MQA takes off or not, and even if it takes over the market (rather doubtful) you do NOT need to buy any new hardware to still be able to play that content. It'll play just fine on your Modi MB or your Yggdrasil, or Modi 2 for as long as you have them.

    ...

    Again, I have no vested interested in MQA. I run a variety of Schiit DACs, a NOS R2R unit, and a couple of other odds and sods, and am not planning on changing that. This year is going to be a vinyl/analog year for me!
     
  3. Dino

    Dino Friend

    Pyrate
    Joined:
    Jun 21, 2016
    Likes Received:
    3,332
    Trophy Points:
    113
    It is possible. I wouldn't worry about it, personally. I have always went with what gives me more pleasure, regardless.

    One example is my preference (all along) for LP sound over digital sound. I had so many people want to debate that preference. They had so many facts and numbers to prove CD/digital sounded better. All I had was descriptions of my experience. Sometimes they would get so frustrated with my stubbornness to facing facts. Things would really escalate if I would mention LP playback also feeling better. :) Now my preference does not seem so weird.

    Anyway, if you feel Hi-Res sounds better to you, then it does, IMO.
     
  4. Dino

    Dino Friend

    Pyrate
    Joined:
    Jun 21, 2016
    Likes Received:
    3,332
    Trophy Points:
    113
    Yeah, people hear DRM and go off. From my (limited) understanding of MQA it is similar to HDCD in several regards. I don't remember anyone saying that HDCD included DRM. I suspect that Meridian just uses the term DRM to entice the record companies.
     
  5. yotacowboy

    yotacowboy McRibs Kind of Guy

    Pyrate Contributor
    Joined:
    Feb 23, 2016
    Likes Received:
    10,695
    Trophy Points:
    113
    Location:
    NOVA
    Home Page:
  6. Dino

    Dino Friend

    Pyrate
    Joined:
    Jun 21, 2016
    Likes Received:
    3,332
    Trophy Points:
    113
  7. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    You're correct and a comparison to HDCD is the best way to think about how MQA "behaves" when played on a system that doesn't have native MQA support from a "DRM" perspective. Though Meridian claim a benefit even without an MQA decoder in the chain.

    He's absolutely right, of course.

    I built a playlist of MQA and non-MQA tracks in advance of having an MQA enabled DAC to do some comparisons with and it becomes quite apparent that different masters are used in both cases. In fact, at least for the stuff I've got in that playlist so far, I'm not sure ANY of the MQA tracks have a corresponding non-MQA version from a common master ... which makes proper comparison impossible - at least from the content available on TIDAL.

    So, right now, I'm limited to testing how MQA encoded content sounds through a non-MQA DAC processing the raw MQA PCM file versus the same DAC and software-Level MQA unfolding. Tomorrow I'll also be able to compare MQA processing in software to a native MQA DAC. But what I can't do at the moment is a proper Redbook to MQA or hi-res to MQA comparison, beyond a handful of files that are mostly stuff I don't want to listen to in the first place.

    And, as @baldr says, it is both inconvenient, and a bit fishy, that such comparisons have not been more easily facilitated. If MQA is as clearly superior as Meridian, and much of the press, claims then they should be making such comparisons as broad and easy as possible for their potential customers.

    Of course, if they (Meridian) see their customers as studios, labels and manufacturers ... which, directly at least, is very much how it seems, that's a very different proposition to regarding the music-buying public as your main customer.

    The most positive thing I can say about MQA at the moment is that it already has more content, and content of relevance, than is available for native DSD. But that's still a bit like saying you'd rather be stabbed with a fork than a knife.

    And the best advice I can give is that, right now, if you were considering a non-MQA enabled DAC purchase last week then there's no reason at all to change that right now. If I was buying another DAC tomorrow, it'd still be an Yggdrasil.

    Finally, at least in the case of Schiit's DACs, if MQA were to deliver on its promise, AND all relevant content was easily available, and if, as a result, Mike wanted to directly support it, it's worth remembering that three out of four of his DAC products can be upgraded to do so without you having to buy another DAC.

    That makes Schiit's gear even MORE compelling, to me, today than it was BEFORE TIDAL added MQA content. It means I don't have to worry about it at all. The content will play either way. If it tanks then I've lost nothing. If it succeeds I can upgrade what I have. And in the meantime I can enjoy the music without having to play "wait and see".
     
  8. Dino

    Dino Friend

    Pyrate
    Joined:
    Jun 21, 2016
    Likes Received:
    3,332
    Trophy Points:
    113
    Great work (as always) Torq.

    HDCD claimed to sound superior to CD even when not decoded. I recall reading that numerous times, but I am not finding a source, ATM.
     
    Last edited: Jan 8, 2017
  9. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    Thanks!

    Now that you mention it, I recall reading similar things about HDCD encoded content. I'd have to go research the specifics. I do know that a conceptually similar "special dithering algorithm" is used in both HDCD and MQA to bury the "extra" encoded data in noise. MQA uses a different implementation (and many other techniques) to do its thing.

    But another thing common to both is that there can be measurable (and possibly audible) degradation on a non-supporting replay chain that is dependent on the source material. In MQA's case the first instance for this is with source content that uses more than 13 bits of dynamic range. It's not clear, yet, if the dithering has other low-level audible consequences.
     
  10. Kattefjaes

    Kattefjaes Mostly Harmless

    Pyrate
    Joined:
    Sep 5, 2016
    Likes Received:
    4,521
    Trophy Points:
    113
    Location:
    London, UK
    Here's @baldr's take, posted on HF:
    Do you know what? I'm going to go with what the guy who designed my ridiculously nice DAC thinks, for now.
     
  11. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    For what it's worth*, once this LP has finished playing, I can now start comparing software-decoded MQA to the results from an MQA-enabled DAC ... as the Explorer 2 just arrived:

    Explorer 2.jpg

    *May not be worth much at all given the lack of obvious content from a common master that has both regular and MQA encoded versions available.
     
  12. Wfojas

    Wfojas Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    848
    Trophy Points:
    93
    Location:
    CA
    I do think the noise of MQA can die down when the players integrate it, as Audirvana has announced.

    https://audirvana.com/?p=4153

    At that point, the arguments will be to the format it was released, which, I guess it the core of it. To measure the merits of MQA with from different mastering is useless.

    To measure MQA encoded vs nonencoded from the same masters then falls in the provenance of finite numbers. If the numbers don't show anything (and they shouldn't, it shouldn't really) the suckers and bad subjectivists will pick camps, objectivists will go on their merry, condescending ways, and I will read reviews from people whos ears I trust. Anything night and day in digital falls in the wondrous alchemy @baldr and Berkeley Audio traffic in, as their long track record of success shows.

    Interesting that Audioquest will offer a software upgrade to support it on the Dragonfly, interesting but unknowable as to the end result, for now.

    http://www.theabsolutesound.com/articles/mqa-amplifies-hi-res-audio-activity-at-ces-2017/
     
  13. Dino

    Dino Friend

    Pyrate
    Joined:
    Jun 21, 2016
    Likes Received:
    3,332
    Trophy Points:
    113
    Last edited: Jan 9, 2017
  14. trung225

    trung225 Facebook Friend

    Joined:
    Oct 22, 2015
    Likes Received:
    123
    Trophy Points:
    43
    From what I have seen from this report, audirvana can decode but only part of MQA file. If you have Hires music encoded to MQA and play with non-MQA DAC, all information from 17 to 24 bit will be lost. If you have Hires music encoded to MQA, and play with MQA DAC, the rest information will be decoded by hardware decoder on DAC
     
    Last edited: Jan 9, 2017
  15. Wfojas

    Wfojas Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    848
    Trophy Points:
    93
    Location:
    CA
    @Dino, there are comments? :) Audio can be such a war zone, lol.

    Cursory listening over Roon comparing CD rips of Christopher Cross and the MQA release that Tidal can now stream in Roon, you can tell its remastered. But it does sound much less congested and is more listenable. I haven't checked loudness and compression, which I suspect is in play here for the mqa instance.
    Because there is no mqa decoding, the source itself is different, for sure.
     
    Last edited: Jan 9, 2017
  16. Wfojas

    Wfojas Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    848
    Trophy Points:
    93
    Location:
    CA
    I wonder if that's a limitation Audirvana had to comply with to get the MQA seal of approval. In theory, all of it can be done in software.
     
  17. trung225

    trung225 Facebook Friend

    Joined:
    Oct 22, 2015
    Likes Received:
    123
    Trophy Points:
    43
    If you are Mytek, Audioquest, Pioneer, Onkyo, the brands have paid the license fee for Meridian to acquire MQA hardware-decode, do you accept the fact that MQA can be decoded fully on software? No. So this is a clever move (but a bad way to music lover)by Meridian, everyone can play MQA files, but if you have full content, please buy a MQA DAC, and we (Meridian) will receive the license fee.
     
  18. Dino

    Dino Friend

    Pyrate
    Joined:
    Jun 21, 2016
    Likes Received:
    3,332
    Trophy Points:
    113
    I was just getting ready to post something along those lines @trung225 . Meridian can keep software and hardware revenue streams going this way (along with their other sources of revenue).

    I don't mind if this all turns out to be a goldmine for Meridian, if the sonic improvements turn out to be spectacular. I am concerned about one company having so much control over such a wide spectrum of audio though (assuming that MQA is a total success).
     
    Last edited: Jan 9, 2017
  19. Kattefjaes

    Kattefjaes Mostly Harmless

    Pyrate
    Joined:
    Sep 5, 2016
    Likes Received:
    4,521
    Trophy Points:
    113
    Location:
    London, UK
    Either way, have fun storming the castle- and thanks for taking one for the team. Not expecting you to be unbearably whelmed, mind.
     
  20. Wfojas

    Wfojas Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    848
    Trophy Points:
    93
    Location:
    CA
    I see that the wiki entry for MQA states that
    "Based on information from Auralic, a manufacturer of Audiophile Wireless Audio Streamers, Meridian Audio prohibits digital output of unpacked MQA in any digital format, only allowing the unpacked data to be fed to an on-board MQA-compatible DAC and output in analog form. Some claim this to be a part of DRM process[citation needed], which allows a proper MQA file to be authenticated and the full quality of the signal decoded only on commercially-licensed equipment."

    So not even sure what Audirvana has going on. In theory they can pay the same license fee for the software, but piracy can rear its ugly head. I forseee someone making an unlicensed plug in in Foobar for this.
     

Share This Page