Digital dilemma

July 11, 2022
 by Paul McGowan

Perhaps the most difficult technology to wrap one’s head around is digital.

Analog? Not so much. Consider that it’s not that hard to understand how the quality of a tape head or a phono cartridge has a direct and obvious impact on sound quality. It has to.

Digital is a whole different can of worms. 1’s and 0’s should be easy to maintain quality. It’s far more difficult to try and understand how bits on a hard drive or, for that matter, bits sent across millions of miles of space can be affected.

Inviolate performance. Perfect sound forever. That was the promise.

And yet, DACs and transports sound different. One USB cable vs. another makes the difference between good and great.

Over time we’ve been learning what makes digital audio sound different. We’ve come to recognize and own up to the fact bits are not just bits. That the timing, noise levels, and quality of those bits changes that which we hear in music.

We never perfected analog and I sincerely doubt we’ll ever perfect digital.

But, we’re moving forward in positive ways and music is getting better for it.

That’s got to be a good thing.

Subscribe to Paul's Posts

29 comments on “Digital dilemma”

  1. If nothing is ever going to be perfected.. then how long can improvements keep being made? (Do Steps just get smaller and closer together?)
    Is good enough ever reached?

    I’m not convinced music is getting better.
    Maybe some techniques in recording and playback are headed that way.

    Thanks for the reset ✌️ 😀

    1. In case recording technology improves on the way, I’d say improvements don’t end.

      The problem is, this affects only new music from then. Old music doesn’t get that much better than from today’s media made from the analog tapes or digital sources I guess. And new recording technology and media production still works hard to surpass the level reached in the 50‘s.

      And the main problem probably is, there’s no interest of industry and most listeners, to noticeably further improve mass produced music’s sound quality. And if all the effort is worth listening to a few boutique labels in future is at least questionable.

      So it may come, that…just as today…music of the 50‘s to 70‘s stays the one with in quite a few terms the best sound quality and in classical music not rarely the one with the best interpretations. Could be that sound quality also doesn’t get much better for digital productions than today…just due to lacking interest of the industry and typical music listeners in high end.

        1. We’re certainly much better in single technology aspects, but if you ask for the overall result in interpretation, music quality and sound quality…phew…at least it’s hard to convince anyone we‘re much further.

          If we look at those ultra high end freaks, today listening preferred to tape machines, record players, tube amps and say e.g. planar speakers, Tannoy speakers, horn speakers…well speakers in general (which all didn’t change that much in technology details)…the question for groundbreaking progress is justified.

          In digital progress definitely happened, just for the most part in direction of convenience, better standard quality and improved mass production and distribution. But digital also achieved to be roughly on the same level as the best analog on an overall level (better in technical details certainly)…which again is not groundbreaking after 40 years, but great for us high end folks who not only want to listen to golden era music.

  2. Imagine how many home-audio brands would evaporate & employee’s would be without a job
    if canned music ever reached perfection…no more need for R&D or listening tests, or
    measuring stuff…& of course once A.I. learns how to perfect digital, then it’s ‘game over’ 😉

  3. What is the specific digital (!) dilemma here, Paul? As far as I understand the ADC-process the 1-bit way requires no sophisticated computational reconstruction process with more or less tricky interpolation schemes – but produces a lot of noise. However in comparison to the pure analog signal transmission digital transmission is always (bit-) perfect concerning the relevant data. I am pretty sure that the biggest degree in imperfection is found in loudspeaker designs. Simply compare the resolution of finest details revealed by a decent pair of headphones with the more or less strange sounds produced by a loudspeaker in a non standardized listening room.

  4. I attend live non-amplified concerts for perfection. I don’t expect it from my home audio system, nor do I need it to fully enjoy the music. That’s what keeps the audio industry alive.

  5. Does everyone really want perfect reproduction? We all listen to music on different systems, all with various colorations(both statically and dynamically). And we all seem to pick out own combination of colorations. I have my own biases on what I consider a good system and they are often quite different from other listeners. I lean towards dynamics and detail and many of my friends lean to consistent rich warm reproduction no matter how the recording was done.

    1. Yes indeed hahax! And if you do want perfect reproduction, your definition of ‘perfect’ may be different from mine. Like you, my preference is for dynamics and detail and I believe the more practical way to improve those qualities that appeal to me is going down the digital path. Now before those who prefer analogue sources jump down my throat I quickly admit I haven’t experienced, for example, a high-end vinyl rig and that may well sound just as glorious as (or maybe even better than) what I hear form my digital-only rig. The point I wish to make is that the sound quality produced from the technologies involved either way are comparatively very close to each other. As paulsquirrel so often points out, the bigger difference with which we should concern ourselves is the sonic ‘gulf’ between the output transducers and our ear/brain.

  6. I simply do not have the time this morning to write a comment that would cover all of the differences between analog and digital audio. This is a thorny subject and very controversial. Maybe we should start with something easier. like politics! 😉

  7. The positive emotional aspect of Music, is manifest when we are open and receptive. I can try to just sit down and listen to enjoy, but cannot ‘force’ myself to do so, like I do when exercising. Music has a deeper connection to our vibrational and emotional state, hence we can find pleasure in the oldest Edison Recording that a Digital System does not provide when we are calm, relaxed and musically receptive. Digital perfection is relative to the observer and the transmission medium, we can perceive audible differences between Cables, Circuits and Components. Just as technical specifications sheets often, do not correlate with anything musically related. So, I manage my enjoyment of Music, when my body and mind are receptive. Even the Birds singing, can make me feel happy inside, just as barking Dog can vocalize joy, pain or concern. We in this wonderful hobby, continuously train ourselves to be better, like in any profession, we should strive to be great, and not settle for any less. But the metrics we use to describe sounds and perception, make it difficult to explain and quantify. When it sounds good, it just sound good. When we can hear distortions, we audiophiles react differently than the non-audiophile, as we have dedicated thousands of hours as focused listeners and avid hobbyists.

  8. I appreciate very much that you are always open for fining out about the little things that are not as perfect in digital as initially assumed and proclaimed by many and address them in your products. The feeling that you are at the forefront and at the same time try to keep it affordable is always a good reason to buy [email protected] (Ted included).

  9. Forgive me, but I don’t get it. There is of course much room for art in ADC and DAC moments in the digital music chain. But moving bits around should be unaffected by cables and internal wiring. Otherwise, there would chaos around the world as digital banking transactions are changed by jitter and copying files from computer to computer doesn’t work as you expect.

    The argument that moving bits and interpreting them in real-time somehow is subject to misinterpretation because of slightly deviating clocks is theoretically reasonable, but easily remedied by sufficient buffering by the receiver. As long as the receiver stays ahead of the sender, it’s the same as transferring a file, and we know that’s bit-perfect. If necessary, a protocol that inserts error correction bytes occasionally to enable fixing any errors in the received buffer fixes even theoretical vulnerabilities. So, what am I missing?

    1. Jitter would not impact static binary bits as in banking transactions. Jitter, however, is a major concern when those binary bits must be clocked out at exactly the right moment. If jitter was as easy to remove as you make it sound then there would be no need for further DAC improvement with respect to jitter.

      1. But that’s my point. Either way, with the bits at rest or in motion, the process is the same. No hardware can inspect all the bits at the same time, for example in a huge file, so they must be examined in some sequence, as a stream.

        So, in both cases the problem is the same–look at the electric signal every nth of a second to tell if it’s a one or a zero. Whether it’s a bunch of bits coming over a wire from a different audio component, or a CPU in a computer examining bits that are streaming in from the hard disk. If jitter caused some zeros to be misread as ones, or vice versa, then there would be chaos everywhere. And there isn’t, so that doesn’t happen.

    2. You’re both correct. A buffer (which we call a Digital Lens) is indeed part of the process needed to clean things up. However, just building a buffer that stays ahead of the data isn’t all that’s required.

      That buffer’s output clock must itself be jitter free or you haven’t accomplished much. Simply buffering of the signal is in and of itself helpful but not the reason you do it.

      The entire effort is based on the need for a perfect, jitter free data stream to be presented to the DAC. Buffering is only partly what’s required to do that.

      1. But that’s my point. Either way, with the bits at rest or in motion, the process is the same. No hardware can inspect all the bits at the same time, for example in a huge file, so they must be examined in some sequence, as a stream.

        So, in both cases the problem is the same–look at the electric signal every nth of a second to tell if it’s a one or a zero. Whether it’s a bunch of bits coming over a wire from a different audio component, or a CPU in a computer examining bits that are streaming in from the hard disk. If jitter caused some zeros to be misread as ones, or vice versa, then there would be chaos everywhere. And there isn’t, so that doesn’t happen.

        1. With respect mfuchs, I think you may be missing the point. You’re correct that ‘the bits [are either] at rest or in motion’. Another way of putting it is that bits are accurately transferred through cyberspace independant of time. ‘Timing’ is the key word in this context. It’s not really whether a bit has been incorrectly flipped ‘on’ or ‘off’. With modern DACs, I think the probability of that occuring up to the point when the signal is converted to analogue is very low.

          I think it’s commonly accepted that the human ear/brain system is exceptionally sensitive to timing discrepencies. So the better ‘clock’ that’s used in the digital to analogue conversion process, the better (more accurate) sounding the analogue signal will be.

    1. Your idea of white could also be different than another one‘s, but white is somehow defined. The rest is your preference or subjective opinion.

      As long as no perfect, black or white exists, but just approaches from different directions, your question is valid.

  10. Bits have personality, good and bad days, mood swings. They don’t like to be thought of as Cybermen, marching in perfect unison. They want audiophiles to care about them. Look after your bits and and your bits will look after you. (Now there’s a punchline.)

    Fat Rat, before you ask, the thing closest to perfection is Root 66’s switch flick over deep third man (or deep backward square, depending how you look at it). I think he actually apologised to Tim Southee, who looked deeply offended, at least after he stopped giggling. Pure joy.

  11. A digital audio stream is like a DNA strand. A few bits or genes out of order can be innocuous or it can be disastrous. Mutations can even result in something better, though that is rare. Both digital audio and genetic streams are order and time dependent. Things have to happen in the right sequence at the right time.

      1. I was just reminiscing what long-distance telephone calls were like before direct-dial, digital and satellites. We had to set up landline calls through operators, and wait for the operator to get back with the connection (station-to-station or person-to-person). Then during the call we had to shout into the handset mouthpiece in order to be heard at the other end, due to the landline signal strength weakness and dropouts. We kept the conversation to only a few minutes due to the high cost. Long distance calls were so expensive that when someone knew someone else was traveling through another city where they had close friends, they would ask to traveller to call their friend and tell them such and such says hello! Whenever you got a long distance call, your first thought was, “Who died?”

        1. Yeah, “Who died?” was very important news…still is I guess 😎
          By contrast, these days, when I ring friends that are overseas, half a world away, on my cell-phone through ‘WhatsApp’ & we talk for an hour or so, the speech is as clear as the proverbial bell & it’s free…well, the cost is covered by my monthly internet fee.
          My, my, look how far we’ve come in just a few decades.

Leave a Reply

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

4865 Sterling Dr.
Boulder, CO 80301
1-800-PSAUDIO

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