Hey morricab,
How are you doing? Forgive me for the length of this post as I don't have time to write a short one.
Ah, the ol’ “accuracy” thing, huh?
Fundamentally, though I agree with you on many things, I disagree with your hypothesis here because for me “accuracy” as an independent arbiter is nothing more than individual preference expressed with a level of conviction.
Unfortunately, we are now only repeating arguments covered here previously. Though there are many who wish to discuss this topic, hence perhaps why it seems to emerge into new threads each year, there is still no single, fundamentally agreed upon concrete ideal as “accurate”. There is only “accurate to within an objectively measurable threshold” and “accurate to how I remember it or think it should sound”. One is an objectively defined correlation apropos audibility, the other a subjective preference formed into an opinion. Both of those are completely problematic, and have nothing to do with the experience of live unamplified music.
We can have two world-class classical engineers show up to the same scoring session asked to bring their most accurate mics, and they will bring different ones. They can then be asked to record/mix/master the results as accurately as possible to what they just heard live and will be doing so via memory relative to preference. We can ask two world class component designers to create the most accurate amplifier possible and they will diverge in topology, implementation and parts selection. What’s more, we can have those two designers produce amplifiers that by accepted methods of measurement are completely objectively “accurate” yet still diverge in how the subject perceives the way in which those amplifiers make music when paired with a transducer to convert the signal into sound waves.
Yes, we as a species have a long history of listening to unamplified (non-electronic) music as I mentioned above. But even acknowledging that fact, with thousands of years of experience burned into our socio-cultural DNA, if not our biophysiological makeup, we can still not get two people to agree on what the most accurate way of recording and playing back the same acoustic performance will be, even if they were there at the time.
So “accuracy” in-and-of-itself has no meaning for me. It’s a term that has no utility value apropos my perception of music. Not just when applied to the question of “how do you know your favourite electronic music is being played back accurately?” but also when applied to the question of “how do you know any type of music is being played back accurately?”.
Why? Because the answer will mean either one of two things with the following justifications:
A) Yes, because it is “accurate” to within an objectively measurable threshold (though directly comparing the musical waveform of the encoded format versus that same waveform when played back and measured at the speaker/room interface will of course render any illusions of accuracy to indeed be illusions), or;
B) Yes, because it is “accurate” to my individual preferences expressed with conviction (or in other words, yes, it matches what I like to think/remember/believe “accuracy” sounds like based on my prior experience).
Therefore, if it’s possible that two individuals possessing a lifetime of attending and/or playing/conducting classical concerts at the highest level will still be overwhelmingly likely to diverge in their preferences apropos their choice of hi-fi system, each claiming theirs must be “more accurate” because they can back it up with superior knowledge of what live unamplified music sounds like, what does that tell us about the utility value of that reference?
Nothing. Again, it only tells us about their preferences, and the justifications they use to make them.
Granted, it has some utility value for the manufacturing and measuring of components. But even in cases in which those measurements are off the chart amazing, we still have plenty of people who do not prefer those components in spite of their objective “accuracy”. I’ve heard plenty, as perhaps have you. Their designers are adamant I am wrong, since they have many, many years of unamplified live music attendance, and can produce a chart showing me how accurate their amplifiers are when playing test tones on a bench.
Therefore, if it’s clear that despite the fact we can measure degrees of accuracy objectively but still not prefer it subjectively - or worse, consider the possibility the first-order effects of achieving objective accuracy have come with second-and third-order effects of emotional detachment or psychological objection, how useful is accuracy as a construct in reality?
Of course, the argument goes that in order for a component to be “truly accurate” (we now get even more qualifying), it must reproduce both the waveform without deviation and the emotional content. And we measure the emotional content of the original event how, exactly? By relying on the subject’s opinion? Do we then use a subjective criteria to judge an objective one? Wouldn’t that be just as problematic as relying on the subject’s… er, subjectivity?
We could measure the brain’s response directly to see what areas were stimulated and certainly I am all in favour of such an approach. I’ve posted many links to research to those who are studying the brain’s neurobiological response to music. By mapping the brain’s emotional centres relative to musical stimuli we could indeed bypass the subject’s subjectivity and get closer to some ideal of what the brain considers to be accurate, but even then, all we are ever really doing is understanding more about individual preference and how the subject perceives reality - it does not necessarily tell us anything about reality itself.
I get that in theory we say want accuracy. In practice, our divergence of how we go about achieving that suggests we just want to be happy. Claimed preference (what we say we do) is often very, very much at odds with revealed preference (what we actually do). All our systems are testament to this, with very, very little consensus among us. Even in cases in which two parties will have areas of mutual agreement as you and I do and have a lifetime of live unamplified music attendance as we also do, how we would actually set up and fine tune the same system would likely diverge in our quest to achieve the most “accurate” sound (though I, of course, would attempt to be more intellectually honest and simply say I prefer it, and give the reasons for my preference).
As I’ve said before, my perspective is that the live performance in-and-of-itself is not a particularly useful arbiter for assessing a system as the live performance begins and ends in time relative to the subject’s presence. If the subject is not there in time then they cannot experience it. Even in cases in which that performance is recorded, it is no longer live, but a series of electrical impulses stored in a medium. Therefore we can no longer experience that live performance, only the recorded one, captured relative to the preferences of those who designed and built the recording chain, relative to the preferences of the ones who recorded/mixed/mastered it, and played back via the preferences of the ones who designed and built our playback components. Preference has shaped the process far more than “accuracy” ever will.
Those variables in themselves distance us from the notion that we can ever be listening to live unamplified music via the reproduction mechanism, because fundamentally that is not what we are listening to. We are listening to sound waves transformed into tiny voltages and amplified and reamplified and reampllified again before being turned back into sound waves. Via a hi-fi system we are never listening to live unamplified music - we are listening to dead reamplified music.
That it has the incredible power to render us emotional sobbing messes at the end of the the forth movement or indeed, at the end of the first second is not lost on me. Many, many times have I have felt like I’m listening to live unamplified music, but that’s far more a testament to our brain’s ability to find meaning in the medium, than it is the medium itself (which of course, we never mistake for the real thing in the same way I never walk into the living room and actually believe Harrison Ford has a best-friend who is an eight-foot dog/bear thing and really hope he makes it out of Jabba’s palace alive).
My point in my previous post is much simpler than all the above. A girl and guitar recording tells me nothing about the ultimate performance of a hi-fi system, expect how that system plays back that girl and that guitar. That’s a useful data point. But if I want to know how it plays Beethoven, or Bizet, or Bang on a Can, or Bjork, or Tim Berne, or the Beastie Boys, to be intellectually honest with myself I’m going to need to assess those on their own terms, not the girl and her guitar’s terms.
Extrapolation of individual data points that result in a conclusion one then labels with a moniker of certainty justified via reinforcement of preexisting biases (“I heard one song on the Lyra and immediately knew it was the most accurate cartridge based on all my years of attending live classical concerts…”) is a form of hasty generalization, a logical fallacy - no more and no less. We should also not conflate consensus with robustness of evidence.
Live unamplified music is a reference for those that wish it to be so. Yet it confers on them no special ability to assemble a system or design components all others will deem accurate other than perhaps themselves, even in cases in which a degree of objective accuracy has been achieved.
Back to my main point: A square wave in-and-of-itself is a useful test for a system because it subjects the system to phenomena that do not occur in acoustic music. Given that occasionally electronic music contains multiple square waves, at multiple amplitudes, all modulating over time, combined with waveforms of other shapes, suggests it may be useful to use such music in addition to acoustic music when evaluating system suitability.
Be well, morricab!
853guy