Why Your Objective Sound Quality Argument Sucks 2005-07-15 04:31:27 In the course of a Head-Fi thread in which I apparantly played a key part in getting closed, I realized that many/most audiophile subjectivists (and more specifically, those who consider their ears infallible) cannot be persuaded by reason.Gasp. Well, I guess everybody else here knew it already.But I also realized something much deeper, that the competing schools of thought of audio look remarkably like paradigms, in the Kuhnian sense. One of the main implications of this is that both sides, objectivist and subjectivist, can explain all observable phenomena in an acceptable manner (to that side). The other big implication is that one school eventually wins out for reasons indirect to the correctness of that theory - often for social or pragmatic reasons. In other words, we can't reason with subjectivists because our observations already have explanations that are more believable to them, and even if we are in fact right, it doesn't really matter. Therefore, the way to argue objectivism is through other effects.Since the original thread is closed and I wanted more discussion on the topic, I'm reposting the relevant part of the post here. I hope this will stimulate discussion on how to best persuade people that a fully objective approach to evaluating audio is best.QuoteAnybody can find a difference between two different things, even two of the same make and model, but it takes hard proof to say that difference means anything, backed up with logic and experience. And it's here where I think we're never going to agree. For every device, there are always going to be any number of parameters and failure modes. Without exception, the audiophile community will attribute differences in sound to every parameter and failure mode of the device!Thus, interconnects sound different because of different dielectrics, resistances, capacitances, characteristic impedances, electrostatic forces, microphonics, crystalline structures, quantum alignments, conductor widths, conductor lengths, connector types, connector materials, conductor materials, conductor distances, numbers of conductors, solder types, and braid topologies. In other words, according to various audio manufacturers, audiophiles and Head-Fi members, every conceivable difference to a cable changes the sound.This is not sound science! Or sound engineering for that matter! Occam would roll in his grave if he saw a state of the art engineering theory such as this, practiced by the most famous people in the industry. No, I'm not saying everybody believes every effect is important - but that's an even more complicated situation, because then everybody has their own little theory as to how cables work, and will choose different effects to work towards the same goals.What does all that mean in the end? Surprisingly, not as much as you'd think I'd say, but still a lot. Even if you attributed audible differences to all those effects, you'd still have a consistent theory of interconnects, and there wouldn't be any evidence I could show you to sway you towards thinking one of those effects didn't matter. You could even build cables based on your theory, and they could sound good, and people would buy them.It's on the fringes that this sort of thinking breaks down. * When some people think a $200 cable sounds as good as a $1000 cable, or a $10,000 cable, that's because the $200 cable got it right and everybody else is overcharging. Or the designers got lucky, or they've tapped into a hitherto-unseen effect. * When a cable that takes all these effects into account sounds only as good as those that don't, the other cables got lucky, or they've tapped into a hitherto-unseen effect. * When some differences are clear as day sometimes and impossible to detect in others, it is due to emotions/stress in the listener, or a flawed audio system, or a flawed detection system, or a hitherto-unseen effect. * When RCA connections still invariably used for even the top of the line gear when the optimum characteristic impedances are obtained with BNC or coax, it's because of the entrenched standard, vendor stupidity, or a hitherto-unseen effect. (Anybody who doubts me on this can show me a 20Ghz switch with RCA inputs.) * When 6N copper so highly desired for interconnects, yet the amplifiers themselves use regular copper and (heaven forbid) 66/37 solder, it's because the interconnect can affect the sound independently of the materials inside of the electronics due to some hitherto-unseen effect. * When audio salesmen tell you to purchase cables based on a percentage of the total value of the system, rather than how much intrinsic value the cable adds to the system, it's because the cables are always less important to the final sound quality than the other components of the system, regardless of how any of them sound or cost.Of course you can answer all of these questions. Everybody can. You might be able to answer them better than I have. That's not the point. The point is that to answer them you will almost invariably need to appeal to ad hoc hypotheses. Either here's a flaw somewhere else in a system, or a new effect is discovered, or an existing effect is less important than some other effect. Never is an effect considered completely inaudible based on new evidence, nor is an effect generalized to explain more evidence. Rarely (if ever) does an experiment in a new cable result in a poorer sounding cable. Ever since audiophiledom has started, and people started caring about cables, the theory of their quality has never simplified over time, and it remains a collection of guidelines about how certain parameters of construction affect certain dimensions of listening, without significant predictive power about how not to build a cable beyond what has been already manufactured. Of course I'm singling out cables in particular here but I could repeat this argument for all sorts of other things.Truth, per se, is not directly a part of the conversation. Everybody can explain the evidence, and if you are comfortable with believing that every effect is audible and everything sounds different I won't be able to convince you otherwise. But if you don't - if you admit that some effects are not audible, that some audible differences may in fact not exist, that we humans are smart enough to agree on what is and is not audible, and that we can program a computer to test for everything that is audible - then everything else must fall into place.Original post here.