Monday, 25 July 2016

A Response RE: MQA, "blocking" of Meridian's responses, and A/B Tests...

Just a quick note (remember... I'm on summer holiday and all!). Someone passed along to me this article from Hi-Fi+. It's an interview with Bob Stuart and Morten Lindberg (2L) apparently conducted in May. I'm bringing this up because the first Q&A is about "detractors" and there's this curious response:
BS: We’ve had some detractors, but frankly where we sit, we’ve been actually rather pleased how few. We’ve come in with an inconvenient truth – digital audio hasn’t been done right – so we thought the community might attack us. And one person put up a long blog about MQA that was just inaccurate, but blocked all our replies. As we say in England, “That’s not cricket!”
Now I don't know if this is definitely in reference to this blog and specifically to posts here and here but over the last few months I have heard this claimed... That somehow Meridian tried to post a response but the message was "blocked" (presumably by me) and that presumably the message would have contained corrections and answers to the issues raised.

The interesting thing is that I know Bob Stuart did post something on Blogger and I actually have a copy of what was written because Blogger sends me a copy of all messages posted at the time of submission. For some reason it did get erased but the interesting thing is, I did not erase it. In fact, over the years, I have never turned on comment moderation nor removed any genuine comment from this blog site (I will erase the annoying spam). Presumably it's Blogger/Google that removed it for whatever reason then, possibly due to automatic spam detection
.

In any event, back in April I responded to the question already:
Archimago17 April 2016 at 14:50
Hi Richard,
All I can say is that I did not block anything at any time. In fact, as you can see I don't moderate the messages at all. The only time I ever erase any messages is when I see a clear case of spamming or if the writer posted something then deleted it and I just want to clean up any "residue".
I do not know if Blogger erases messages on it's own. If so, then this must be an example.
Now as for the message itself... I have it opened up right now in another window and I can tell you that it was thoughtfully composed and did not appear rushed or incomplete. Instead of posting the whole note to respect whatever reason resulted in the deletion, I will just summarize a few points which I don't think would be contentious:
- There may have been different versions of encoded files released so noise floor could have varied.
- He's grateful for 2L for releasing the test bench. There's potentially "cross-checking" that needs to be done to make sure the versions released are from "consistent sources". (I of course agree and thank 2L for allowing us to compare the different resolutions.)
- Acknowledgement of working on questions posed - I assume he's referring to the eventual Q&A posted in Computer Audiophile.
- ** The main objection to the results: He feels the noise floor is equivalent to 22-bits below 10kHz and "always below the noise-floor of the recording".
- Regarding the PS Audio blog post: the development decoder was not updated to production version and "would have been operating in 'bypass'" mode so he felt Mr. McGowan and company were not experiencing the benefits of MQA.
That's pretty well the gist of it... I trust there's no "mischief" here and most of these points are covered and expanded on in the Q&A post on CA.
Unlike recent political events such as the "debates" among Republican candidates :-). No, I trust there's nothing like a bar brawl going on here. As I noted, Mr. Stuart's response was well composed and gentlemanly throughout - as I would of course expect from him and for these discussions basically about academic and empirical matters.

That's all I know about this. And as far as I can tell, the Computer Audiophile Q&A didn't really say much we didn't know already and still there are lingering questions about the nature of the claimed "losslessness", questions about the "de-blur" algorithm, and noise floor characteristics across the spectrum.

By the way, while you're on the Hi-Fi+ article, check out page 2 and the response on why A/B testing isn't used in their demos:
BS: We didn’t want to do public A-B tests, because they are completely uncontrolled. You know what it’s like – you put three audiophiles in a room and you get nine opinions. In fact, we do A-B tests all the time, with people who understand the context. So, for example, whenever we were with someone who had made the recording, we’d do the A-B test on the spot.
But ultimately, we didn’t want the discussion to devolve to A-B testing, when in fact it was about bringing the sound from the studio.
Wow, talk about a slap to the face of the audiophile public as people who must not "understand the context"! The most important thing A/B listening would have answered is whether it sounds any different to those in attendance. If it does then I imagine this would at least spread the word and promote curiosity for the average audiophile. Of course demo rooms would be suboptimal for ultimate critical listening but companies typically bring out their best gear with easily 5 if not 6-figure systems and can treat the rooms to some extent. If not under the guidance of a Meridian-approved set-up where perhaps time-coherrent playback is ensured to bring out MQA's performance, then where practically is an audiophile supposed to audition MQA? Surely "revolutionary" technology can be demonstrated to be obviously different, right? Are we to go on faith - "trust me, this works!" - and just buy a new MQA DAC and demand MQA-encoded music when the technical details appear unclear and there's not even an attempt at demonstrating benefits?

I would imagine that only if most people heard no difference would there be a devolution into doubts or debates around the A/B procedure used. Is that the concern?

------------

Anyway... Back to summer :-).

27 comments:

  1. I think enabling comment moderation might be a good idea. Not to filter out competing points of view, like what Michael Lavorgna does constantly on Audiostream, but stop the spam comments from being published.

    If you post a comment on your blog and check "Notify me", you get emails after every spammer posts some nonsensical link to cracked software or the other stuff they post, and comment moderation would stop this.

    Turning on "Notify me" for this thread. :)

    ReplyDelete
    Replies
    1. Count me as one of Michael's banned.

      Delete
    2. So Rattus...

      What horrific thing did you say/do to suffer such indignant fate? :-)

      Delete
  2. Blogger does have spam detection, however it puts any comments it flags as spam into a folder you can view:

    http://googlesystem.blogspot.com/2010/08/blogger-adds-spam-filter-for-comments.html

    So if you got a notification about a comment from the Meridian guys, and it is not showing up, either it is in your spam folder, or the author deleted the comment after posting it.

    ReplyDelete
    Replies
    1. Thanks Jim.

      Notification on and I'll endeavor to remove the spam messages more!

      I checked the spam folder... Nope. Other than the notification I received when the post was sent, it's not there! Bizarre. In any event, the post above summarized the contents.

      Delete
  3. Any audio format that contains DRM is a huge no no for me.

    ReplyDelete
    Replies
    1. I agree. Whether one calls it "DRM" or not, the loss of freedom in being able to fully decode the data must be commensurate with true gains in whatever the proprietary technique claims to provide.

      In this case, MQA claims to provide better sound with the data compression. If we can't be sure of the audio quality gained, why should the public give up the freedom to fully decode their downloaded hi-res files?

      From my perspective, the onus is on Meridian/MQA to demonstrate the superiority of the technique. How are they doing this by not even playing MQA vs. regular file in at least some kind of A/B comparison!?

      Delete
  4. They're being incredibly cagey about the whole thing, which suggests that it's smoke and snake oil. Heck, even Neil Young got people to listen to his stuff in a car...

    ReplyDelete
  5. Last fall a local high end audio shop invited me to attend a Meridian MQA demo/event, so I brought my audiophile friend and we attended. I expected to 1) hear about what MQA was, how it worked, etc. and 2) hear an A/B comparison of the same tracks with and without MQA.

    Instead, there was virtually no information from the Meridian folks on how MQA worked and no A/B comparison of any tracks! Furthermore, many of the tracks they played were 1950's and 1960's recordings. They did serve free wine and snacks, so got something out of it. And the KEF Blade speakers they demo'd the material on sounded great. But my friend and I were amazed at how the event had no informational value at all. Perhaps there's a reason behind that!

    ReplyDelete
    Replies
    1. Sounds like their room at CES. No A/B comparisons and despite using a $$$$$ Meridian system, there was nothing special about the sound whatsoever (It even sounded a bit bloated and muddy to my ears). I don't know if this had more to do with the Meridian gear, MQA or a combination of the two, but either way, they really should pay as much attention to system setup and SQ as any other vendor attending CES. I certainly left with a negative impression of both Meridian gear (which I've owned and enjoyed in the past) and MQA.

      Delete
  6. With the last couple of audio shows that I've attended, I was hoping to get a proper A/B comparison of music with and without MQA encoding, but no such luck. We got a explanation of the theory behind MQA and heard some encoded tracks that sounded great in and of themselves but that's it. I respect Bob Stuart and the over-arching idea of improving the recording thru-to mastering process and ensuring a music file's provenance. But without a little more transparency to the buying public, this doesn't do much to convince people that this whole thing isn't just a licensing/money grab scheme.

    ReplyDelete
  7. Hi Archimago,

    Great blog and very informative (even though I'd never claim to fully understand some of the discussions).

    However, I must say, that as far as MQA is concerned, there's a baffling number of people out there (not yourself of course) who seem to be absolutely convinced that DRM, in the traditional sense, is part and parcel of the deal, which simply isn't true. Yes, to fully obtain the benefits as claimed by Meridian you will need a MQA enabled DAC, but how this is any different to having to own a turntable to play vinyl records or an SACD player to play SACDs, is beyond me! In fact, it's much less of an issue, as MQA files will still play using a non MQA DAC. The point I'm making I suppose it's that MQA is dismissed without being given a chance in some quarters based a patently false premise.

    I do agree that the onus is on Meridian to convincingly prove MQA's superiority and I also agree that the jury is still out on that one. They need to try harder.

    MS

    ReplyDelete
    Replies
    1. Hi MS,
      Thanks for the note. Right, it's not the traditional DRM we've been exposed to over the years. Those systems prevented digital data to be copied - for example Verance/Cinavia watermarking preventing playback of copied data, or Sony/BMG's preventing CD ripping.

      In this situation, so far at least there does not appear to be any attempt at prevention of bit-perfect copying of the files. We don't know if there is embedded ID's in the file of course. The only loss of freedom is the ability to fully decode the data on "unauthenticated" DAC's and no software decoder available for computer decoding.

      Delete
  8. I simply don't understand why, if Stuart wants to engage with you on the topic of your concerns or questions on MQA, he doesn't do so simply and openly. What say you simply send him an email, identify yourself transparently, and have an email discussion that you premise with an offer to upload here?

    Sound engineer and educator Mark Waldrep also attempted to engage Stuart with some doubts on MQA, or at least requested some clarification and discussion. Mark thought he had a decent cordial pre-existing relationship with Stuart, yet Stuart has 'gone quiet' on him.

    Remember this: Stuart's entire audio career and business model is built on one presupposition, namely, that human hearing has near-infinite resolution that cannot be matched by audio products in the reasonably priced bracket, or using widely and cheaply available technologies. If a $1000 Pioneer CD player meets the ear's requirements, he's dead. If mp3-320 is indistinguishable from a master tape, he's dead.

    ReplyDelete
    Replies
    1. Hi tnargs,
      Over the years, I have E-mailed a number of well-known folks in the audiophile world and have had great discussions. I'm sure he can find me if he wants... But like you say, if someone like Mark W. can't even get his recordings with unquestionably hi-res provenance converted to MQA for testing, there's not much that can actually be "said".

      We are now >1.5years since MQA's introduction. Without even demonstrations with comparisons, there's only so much that can be said at this point unless some practical sonic differences can be shown.

      Delete
  9. I do not think that is a totally accurate characterization of Stuart's viewpoint. The Meridian Explorer 2 DAC belies that. And, I personally do not feel that the examples you cited demonstrate that digital audio is a fully solved problem, though some mid and mass market items are good enough for many people. I do think that digital audio is relatively mature, having overcome the major problems, the "low hanging fruit" if you will, even at remarkably low price levels. But, the rate of sonic advancement in digital audio has slowed considerably, as it does with most all technologies outside of audio as they mature. The problems that remain in audio are relatively small ones unlikely to be solved in any kind of major breakthrough. I think that applies to MQA.

    At the same time, recordings still do not sound indistinguishable from live music even on the priciest, most high tech systems. But, I am still open to the possibility that MQA and other key developments in the recording/reproduction chain might slowly nibble away at the gap vs. live sound one small bite at a time.

    It does seem that the MQA rollout has many flaws. It does not appear that they correctly anticipated the huge opposition from audiophiles, some of that valid, but much also spurious. Stuart has gotten hit from so many different directions, he appears to be unable to respond to all his critics.

    I agree that it would be nice if people could sit down to a simple demo of MQA on vs. MQA off. But, that might be a problem introducing considerable complexity in the design of DACs that could easily "switch hit" between use of MQA filters vs. an internal signal path that completely bypassed all of MQA's thinking in their filter design. Even if such a DAC existed, I expect audiophiles would remain highly skeptical and accuse Stuart and his adherents of tilting the DAC design so as to favor MQA sonics.

    Using, say, an MQA DAC vs. a comparably priced non-MQA DAC might be the best comparison we could hope for. But, intense skepticism would continue because obviously many things in the signal path would have changed beyond just the use of MQA filtering on the DAC side. Still, I for one, would welcome such a comparison, flawed in theory though it might be.

    Such a comparison should also include the ability to hear what difference MQA makes when applied to just the a-d and mastering side when played through a non-MQA DAC.

    These comparisons are possible now, so I wonder why there are not more of them. Perhaps, it is just an early issue because there are not a lot of MQA DACs available currently.



    ReplyDelete
    Replies
    1. Hello Fitzcaraldo,
      Good points.

      Yes, audio technology progress has slowed. It is IMO mature. But you are right, reproduced sounds still do not sound "live" yet. My sense is that this is primarily the fact that we are stuck in 2-channel stereo and there is no way that 2 speakers are going to be able to reproduce the spatial characteristics of a real performance... We could get a good facsimile especially with simple recordings (like slow-paced vocal jazz), but it's still not "real".

      I like multichannel because it does get us closer. When done right, it's truly fantastic! But it's expensive of course and for the most part impractical... Not to mention the lack of much material produced and mixed properly. Maybe in time this will improve but with space at a premium for most homes, I don't anticipate many folks will truly reap benefits (hence few companies will put resources into it).

      I think we can even wonder if audio has *retrogressed* with the rise of LP/vinyl. The meme out there that somehow the old analogue technology represents a step forward of sorts in better resolution. I can appreciate that some folks prefer the sound of an LP, but IMO this is subjective preference rather than an improvement in the state-of-the-art.

      As for MQA comparisons. As far as I'm concerned, MQA could easily release 2 files - standard CD 16/44 upsampled to 24/176 with typical linear sharp digital filter and the same master but MQA processed into 24/176. Let people listen. The MQA-processed 24/176 version will contain the majority of the MQA effect in terms of the claimed temporal accuracy and whatever benefits the "deblur" algorithm yields. Sure, they might claim that the real thing will sound better, but in practice, it'll only be a slight improvement at best IMO.

      Delete
    2. Thanks, Archimago. I am a devoted fan of your level headed approach.

      I agree totally about multichannel. I find it a substantial step forward over RBCD stereo for the classical music listening I mainly favor. I have thousands of Mch albums on my NAS. For the past 9 years, I have only seldom listened in stereo. Mch sounds much more like the real thing live to me, but there is still room for improvement and there probably always will be.

      My expectations are not too high about MQA sonics, but I respect Bob Stuart, his research and the huge effort he has undertaken. It should handle Mch as well as stereo, as has been promised. My main technical concern with it is the integration of PC-based DSP room EQ, which they say could be done, but I worry about the details.

      But, even bigger is the question of whether it will succeed in the marketplace with a good range suitable equipment and music choices. Effective demos will help consumers decide, but not many convincing demos seem to have occurred yet to make the case for it. Time will tell. However, it may be a case of much pain, little sonic gain, as you suggest.

      Delete
    3. Wow. That's a huge lot of multichannel music you got!

      Good show :-).

      You're right to be concerned about the integration of DSP room correction in an MQA system. Until something like this is announced, I just don't think it's possible or part of the "authenticated" MQA mindset. The whole idea of "end-to-end" authentication implies that they do not want anything to intrude on the signal. Of course this is all a false sense of security because the biggest variable are the speakers which MQA can do nothing to correct (and we would all be right to optimize with DSP processing).

      As much as Bob Stuart can be respected in the field and has done great work through the years. I just don't see MQA as a major innovation that will "up the game" meaningfully.

      Delete
  10. https://dubbeers.blogspot.com.eg/2015/09/a-players-guide-to-my-pbp-game-on.html?showComment=1471214640783#c7036164558935221844

    ReplyDelete
  11. Hello Archimago and co-readers of this Blog,

    Perhaps this article may be interesting: http://thehbproject.com/en/articles/37/Meridian-MQA-Explained.
    It's not hard to read and understand (for people with some technical background I assume) and gives some idea about what MQA is and tries to achieve. It does not mean I'm turning around in favour of it, but it makes some sense why MQA may have its purpose ... (If you like you can also watch this video for a condensed version of the article: https://www.youtube.com/watch?v=r_wxRGiBoJg&feature=youtu.be)

    Enjoy !!

    Regards, Menno.

    ReplyDelete
  12. It's easy to do your own A/B demo of MQA. I have an Explorer2 with MQA decoding. I've downloaded several MQA albums from 2L. In a couple of cases, particularly the Thorsten Mozart, I also have a PCM high-res version of the same album. In direct comparison, my view is that MQA sounds substantially clearer, less muddy, and is completely non-fatiguing, with much smaller file size.
    Of course, the Explorer 2 is a USB DAC, with all the compromises that entails. When I switch back to my Teac UD-501 (same one archimago uses!) PCM generally sounds at least equal to the MQA files on the Explorer. My initial conclusion is that there's definitely something there with MQA, but if you want it to sound its best, be ready to spend another small fortune.

    ReplyDelete
    Replies
    1. Thanks for the note Don.

      Over the last months I have received a few 24/192 rips from MQA DACs that are supposedly representative of the MQA sound. I cannot say I've heard any special clarity or lack of muddiness. In any event, if MQA could deliver such quality, I suggest they start running A/B demos ASAP!

      Delete
    2. That's a rather gentle reply, Archimago. Given the level of expertise I find after reading your blog and the responses in more detail, I can see that my post came across kind of simplistic. Since posting it I've put the Explorer2 in a drawer and gone back to my PCM files, everything from 16/44 up, played through the TEAC UD-501. There's some wonderful music and great sound to be had there, especially with DSD upsampling via Audirvana. I owe that tip to you. It can often be very very difficult to discern any improvement in high-res PCM over very well recorded 16/44. I don't have anything that sounds better than my rip of Cohen's Ten New Songs CD, whether it's played as straight PCM or via DSD upsampling. Your posts have got me downloading some 16/44 files again, especially from Hyperion. Great music, cheaper, and is there really a difference?
      Don

      Delete
  13. I just posted in the April section. In 2 months, none of the files downloaded activate the MQA light on two different Explorer2 units. I think Windows computers based on my three units often cannot play a file bit perfect.

    ReplyDelete
    Replies
    1. see http://www.meridianunplugged.com/ubbthreads/ubbthreads.php?ubb=showflat&Number=186420 for playing bit perfect via Windows

      Delete