Bending with the crowd

March 12, 2017
 by Paul McGowan

Sometimes we set the pace, other times we try and keep up.

Take MQA for example. As my readers know I’ve not been as big an advocate as others have. I have heard impressive demonstrations, but none that would make me run home and stop other engineering projects in its favor. Yet, it’s important to enough people that it finally made it onto the engineering schedule.

We’ll be launching both MQA and Tidal in a new, free, release of the Bridge II firmware in April. Owners of PWD, DirectStream, DirectStream Juniors (with Bridge II installed) will now be able to stream Tidal and decode MQA.

There will also be a new update for both Ted Smith designed DACs—DirectStream and DirectStream Junior. Ted’s done some rather wonderful work on improving sound quality, yet again, and this is very exciting. A new mountaintop to name!

What truly excites me is the fact that with both products, our DACs and Network Bridge, we can transform both their functions as well as sound quality to thousands of owners through a free download.

Perhaps bending to the crowd is ok sometimes.

Subscribe to Paul's Posts

49 comments on “Bending with the crowd”

  1. Hi Paul,
    Your initial comments on MQA last year highlighted your concern that implementing MQA in the Directstream would require compromising performance (tweaking filters, etc). Have you found a way to avoid that, or is this “bending” you describe a performance tradeoff for giving the masses what they want?

    1. JM – that’s been addressed in other threads – the implementation is not in hardware, which causes the compromises Paul discussed. This is a “partial unfolding” sort of deal in software, as I understand it, so not a full, end-to-end MQA hardware implementation. So the benefit of the ability to stream it in Tidal to the Bridge II, without the sonic drawbacks Paul found — though not a “full” implementation.

      Makes you wonder if other DAC manufacturers accepted sonic compromises in order to jump on the bandwagon – or if it simply doesn’t affect other sorts of DACs, or they can’t resolve the differences.

    2. As Badbeef suggests, no hardware changes have occurred and the DAC is untouched. This is a pure software change in the bridge II and does not alter the DAC and its performance.

      What we have determined is that MQA altering DirectStream itself is not acceptable. It’s worse, something we will not allow. This solution we just announced is a good compromise, allowing DirectStream to remain pure while unfolding MQA to its first level.

      MQA’s second level upsamples the unfolded data and reconfigures the DAC—neither of which are processes we will permit because the results (while different) aren’t better. Quite the opposite, in fact.

      1. Listening now to selected Tidal “Masters” as a HIFI subscriber into the DirectStream via BridgeII via a MacMini /Roon Server, the streamed files even “unfolded” impresses and tickles my ears with the current resolving capability of the DirectStream.
        These files not being “unfolded doesn’t” seem to harm the resolution of what I’m hearing and seems to be an improvement over the regular HIFI or ripped PCM Rebook AIFF files out of iTunes. Curious as to why that is? The DirectStream seams to really like these files unfolded… so you hear the same?

        Looking forward to what other “magic” might come from an unfolded level 1 presentation through the BridgeII.

  2. BTW, a PSA PSA: Change your clocks, (at least) folks in the US!

    Reminds me of when DST was first rolled out in Michigan where I grew up – someone wrote in to the paper (remember that?) complaining that their lawn was going to grow an hour more a day, resulting in more mowing for them to do.

    Or when the King changed to the Gregorian calendar, and the peasants revolted because he had stolen a week or so from their lives! Forgive me if my history is shaky or apochryphal – I may have gotten that last one from Sherman and Dr. Peabody ; )

    1. I think it’s Mr Peabody, and let’s hope that MQA turns out to be more than a “fractured fairy tale.” Otherwise,
      our own “Dudley Doright” will have to come to the rescue.

      1. Right! “Mr.” – thanks. Wondered why that looked wrong.

        Well at least the MQA team has relented to the extent that they are allowing software implementations. I’m guessing they were feeling enough pushback on the need for full-on hardware implementation that they realized world domination would be more readily achieved by compromising to some extent.

        The “Audio Deep State” marches on! ; ) ROFL, etc.

        1. Paul, being a Tidal/HF subscriber and a DSD owner without bridge II at present, I have a question. The Tidal desktop app (my only source for MQA files) only does a partial software unfold on MQA. Would the upcoming PSA firmware for the Bridge II take this any further, and how close to the hardware unfold are we to expect?
          I ask this as I have two options as I see it.. Either a Lanrover purchase to get a clean USB connection to the Tidal desktop app, or a BridgeII purchase for the PSA unfold. Which is the right path for me and others?
          Sorry, but you opened the door.
          Thanks in advance,

          1. No, there’s only one stage of unfolding available in any of the MQA apps and they are all the same. No advantage to having two. Once unfolded, there’s nothing more to be done unless you have an MQA enabled DAC with the second stage.

            That basically takes the unfolded music, upsamples it (as to my understanding) and modifies it to fit the personality of the DAC. We’d not permit that in our hardware.

            1. Thanks Paul. Noone can blame you for not wanting to wreck a wonderful device like the DSD. I look forward to the new firmware for it, and I guess I’ll give the Lanrover a try. I need a decent link between the computer and the DSD. So, my choice is obvious. Thanks again for a great product and your honesty as a business guy. You really make this all a lot of worthwhile fun.

  3. Great news… well received here! Thanks to Paul and the PS Audio team for keeping its principles and priorities aligned. We always have a choice when it comes to where we invest our Audio $$$ for what we get in return.

    You guys do a great job in striking a balance between leading innovation and following with the proper implementations of new industry features and functionality as well.

    I’m looking forward to even more and improved musical experiences from any advancements that can be made to reveal even more to our listening ears!!!

  4. What are you going to do when you run out of mountains Paul.

    “Bending with the crowd”

    I call it pandering. But you have to give the market what it wants or you go out of business. What the suckers…er…..I mean customers don’t realize is they pay for it whether they use it or not.

    I’m sure this one blows the last release away which blew away the one before that which blew away the one before that….

    Too bad I can’t get out there. I’d like to hear a playoff between the latest release and my $30 Toshiba. 🙂 EMOJI ALERT!

    1. Well if each version didn’t improve (or in Paul parlance, “blow it away”) it wouldn’t make sense to spend all the time and money we do to make the improvement.

      I know your cynical streak is on fire when you read all this – but consider we give this for free. It’s not like I get excited with “blow away” performance and then put my hand out for more money.

      We give huge improvements to our owners, free services like Tidal and MQA, and all for just the original price for admission.

      Surely there’s room in your heart to appreciate the importance of this to our customers.

      And yes, I would dearly love the opportunity to embarrass you and your Toshiba in Music Room One. My black heart would be in heaven! 🙂

      1. It’s not my cynical streak…it’s my funny bone. Got a nice challenger for your power plant to play with for the next few weeks. A 150 kva Mitsubishi ups. It was company surplus. Ups was free. Installation will cost over $100 k.

          1. For a home hi fi system yes, even Dick Burwen’s 20 kw sound system. For a sports arena that holds 100,000 people, no. For a data center, it’s a peanut. Once they get to about 500 KVA or much above 750 KVA we start using them in many types of networks where they not only provide enormous quantities of power and are backed up by banks of huge generators, they provide redundancy that makes any one component failure transparent to the end user. If you think I argue about such trivial things as hi fi equipment, you should have heard the wars I fought over vulnerable networks that could collapse like a house of cards. The reasons are sometimes very arcane. Line finding a needle in a haystack. Imagine the impact failure has on an investment bank or a large hospital complex. It better be right the first time. No “patches” to fix inherent flaws allowed.

        1. In the shootout between the $6000 Direct Stream and the $30 Toshiba, I get to pick the recording and the sound system (It will be respectable.) Your equipment is optimized for your equipment and equipment you like. Everything is pretty much fixed. That alone puts you at a severe disadvantage. It’s like a sports car that can do 0 to 60 on a dry level road in 3 seconds but can’t handle a left turn in the rain at 60 mph without going out of control. Good recordings are defined by audiophiles as sounding good to them on equipment they like. Other perfectly fine recordings that don’t are considered poor recordings even though they are among the best recordings and best performances of the most important music ever written.

  5. Paul,

    Re: Tidal MQA support, will I be able to stream Tidal Masters via Roon to the Bridge II and have them unfolded to 1st level? In that case, I suppose the SQ will be similar to having the DSD connected via USB to a PC and streaming MQA Masters via the Tidal client. Am I thinking about this the right way?

        1. The Tidal app can’t but the MConnect app can. Once we update Bridge II next month, all you need to do is use the MConnect app to play anything you want on Tidal directly to the Bridge II. If it happens to be MQA encoded, the Bridge II will unfold it to its highest level, 88kHz or 96kHz.

          Anything higher than that with MQA is upsampled in other DACs.

  6. Having just installed the Bridge II & loving it, I was wondering if we would have to use third party software such as Bubblepop for the decoding of MQA or will the bridge II now take care of that ?

  7. You are saying that MQA decoding will be available through the Bridge 2, can it be built into the Directstream DAC? I don’t do streaming, but I would like to be able to download files and play them back on the DS DAC itself. I am almost entirely using physical media.

    I look forward to the updated FW for my DS DAC. The FW updates have all been positive in my system.


    1. You don’t have to stream them, but you do have to play through the Bridge. One way or another, you have to get them into the Bridge. I don’t believe there is any physical media with MQA decoding, or, am I missing something?

    2. If you don’t have a B II but your PC is connected to the DSD via USB, you can already play MQA via the Tidal app. It does the same L1 unfolding that the B II will eventually do. So I think you have choices depending on how you want to connect to the DSD.

      1. H there, I’ve got a Tidal hifi account and about to install a BII into my Directstream. I control everything via tablet (Samsung, hoping to get it working with IPad) but my understanding is Tidal Masters (MQA) is only available on MAC Tidal App. Presumeably the integration of Tidal into next release of BII firmware doesn’t getvaround the requirement to use a MAC for Tidal Masters? Thanks, Mike.

    1. LOL, Soundmind. That’s twice this week you’ve displayed a sense of humor that tickled my mind. Must be we have a funny-gene in common. Actually, I’ve been toying with the idea of pulling my cassette deck out of mothballs lately. Hopefully the rollers haven’t developed a flat spot over the years.

  8. Umm. Any timeframe for the new mountain top?

    And Paul. I think your following recognizes what your company offers and the reason I bought a DS was for THAT. Never heard it – but the company is so engaged – it is rare. I think everybody recognizes it – and loves you for it!! 🙂

    1. Thanks, Tim. We’re shooting for a release next month, April. I can’t promise that because, well, it’s software – the big unknown, but Ted’s pretty reliable about such things so there’s a good chance.


    In the issue of time resolution he got it dead wrong. The 5 microseconds has to do with the changing difference of the time of arrival between your two ears, not the highest frequency or “resolution” of what you hear. This plays a role in perceiving direction of sound, not its tonal quality. Anything based on this error is just plain wrong and worthless. This is why the eardrum is directly adjacent to the cochlea which tells your brain the orientation of our head. The two are compared when you move your head and the change is immediately sensed as the direction the sound is coming from.

    96 db dynamic range is sufficient for all music. With background noise in a dead quiet concert hall the background noise is 25 db. 25 db noise + 96 db signal is 121 db above the threshold of pain. Do that enough and those hairs in your cochlea will be destroyed for the rest of your life. However loud you play pop music, the dynamic range is rarely more than 10 db.

    In part two, the crux of the technology is revealed. It’s something the inventor calls “audio origami.” Origami is the art of Japanese paper folding. Here the higher frequencies are folded digitally into lower frequencies in the digital domain twice and sensed at a lower signal level being segregated from the signal in the main audio passband.. This is the essence of the idea.

    The speaker’s listening experience and review were of course a joke. There were so many variables and so much confusion it was meaningless and it was only one demonstration. The only valid test is of course the same signal downconverted from Hi Rez to RBCD repeated with significant correlation many times. The so called “time smear” on the piano was attributed to the room reverberation and disappeared when the sound level was lowered.

    The simplified but accurate explanation for why you only need 20 khz even if your hearing is excellent is that your ear cannot respond to a change in loudness faster than in 1/40,000 seconds. That’s the time of a half wave at 20 khz. Therefore wider bandwidth systems are of no practical use for sound quality. However, oversampling by 8X (320 khz) 20 bit technology was used in my 1987 Denon 1520 that got the steely sound of strings resulting from the low pass filter so far beyond the audible range that it was for all practical intense and purposes gone. The presenter’s explanation of this would make you think it was discovered yesterday. Nothing new there.

    Putting things into perspective, the trashy music most people including audiophiles listen to whether antique which has no signal above 20 khz or the trashy music being recorded today with the most sophisticated equipment doesn’t merit anything beyond MP3. That’s my observation and it doesn’t matter who recorded it or what it is. In fact even at the MP3 level it doesn’t merit paying money for or for that matter listening to for free. Just my opinion. I think I’ll listen to a nice recording now….just to see if the equipment still works as my interest in music and audio equipment is still in the deep freeze. I’m thinking about that Brahms double with Heifetz and Piatigorsky on RCA.

  10. Paul,

    I’ve spoken and skyped with some other builder of DACs since you published this post.

    The purport of each answer about MQA was that they do not like it.
    Further they see it as a kind of DRM and that it forces the customers to rebuy the music.
    Finally most said that they think they will have no choice but also offer MQA in future releases of their products only to stay competitive. That’s what I call market power – to praise MQA to the skies like most of the opinion leaders in HiFi did.

    So bending in my POV is not OK but necessary because of a further misdevelopment in the music and HiFi area.
    It’s too bad how much influence the gluttons in audio business have.


    1. I don’t know one DAC manufacturer that supports it. Some, like Mytek, have capitulated – maybe he even likes it – and I suppose there are others. But, every DAC designer I know really dislike the sound of it, the idea of it, and certainly the notion of “tailoring the DAC’s performance”.

      Then, there’s the DRM aspect of it.

      Aside from the compression benefits to streaming companies, there’s little to like. Why some magazines have gone so gaga over it is beyond me, but then perhaps they’ve heard things I have not. I agreed to add it to the Bridge for a couple of reasons:

      1. Our customers asked us for it
      2. We are only unfolding it – not unlike what we do for FLAC or ALAC
      3. We are not giving in to changing our DAC’s performance

      So, in the end, it’s really no different at this level than unfolding any compressed format, like FLAC, ALAC, OGG, MP3. We simply added a software CODEC so customers could hear the higher rez unfolded version. That’s not only the right thing to do, there’s no downside and plenty of upside.

      No values have been compromised.

  11. One more thought on bending with the crowd. Usually some stupid guy, usually not too clever gets a bee in his bonnet. He’s discovered a new silver bullet. Take for example “time alignment speakers.” He advertises the hell out of it using some smart ad agency and some slick papers and suddenly the world follows. If you don’t have time alignment, what you do have is no good. It could be a lot better. Words get mixed up, tyros are agog at the insight, can’t wait to hear it, try it, buy it. Problem is, it’s pure bullshit. The reason is that in acoustics which is has three dimensions, time alignment is only one criteria for phase coherent. The other criteria is same geometric center of propagation. All of the time aligned speakers missed it by a country mile. Bottom line it’s of no value and would be of questionable value even if it meth both criteria. Doesn’t matter. Soon it will be forgotten when another silver bullet comes along and the crowd bends to that false idea instead. Example, high resolution. Problem, if a real breakthrough ever comes along, how would anyone know? All the superlative words have been used up on previous silver bullets. Just look at ads for radio phono consoles from the 1930s and 1940s. Same words. You have to give the admen credit. They always find a new way to tell the same lies.

    1. I’d suggest you stop reading the “BS” – it seems to stress you so much, the cortisol generated is going to kill you! And all on subjects you claim not to care about anymore, and a hobby you are not involved in.

      But it seems you are more than delighted by pointing out various Emperors’ lack of clothing. So, knock yourself out! If you could do it more often without the tone of “I’m smarter than all of you easily duped audiophools put together”, it would likely go a long way toward people accepting your viewpoint.

  12. Thank you Paul for the step forward being planned for unfolding MQA. I’m very much converted to Tidal streaming via Roon to Bridge 2 and Directsream and like others I find the unfolded MQA for most albums sound better than my old Cd rips but probably only in a similar way a remastered CD sounds better than the original CD release. A couple of questions for me:

    1. Roon themselves are planning on doing the unfolding of MQA in their software so presumably i just need to wait and see who gets there first. Could there be any differences or advantages on using Roon or Bridge 2 to do the unfolding?

    2. Roon do have a ‘DSP’ engine for upsampling options / DSD playback etc but is a feature that seems disabled for Directstream users. Noting your principle to not do further MQA upsampling within the DAC itself would there be any potential advantage to allowing the user so select such options in Roon? Could this then allow users to try out further upsampling beyond the MQA unfolded version?

    1. To the best of my knowledge, the unfolding planned by Roon is identical to what we are doing so likely there’d be little advantage to one vs. the other.

      Upsampling probably doesn’t buy you much. DirectStream already upsamples everything you put into it and it probably does it better than anything else I have yet encountered.

      1. Thanks Paul – that’s good to know.

        So in effect after the first unfold of MQA, Directstream remains in control of its upsampling and is not being forced to do a second level MQA process like MQA enabled DACs do.

        If the MQA Master file is greater than the first unfold limit of 96khz then presumably we would still get any sonic benefit through Directstream’s own processing?

        That’s good enough for me and look forward to trying it out soon.

Leave a Reply

© 2023 PS Audio, Inc.

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram