Looking in the wrong places

September 15, 2022
 by Paul McGowan

Digital audio is a very complex subject, one most of us don’t really understand at a level deep enough to make a reasoned technological change.

Yes, we can change cables, equipment, programs, and even formats and hear the differences. But most of us aren’t able to dig deep into circuitry and programming to make those changes.

Our changes are broader—surface level as opposed to digging deep into the machine.

As someone on the other side of the fence (with just enough deep knowledge to be dangerous), it’s always a balancing act helping people get to where they want to go by separating fact from fiction as best I can.

For example, in our ongoing quest for getting better sound out of our digital audio systems we work with cables and purifiers to improve the resulting audio. And it works! But why?

Lower noise? Less jitter? Better signal shape?

Without a clear understanding of what we are actually attempting to control or improve, most of what we do is more of a crapshoot than a reasoned approach.

I suppose the point of all this is that as we delve deeper into the benefits of galvanic isolation—benefits we will all have access to when we soon release the new DirectStream MK2—it might set a lot of folks to scratching their heads as to why those fancy cables and purifiers have so much less of an impact.

Once you know where to look it’s easy to see.

Subscribe to Paul's Posts

39 comments on “Looking in the wrong places”

  1. Hi – John Darko made a great podcast on the topic a week or two ago. The only idea I got from more than an hour rumble is – the topic is much much more complicated then i thought. Much more than I can understand (as a graduated technician and former IT). And I didn’t think it was any simple.

  2. I like simplicity and if it isn’t broke don’t fix it. My made in Belgium Philips CD630 16 bit with 4x sampling and dual TDA1541A DAC’s with 27 bit digital filtering with digital output just in case I ever want to use the great Philips CDM-4/19 transport on another DAC is all I will ever need. Some don’t have the digital outputs but sound good. I have a 1 bit Philips based by Conrad Johnson and a 14 bit by CJ. A few early Magnavox too. My Philips CD630 which is hard to find in any condition let alone new condition with remote I was lucky enough to purchase recently and is what I’m using now. Don’t ask me why but I’m a vintage CD player nut and some of the sought after players can fetch a big price tag. I’m just a plain vintage stereo nut. Some others can be found reasonably priced. Early Philips made some great bullet proof transports. Much better than the Sony transport found in many CD players.

    1. My sentiments exactly Joe! Since 1988, I’ve had the Philips CD880 (same dual DAC chipset), but additionally with the single Crown S1 designation! For the past 34 years, it has always been extremely musical, non-fatiguing and an accurate fidelity producer of redbook CD’s! Last year I sent back a Marantz SA-KI Ruby SACD player, as it’s sonics could only equal that of my venerable Philips!

      Currently, I have the CD880 variable outputs directly driving my new GaNFET Class-D power amp…absolute STUNNING soundstage width, depth, imaging and You-Are-There holographic presentations!!

      1. The person I bought the 582 from said it was modified, I opened it and noticed a few other changes besides the DAC being a 1541A S1 Crown. That must have been the other modification they did. The player sounds awesome.

  3. I was bitten by the digital bug back in 1969 when is was taking a math course called Integral Transforms. It was there that I first learned about Fourier series, transforms and theorem, all so essential to the understanding of acoustic signals and digitization. I first heard the word digitization when the professor said that if you know how to integrate then how to digitize is simple. Let me try to explain.

    If you have a curve ( a signal ) of some type and you want to know the area underneath it you integrate. When you integrate you take a amplitude values along the curve at precise intervals. You then multiply each amplitude value by the width of intervals. Thus each amplitude sample times the width of the interval becomes a tiny sliver of the area underneath the curve. It looks like a picket fence with the top of the pickets defined by the curve. You sum all of the individual samples and you have an estimate of the area underneath the curve. The smaller the interval the greater the number a sample points and the better the estimate of the area.

    As an aside the first electronic integrator was built in 1946 using vacuum tubes. The theory of all of this started when a guy name Newton invented calculus back in 1665. Thus, it only took about 300 years of more mathematics development and a bunch of vacuum tubes to make the first electronic integrator. 😮

    Now, once you have all these tiny slivers ( pickets in the fence ) of area, instead of multiplying the amplitude of each point by the width of the interval, simply store all of the amplitudes individually and you have digitized the curve. Simple. 😉

    1. I took similar courses but they were purely mathematical and didn’t deal with the sub topic of digital audio. They were just talking about digital technology which I used extensively as a logic design engineer many years ago. There are a lot of differences in the use of digital technology that audio was not even a twinkle in the eye of most engineers. The purpose of higher mathematics is in large measure necessary to prove out physical principles and I believe it had nothing to do with digital audio at the time. In my mind, seeing what’s going on right now in digital audio it’s more of a minefield and new problems crop up constantly because digital is not analog… It may approach analog but never reach it.

        1. Tim, When I joined IBM in 1982 I was in the Advanced Thermal Technology group. At that time we were working on multi-chip-modules (MCM’s ) that had 100+ bipolar chips in them. Because of how much power they consumed they had to be water cooled. It was clear from the technology plan that in another two or three generations of bipolar chips those MCM’s would be consuming 1KW each and a mainframe computer might have three or four dozens of these modules. Customers biggest complaints were that the mainframes required too much space, too much power and too much water. While the switch from bipolar to CMOS was painful ( IBM handled it very poorly and a lot of people lost their jobs ), but staying with bipolar was not feasible.

    2. I took similar courses but they were purely mathematical and didn’t deal with the sub topic of digital audio. They were just talking about digital technology which I used extensively as a logic design engineer many years ago. There are a lot of differences in the use of digital technology that audio was not even a twinkle in the eye of most engineers. The purpose of higher mathematics is in large measure necessary to prove out physical principles and I believe it had nothing to do with digital audio at the time. In my mind, seeing what’s going on right now in digital audio it’s more of a minefield and new problems crop up constantly because digital is not analog… It may approach analog but never reach it.

  4. And this is why guys like Ted Smith, Rob Watts & others of their
    ilk exist, because they are driven to dig deep into this ‘stuff’.
    Meanwhile, I do what an audio enthusiast was born to do
    & that is to listen to canned music on my home audio rig.

    Recently I discovered a modern-day Jean-Michel Jarre named ‘Madis’,
    from Poland, who plays fantastic electronic music…take a listen:


    …or not.

  5. And even the Maxwell equations are based on simplifying assumptions! And don’t forget:varying electric current in a wire is associated with the generation of a varying magnetic field which can even build up magnetic vortices. And what about the resulting interactions? And what about magnetic monopoles? There are so many unsolved scientific questions concerning atoms and electrons and quarks etc! 🙂

      1. The physical model theories written in mathematical equations allow the existence of magnetic monopoles. – be aware that Maxwell did omit specific terms in his equations! I know a lot of researchers some from the IBM laboratories focusing on these “objects”. We have to accept that we are unable to fully comprehensively describe the real world and that we have to introduce such strange things as “dualism of particle and wave” and that we are unable to “understand” effects of quantum physics. We only can play around with descriptions, equations and experiments.

        1. ps, I have no idea what you background is, but as a retired physicist ( who worked at an IBM laboratory ) I will be first in line to admit that we do not completely understand the physical universe that we live in. However, I do not find the wave particle duality “strange” and I think that scientist are very able to understand the effects of quantum physics.

          1. Maybe it is only semantics? “Understanding” in my point of view means that I can explain “why”. But physics never asks “why” but “what is” and starts describing “what is” by mathematical equations and/or theoretical models. Only agreeing on these simplistic equations and models it is possible to “explain” why. But do these models and equation fully describe a most complex reality? 🙂

  6. After reading this morning’s post, I now understand why I have stated during the past several years that I love my Wadia CD player. It certainly isn’t State Of The Art at present and I can’t play SACD format but my deepest feeling has been that it’s not time yet for me to make sense of what I see as quite a bit of confusion in the digital audio domain to make changes in my music system. It matters much less to me to try to achieve State Of The Art than enjoying the music that emanates from my speakers today.

    I waited over 15 years since the inception of the CD player to find a unit that sounded head and shoulders above any other CD player that I had auditioned during that
    period of my life that soothed my soul by being truly musical to my ears. I stopped thinking about moving forward in the quest to find the Holy Grail when I’m more than thrilled with my home musical experience at present.

    Digital audio technology is just too complex at the moment for someone like me to want to start messing around with happiness once again and potentially winding up angry and frustrated. Any audio enthusiast can always find a way to better sound quality but I’m choosing a safer path. We each need to choose our own path that satisfies our musical senses. This is the possible reason why so many people keep ‘pushing the envelope’ with the hope that digital technology will move forward rapidly towards audio nirvana. I just hope that we’re not heading towards proving the Peter Principle in the audio world.

    1. The Wadia was a great CD player in its day. Instead of just using simple interpolation when it upsampled it had a small computer on board that used splines when upsampling to make the upsampled results more realistic. I know a couple of guys who loved their Wadias, but all have eventually moved on.

      I had a pretty good CD player when I discovered SACD’s. It blew my mind, how could digital sound so good? I went out and bought a Sony SCD-1. After 14 years when it started to become unreliable I was forced to discover FPGA based DAC’s and the world class digital genius of Ted Smith. I have never looked back.

      Take the leap. When the new MK2 comes out get it and the PST transport from PS Audio. It will make you CD’s sound better than they ever have and you can play SACD’s if you want to.

      1. It’s really a matter of financial priorities Tony. I just became a Grandfather of identical twin girls and I’m planning a 2-3 month vacation which will end up in San Diego where both of my sons and my Daughter-in-Law live. I plan on giving each of my these little ones a sizable College gift as well as making plans to move to that area within four years. Then, I’ll need to be able to afford to live in that state. Ergo, the joy of music is greater than my need for the latest technology now.

    2. Wadias are still some of the best Redbook/PCM players in existence.

      The only problem is they can’t do anything above 24/96 or DSD.

      The first time I heard a Wadia was the first time I ever heard a CD sound like MUSIC.

      I readily admit when I was auditioning the PS Audio DirectStream there were many areas of sound that it did not get right as compared to my Wadia, which is why I did not purchase one.

      Ted was great, but often the next revision of firmware would fix one area but break another, it’s always a catch-22.

      Honestly, I was not able to exceed the performance of my Wadia until I spent about 3x its purchase price, and I wouldn’t have done that had it not become clear that Wadia was effectively dead and being allowed to wither on the vine following their acquisition by Fine Sounds.

      1. Thanks for all of that information about Wadia, I just biting the bullet and am hanging in with my current cd player. You don’t have to be a critical listener to hear immediate differences when you audition a component. That’s the way it was with the Wadia, Love at first listen. If I want to listen to DSD I’m going to be downloading occasionally and hopefully make more of a habit of getting higher quality digital sound without a transport.

    1. Ha, if so Aston Martin ought to put it on the options list.
      Apparently, to prevent, they used to put cotton cloth strips impregnated with waxoyl or similar between the two dissimilar metals but I expect with the alloys and composites used in today’s construction it’s much less of a problem.
      Apologies as I fear I may have steered this some way off topic. 😉

  7. I suspect that there are some that think/believe/hope that there is something new and groundbreaking around the corner with DACs and streamers. They have been around for 40 and 15 years respectively and personally I think for the vast majority of people they have already reached the end of the line.

    I use both Wi-Fi and usb, what might be considered the best and worst solutions. Usb is considered bad by some because it carries power down a copper cable, so is not galvanically isolated, but usb is excellent for any kind of data. So some manufacturers optimise usb by generating extremely clean usb power.

    There are now just loads of good ways of doing digital, and for my part power conditioning had always assisted my digital audio.

  8. Paul is correct – much of the high end audio business is indeed based upon “we did THIS and it sounds SO MUCH BETTER.”

    Understanding WHY that is is secondary to the audible improvement created, as most customers don’t care about the measurements or rationale, just that it sounds better than what it is replacing.

    1. Right, and that’s both a blessing and a curse. Do this or that and get a better result is fine but often it leads to pretty strong beliefs that are incorrect.

      Like spending a lot of money on better ethernet cables when the real answer is to be found elsewhere. It works after a fashion but probably has a better answer that’s a lot more effective.

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

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