The live myth

April 15, 2016
 by Paul McGowan

Just a reminder that if you're attending Chicago's Axpona, come by and say hi tomorrow, at 4PM. I'll be one of the cranky old guys up on stage in the Legends Forum.

There are no perfect recording or playback mediums. The much sought after goal of reproducing live music from recordings in your home is mostly a myth, though we often catch glimpses - snippets of the thrill of what it might sound like - sometimes outright fooling us into turning our heads because it sounds like someone is playing in the room. But consistently fooled? Not in my experience.

And yet claims of one playback medium excelling over another swirl like Minnesota black flies in June.

I promised in yesterday's post on mastering limitations that I'd mention some areas where digital doesn't hold a candle to vinyl in a technical sense. Perhaps the most obvious is the fact vinyl's analog. And analog is the gold standard. Analog is continuous, infinite in resolution, and defines the medium conveying recorded music: the output of a microphone. Digital's performance is always referenced to analog's gold standard.  Sony's original marketing claim of Perfect Sound Forever suggested digital is a carbon copy of analog (which, of course, isn't true).

Digital's proclamation of perfection is kind of like the artificial sweetener industry's claim that their product is "indistinguishable" from their gold standard - sugar - a claim we can argue about all day long - but the point is the same. Both are attempting to be as good as the reference. Never better. And if someone's claiming "better", run like hell.

I don't want to focus too much on the good and bad of digital, though most of you know my stance on it. DSD is closer to analog than PCM - and not by just a little. Sure, there are pundits that hear more detail and resolution from PCM than DSD, even if the PCM is a copy of the a DSD recording. That's a subject we'll likely jump into when we take a breath. But for the purpose of this discussion between the two major format groups - vinyl vs. digital - I will simply reiterate that both from a technical standpoint and from my own listening experience, DSD is closer to analog than PCM of any resolution.

But regardless of your opinion on the matter, here's the thing. Both vinyl, DSD, and PCM's goals are the same. The accurate reproduction of the original analog waveform in all respects.

Vinyl attempts to capture the analog using analog means - while PCM strives to do the same thing with numbers, DSD with varying degrees of energy density. In each case, the goal is the same.

And none get it right.

Subscribe to Paul's Posts

28 comments on “The live myth”

  1. I think this is a badly misinformed posting based on intuitive understanding in a very narrow segment of technology that hasn't benefited from enormous progress in the last 50 years to nearly the same degree most other technologies have. The reason is that for those with the most skills in technology, this problem is no longer very interesting to them, they have better things to do with their talents. A lot of research is done where I work and in many other laboratories around the world. I think if you were to ask any of the scientists and engineers in any of them if they'd trade their careers, their projects, their goals for developing hi fi audio technology instead they'd just laugh.

    "And none get it right."

    What do you mean by right? Digital technology PCM can define an arbitrarily small degree of error and capture, store, transmit, and reconstruct an analog signal within that degree of error. Analog systems cannot. We experience the world at our macro level as analog. The reality is entirely different. Matter and energy are made up of individual discrete units. Most of the universe including the densest things we know are mostly empty space even though they seem solid to us. Are time and space also quantified? We don't know yet.

    Identifying, analyzing, and duplicating things are best done in using digital technology. Whether it's a simple forensic lab identifying a common substance, analyzing complex molecules, or synthesizing an exact duplicate of a strand of DNA from its constituents this is done under digital control. All industrial control systems are being converted to digital systems. The legacy analog systems are now inadequate and obsolete.

    As a hobby to entertain those who tinker with electronics or enjoy using them to listen to recordings, the debate over what entertains them more will go on, possibly for a long time. But for the scientific study of acoustics where the goal is to understand and duplicate sounds with accuracy and precision, the future looks to be all digital, Who will be more clever at applying it, those who develop PCM or those who develop DSD remains to be seen but I think so far PCM is winning out. It seems a lot of hardware and software have been developed around this method of applying technology. I don't know if PWM which is not a true digital technology has any real long term future.

    1. I mean, by the statement "none get it right" exactly what it implies. There is no perfection, reproduced music rarely sounds live - even your system which gets some music presentations better than most - doesn't sound live.

      Part of the problem's in the reproduction medium. Other parts of the puzzle you already know.

      We can jump up and down all day long about digital's perfection and yet trained listeners can routinely identify a digital presentation vs. an analog of the same music.

      Just like my example on sweeteners. Many people can't tell the difference between Splenda and sugar - they are, after all, the same molecular structure - just a "smidge" different. But then, many people (including me) can identify the difference easily.

      Just because some people can hear differences and others can't, doesn't get explained in a flurry of technical proof to the contrary - in either direction.

      One of our biggest ongoing debates is my empirical "proof" vs. other's "measured" proof. I think both are valid - or can be valid - we just need to keep sifting through the data on both sides to come to some better understanding - and that understanding usually leads to discovery.

      1. No matter what you are listening to, it is an artifact, key concept is "ART"….not the thing itself. Every human involved….from the designers of the microphone to the manufacturer of the speaker's driver….contributes his/her ARTISTRY to what you hear, and in a concert hall or jazz club what you hear depends on where you are sitting or standing….(Special Relativity). There is absolutely no absolute sound being regenerated. It can't be done.

      2. I most certainly have no problem getting the "live" vibe happening - although the "no problem" bit has to be qualified by saying that I need to go to a far bit of trouble optimising the system at the particular time for it to happen - I don't get it by just switching everything on in the morning, and pushing the Play button ...

        Depending upon my mood, I might be happy enough with a typical lower, "non-real" standard - but if I decide that I want the full deal to happen then I'll put the time and effort into revving up the system to that point - and enjoy the benefits!!

        Of course, it _would_ be nice if it was always on tap, a simple flick of the switch to get there - but current equipment setups nearly always have too many flaws along for the ride, it's hard work getting rid of all of them - but that's what's necessary for live sounding playback to just fall out of the rig ...

      3. "I mean, by the statement “none get it right” exactly what it implies. There is no perfection, reproduced music rarely sounds live "

        In the context that "right" is used here and "sounds live" is used these are comparing apples and oranges. Insofar as the medium is concerned, right depends on what your goal is. For audiophiles, it appears that often right means pleasing, possibly mitigating common problems elsewhere in a home sound system like shrill treble that are objectinable. To this engineer, in this context right means that the difference between the recorded electrical signal after the recording engineer has fiddled with it and the played back electrical signal is within the tolerance of audible detection, that is the difference is below the threshold of audibility. What does that mean? Not necessarily a lot in the way audiophiles use it. For one thing hearing abilities vary. For another we have no way to know how the recorded signal sounded so we have nothing to compare the playback signal to. Therefore in this context, "right" cannot be determined unless you have access to the original signal the way tape monitor switches were used to compare the signal going on to a tape with the signal coming off of it a fraction of a second later. However, that you could make recordings from records and possibly other sources like CDs or streaming audio that sound indistinguishable or nearly indistinguishable from the source probably means that playback device is just about right.

        The comparison of played back signals heard through sound systems and live music is a whole 'nother kettle of fish. In commercial recordings the microphones are placed close to the sources of sound so they don't hear what the listener in the audience hears. Directional qualities of sound components that fall on microphones are lost when they are converted to electrical signals. That quality is not present again until the signal drives a loudspeaker and its directional qualities will invariably be very different from either the source (except in the case of Acuvox's design) or the sound heard in the audience.

        As we do not have access to what was heard live there is no way to compare the reproduction to know if it is "right." At best it's only a guess based on memory of similar musical experiences comparing those to what might have been heard at a live performance when the recording was made. A guess, hopefully an educated one based on a lot of experience with live.

        "even your system which gets some music presentations better than most – doesn’t sound live."

        The sound system I patented was a rudimentary version of a far more complex idea. Only two prototypes were ever built, one in 1976 which was operated in three locations until it was disassembled in 1985 never to exist again, and a second one based on a device that infringed on the patent that has evolved since 2002. The purpose of these prototypes was proof of concept. They exist to demonstrate that there are qualities of sound that are plainly audible to ordinary listeners with normal hearing that are also present to one degree or another in live music that are not present from sound produced by other types of sound systems. The term "accurate" again has no meaning there. I prefer the term convincing which only means reminiscent of what might be heard live to those who have experienced acoustic music at live concert venues. There are a wide range of variables that can be applied to alter these qualities in this prototype and no one adjustment is "right" for any particular recording.

        To be "right" or "accurate" in the scientific sense of the meaning would require a far more complex version of the same idea on a huge scale, anechoic recordings, and installation of the equipment in an anechoic laboratory. It would also require field measurements of spaces to be recreated. At this level, this concept and the concept behind wave field synthesis appear to converge into the same invention. That technology has never been realized to anything like the required degree for "right" either and for the same reasons. Rudimentary examples of WFS make different compromises and yield different results from my prototypes.

        In discussing WFS with professionals experimenting with it, one of their problems is figuring out how to measure what they need to reconstruct. I had the same problem and my solution so far works only on paper, it has never been built or tested in the way I envisioned. The closest approach was surprisingly the microphone array used to produce the audio portion of the video visualization of concert hall acoustics I've linked to several times in the past. This suggests that although that device was not used in quite the way I would need it to be used, its viability as a useful tool or a more advanced version of it just might possibly work.

        As an aside, "Many people can’t tell the difference between Splenda and sugar – they are, after all, the same molecular structure – just a “smidge” different. "

        They are not a "smidge different." In splenda, three hydrogen atoms in sucrose have been substituted with three chlorine atoms. Hydrogen and Chlorine are on opposite sides of the periodic table. While both tastes sweet, chemically there is a world of difference among which, Splenda is metabolically inert. By your logic there would be only a smidge of difference between sodium chloride, common table salt, and sodium fluoride (fluorine is also a halogen like chlorine) a deadly poison in all but the most dilute concentrations.

        1. I never thought that I would recommend a loudspeaker from B&O other than by its design.

          Yesterday I had the chance to listen to a demo of the new B&O BeoLab 90 loudspeaker at our local B&O representative. This was in terms of soundfield reproduction the far best I heard in years if not even ever. It was really convincing although i do not really trust the B&O playback sources like CD player and their amps. Nevertheless I have to admit this was really fine.

          Soundmind have you ever read US 4332979? This could be interesting to you because imho it shows a lot of things that, how I understand it, your system is designed to.

          1. bernd, hahahahahahaha. Did I ever read patent 4,332,979? Of course I did. I WROTE IT! That is my patent. Even that is a simplified version of the entire explanation that came to me like a bolt from the blue 42 years ago. If you have any questions about it, I'd be happy to answer them.

    2. My simple question to all these vinyl-attracted audiophiles is: Where do you want your favourite song or piece of music be engraved in the vinyl? In the outer grooves or in the inner grooves or doesn't it matter at all?
      The astonished looking countenance of most tells me that they do not even know the basics of analog reproduction. And then there are those who say with conviction 'it does not matter to me because I use a tangential tonearm'. I think they know even less.

    3. Art is informed by experience and experience is a great teacher, fallible, but great. The junior partners in my practice know many things about their field but I watch them struggle with concepts that we all know are "right" yet they are not congruent with how the world really works and only time will teach them how to respond appropriately. The musical instruments that we listen to were all created by tinkerers. Would the reeded woodwinds be discarded if designed by principles of science and engineering? Too many odd harmonics. Clearly inferior to other instruments on the bench. Without art, we might all be listening to studio monitors. I have several pair and they are tools, not particularly pleasant to listen to but useful as a reference. I'll take the "tinkerers" products any day.

  2. I confirm, the ambition to ever reproduce "live" by playback mediums will have no end - like looking into the valley from a swiss mountain compared to watching the same in an IMAX cinema.

    With the goal to achieve this, in my perception digital often suceeds a little more to the head, while analog succeeds little more to heart and stomach 😉

      1. I found this remark very interesting and, in my case, true. I guess that is why I'm pursuing DirectStream in an attempt to have my entire collection (couple of thousand CDs digitized and maybe a thousand downloads too) sound as "analog like" as the current state of DSD permits.

        I've got a small dedicated room, and there's only digital there. That feeling of naturalness is something I always hoped to be able to duplicate so that one can feel it in the gut. F.i. in my room you have to be barefoot (no footwear allowed), sometimes more standing (as in a rock concert) than actually sitting, no tight wear, no belt on if possible -unfortunately with digital, PCM etc I can tell the artificial nature to music and so achieving that reaction of the gut not the brain is something I still aspire to.

    1. You didn't necessarily miss much -- unless you own the Meridian speakers and the recordings they were playing, it was pretty hard to tell how "good" it was, compared to any other "good" sounding stuff. Still scratching my head as to how the preamps and mics and consoles can be in any way "encoded" so to speak, in the master, to be retrieved by any process.

      I was interested that Robert Harley is now listening to it on his own system, so at least someone is talking from their own comparative experience. Though how it can be as magical as he claims is beyond me. Heading over to Axpona now -

        1. Sorry - did not find them today, as I was surprised to find that neither Meridian nor MQA, Ltd. were listed in the show guide. I just got home and with a bit of Googling found this:

          - which spells out where to find 'MQA partners' at the show. So I'll make an effort to check out the Bluesound room tomorrow.

          Did end up having some interesting conversations with manufacturers who should ostensibly be jumping on the bandwagon, but aren't. For some, it represents a not-insignificant investment in the ongoing future of the technology, so I got the sense they are waiting to see if the world agrees with Bob Harley or not, so to speak.

          If it turns out to in fact BE the best thing since sliced bread, everyone will HAVE to integrate the technology into their product lines, so no worries. If not, they saved themselves a lot of hassle and expense. It feels a lot less like a foregone conclusion than it did a year ago.

          My brother's here from out of town with his Meridian Explorer 2 (which according to Harley, Has the Magic despite the low price of entry), the PSB M4U2 (one of the sets of cans Harley used - though sans the $5k headphone amp) and 2L MQA-encoded files. I'm going to hook 'em up to my ATC SCM25 studio monitors (and maybe also Quad 988's), and I'll report on our observations, (assuming we don't have too much wine. No guarantees on that one).

          1. Thanks for the info. It seems with MQA less is more? I think with audio, the source is the most important factor and I have been playing with audio for fifty years!

            1. Yeah - based on our highly unscientific tests last night, I'm thinking that, YES, it's a good codec. Perhaps better and more efficient than most. Do we really NEED another one? I dunno.

              Would it be nice if we could get RID of the plethora of codecs we currently have and replace them all with a smaller, superior one? Yes. (One Ring to Bind them All...). Do I think that's going to happen? No. I'd be happy to be wrong about that.

              Reminds me of when the "original" HD format, 1080i, came along. I thought, fantastic! We can just have one, superior video format! Instead it was the beginning of an explosion of formats. And now that we're in the midst of adopting 4K, they're pushing right into 8k.

              And back to audio - yesterday I met and spoke with Michael Bishop, one of the most lauded and respected recording engineers on the planet. An unexpected honor! He is now recording in (get ready for it...) ELEVEN POINT TWO MEGAHERTZ DSD. He's giving a talk today at 11:30, gonna play some of his work.

              1. The Bluesound room had great sound - the same group owns the three companies the system was comprised of - Bluesound streamer, NAD Master Series electronics, and PSB tower speakers.

                Sadly they claimed they didn't have any non-MQA versions of the music they were playing for comparison ("You wouldn't believe how many times I've heard that question" the rep said. I replied, "Yes I would").

                No Meridian devices involved. Had a nice chat with the MQA rep, who (sorta) explained some of the things MQA is doing.

                My sense is that IF it is adopted widely, we'll have (yet another) viable option. I'm not sure it will make all the existing formats go away. Their focus, it seems, is primarily on streaming, though no particular reason it wouldn't be downloads as well.

  3. Paul, your suggestion that vinyl is inherently superior because it is analog, as is the original microphone feed, is fundamentally flawed. Vinyl is the Nth analog copy of the analog original. If more information gets lost during an analog copying / processing step, digital would be inherently superior (not saying it is, but just pointing out the flaw in your argument). Think about video. Source images are analog, yet digital video is far superior than analog for precisely this reason - far more information gets lost in an analog processing / copying chain.

  4. "[The] goals are the same. The accurate reproduction of the original analog waveform in all respects."

    Once again, I'm going to be the picky bastard. My goal isn't waveform reproduction. It's getting in touch with the creativity and skill of the artists. On the liner notes of some LP that I now can't find, the annotator quoted Bruno Walter as saying that at one point in his life he knew he had gained access to Mozart's soul. He was speaking in terms of scores and conducting, of course, but after I read this some 50 years ago. It quickly became one of my hi-fi mantras.

    The motion of molecules in my listening room is a means to the end of my synapses firing in sync with Mozart's and Walters's. My consciousness (heart, soul, and gut if you will) is the final component in the chain. Ancillary perceptions are important, too. When listening to "Taxi War Dance" from a 78 rpm shellac platter, the knowledge of a very short and simple line between Lester Young and myself elevates the experience immeasurably.

  5. Perfection does not exist much as some may delude themselves. That leaves us with less than perfect. If there are more than one less than perfect in a field than the least inferior is naturally the best. Not perfect but best. Enjoy your trip. I hope it is very fruitful. Regards.

  6. Following a waveform accurately does not music make. Even compared to a single microphone, human hearing can detect frequency to greater accuracy in a shorter period of time than a theoretically perfect linear system. And not just a little - the best trained listeners are ten times better than is mathematically possible. Obviously this is taking advantage of high order non-linear detection and math that has not been discovered yet, but this performance advantage is typical of biological systems.

    Nature has been running billions of complex biological experiments a second for the last 4 billion years so our analog and digital data collection is paltry in comparison, and will remain so for another century of exponential growth in hardware and software performance.

    When you start talking about spatial perception, you get into the realm of vectors which are hundreds of times more information than microphone pickup. The bones in your inner ear trace three dimensional patterns which become three dimensional motions on the basilar membrane and turn into massively parallel three dimensional impulse trains. The standard model of stereo ignores this completely, and so recording, speaker design, acoustics, etc are all grossly oversimplified.

    1. The field of study call Auditory Scene Analysis is where it's happening with regard to this sort of stuff - it kicks off on the infamous Cocktail Party Effect, the fact that one can "listen" to a single voice amongst a hubbub of voices all rabbiting on as well; or, "the phenomenon of being able to focus one's auditory attention on a particular stimulus while filtering out a range of other stimuli, much the same way that a partygoer can focus on a single conversation in a noisy room."

      Turns out human, and animal hearing is remarkably sophisticated in how it can perform - the simple theories of masking often tossed around in audiophile circles are almost irrelevant, when faced with the ability of the brain to sort things out.

      1. I'm not sure how this may apply, but it's from Francis Crick (the DNA guy) in his "The Astonishing Hypothesis" book.


        An example of selective attention in hearing is a subject concentrating on the sounds coming from earphones into one ear while trying to ignore different sounds coming into the other ear. Many of the sounds coming into the unattended ear do not reach consciousness but it can be shown that they may leave some trace in the brain and may sometimes influence what is heard in the attended ear. At some level they are being registered by the brain.

        1. Very relevant, in fact. Essentially the brain can lock onto the features of one object that it "hears" in the total sound energy coming in, and keep monitoring that single element to the exclusion of everything else. It even "fills the gaps" - if a loud sound momentarily masks a softer sound which it's focusing on, then the brain will still keep "hearing" the soft sound, through the masking loudness! This is called, "perceived continuity".

    2. Not quite right!
      The origin of sexual reproduction is supposed to be started about 1.2 billions years ago.
      If you do not want to blame Darwin and Mendel it takes more than a second to give the learned or adopted abilities to the next generation. If you want to get further into non-linear detection and math just start with Ciarlet, Schultz and Varga!

  7. Gotta admit, I'm confused about our current discussion. Where are we going and why?

    If one can make a digital copy of an LP and maintain the "warmth" and the "sound of vinyl" then obviously the reason for the difference in sound is not due to anything related to analog vs. digital. I just read a thread by jazznut ( ) and where even Ted Smith asserts "I still suspect that you might be comparing mastering, not formats or devices. When I’ve recorded vinyl in 24/192 or double rate DSD and played it back it sounds just as rich as the vinyl…" Yes, I agree, just as rich as the vinyl.

    It's obvious to me that "There is no such thing as an identically mastered LP and digital counterpart. LPs are far from flat. Many love this sound. Imperfect sound for a limited time. This is also why playing back the LP version sounds different from digital. And why if you record the LP onto digital it will sound just like the LP; digital can capture vinyl, but vinyl cannot capture digital." as posted by Elk on the same thread. Couldn't have said it better myself.

    If you want the "sound of vinyl" just master/mix it that way and put it on the (digital) disc.

    Anyone here have experience doing just that? Paul, you said the upcoming music project sounds "live." Is that better than vinyl to your ears? Different than vinyl? Which do you prefer?

Leave a Reply

© 2022 PS Audio, Inc.

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