In this post, I'd like to discuss a design and marketing practice that's common in the audio industry that I feel ultimately hinders meaningful progress. I call this, "phenomenological audio."

Phenomenological audio is audio that relies on some kind of "trick" to generate a noticeable and, temporarily at least, pleasing effect.

For me, the poster child for phenomenological audio is an iconic loudspeaker design from the 1970s and 1980s that relied on firing 1/9th of its output at the listener and 8/9ths at the walls of the room. Enveloping? Yes. Pleasing? You might think so, at least at first. Accurate? I had a hard time ever hearing them that way.

This example is something of an extreme. Phenomenological audio typically has less obvious manifestations. One of them is something I mentioned in a previous post: the deliberate inclusion and shaping of THD. Another, which has been popular with some loudspeaker designers, is the introduction of a mild frequency response dip in the upper midrange -- sometimes referred to as the "BBC dip." This increases the perception of the sound stage and imaging, but it also zaps a lot of instruments of their vitality. These are examples that are common enough to lump together. There are countless others that are unique to specific products and used to give them a "signature" or "voicing".

Phenomenological audio is incredibly compelling for folks who try to make a living selling audio equipment because it provides differentiation in the market. "Did you hear the incredible [property] that [XYZ product] has? Wow!" The buzz starts and the reviews almost write themselves.

Then something else comes out with a different kind of buzz (figuratively and sometimes literally). And so starts the eternal cycle of buy-and-dump in the hopes of securing an impossible audio Nirvana that's composed of a list of contradictory illusions.

One of the most annoying aspects of phenomenological audio is that it doesn't have to be deliberate. It's surprisingly common for someone to come up with something they love that relies on something phenomenological and then try to convince you to love it too. They may sincerely think they've developed something that does things better without realizing the reason they love it owes to what it adds, removes, or masks.

I'll admit to occasionally being seduced by phenomenology I accidentally designed into something. However, I can say with confidence that I've never deliberately designed phenomenology into a piece of hi-fi gear. Quite the opposite. My intention has always been to minimize the impact a circuit has on the content it's serving. I love and respect music and the people who make it too much to want to editorialize what they do.¹

It's easy to think that what I'm talking about here is just the eternal subjectivist versus objectivist debate but using different words. That's not the case. Objective measurements can tell us when there's something obviously askew, and this is almost always audible. But I've also heard plenty of designs that should have had "inaudible distortion" based on objective measurements but which nonetheless sound different -- and some quite awful. My experience is hardly unique. We're getting closer, but we still don't know exactly what to measure to tell us everything we need to know to effectively predict how a given design will impact the perceived signal.

So, the antidote to phenomenological audio isn't objective measurement. Rather it's intention and understanding -- of the technical and psychoacoustical aspects of what you're trying to do as well as your own personal susceptibility and vanity. It can sometimes be a bit of a challenge teasing out what's phenomenological from what's not. But recognizing that it's a thing and the common traps is a good place to start.

To be pedantic, since design is all about balancing tradeoffs, we'll never be completely free of phenomenological audio. When engineering a loudspeaker, do you go for the tightest, most linear bottom end for a given size or do you provide some extension at the cost of linearity? Which compromise cheats the music the least? Trade-offs between conflicting phenomena like this are everywhere in audio design.² The difference really comes down to whether you view phenomenological consequences as a necessary evil that you do your best to minimize or as something to deliberately encourage.

I hope this doesn't sound like one of those "Only I can fix this" screeds that hucksters use to enlist adherents. I don't mean it that way at all! I really just want to give a name to something that IMHO has consistently been an obstacle to making good progress in audio. Folks supporting phenomenological audio have some of the loudest voices in our field. But there are plenty of engineers who do their best to avoid it in their work. I'm grateful for them and that their efforts not only drive the industry forward but also make music lovers incredibly happy, day after day after day.

---