Beware the Skeuomorphic Rabbit Hole

I recently watched a video touting the new Studer® A800 Multichannel Tape Recorder plugin from Universal Audio. It features a surprisingly mediocre soundtrack, enough marketing hype to make Don Draper blush, and a bunch of talented audio engineers waxing poetic about a digital recreation of a 34-year-old electromechanical device. And as I watched it, something struck me: Those spinning tape reels are ridiculous.

As you may or may not know, Universal Audio has been making a name for themselves lately with their UAD-2 DSP platform and associated plugins. And with few exceptions, all of their plugins are digital emulations of various pieces of coveted (and usually very expensive) hardware, including pieces made by UA itself. Are they beautiful pieces of DAW art? You bet. Do they sound really, really great? I’m sure they do. (Though I’ve never spent time with a UAD system, judging by their ubiquitous ads and ever-expanding plugin roster, I assume people likey what they hear-y.) Is this whole exercise in recreating the past with strings of ones and zeros—being perpetrated by countless developers, by the way, not just UA—a little bit absurd, and a lot bit regressive? I’m going to go out on a limb and say yes.

First, a little detour: A skeuomorph (for those of you who haven’t kept up with the recent craze in iOS UI critiques) is “is a derivative object that retains ornamental design cues to a structure that was necessary in the original.” Does this sound familiar? It should, as DAW and plugin design are probably more guilty of using—and sometimes abusing—skeuomorphs than just about any software out there. In fact, said topic’s Wikipedia article calls out audio plugs by name:

Many music and audio computer programs employ a plugin architecture, and some of the plugins have a skeuomorphic interface to emulate expensive, fragile or obsolete instruments and audio processors. Functional input controls like knobs, buttons, switches and sliders are all careful duplicates of the ones on the original physical device being emulated. Even elements of the original that serve no function, like handles, screws and ventilation holes are graphically reproduced.

Skeuomorphs can be a good thing, and I am as guilty as the next guy of getting suckered by a likable (if not lickable) interface. As with any skeuomorph worth its salt, a little suckering goes a long way towards making any virtual interface approachable and usable. I can’t deny that staring at fake VU meters sometimes just feels right; or that texture, personality, and real-world associations can help users get to know a piece of software—and the thought processes behind its algorithms—better than reading any manual could. Leave me to stare at both Ableton Live’s Amp device and Softube’s Vintage Amp Room while I listen to them do their thing, and I’ll probably tell you that the prettier Vintage Amp Room sounds just a wee bit better, even though they are, for all intents and purposes, the exact same thing. It’s human nature, and it’s all good.

What worries me is that our addiction to pixel perfect skeuomorphs—in all of its bokeh-blurred, dangling-patch-cable, and every-last-scratch-and-dent glory—is feeding our obsession with the past, and in the process, impeding the state of the art of DSP. Why are we settling for mere replicas of decades-old machines? Why not think outside the 19″ box? Why continually recreate—visually and sonically—when we could out-create?

Whenever someone sets out to “decode” a piece of hardware, perhaps they should start by asking themselves a few questions: Sonically, what is it about the hardware that is so desirable—and is it absolutely perfect in every way? Are the controls exactly where they should be, and are they exactly the right size and shape for mousing? Could this tool benefit from the flexibility of an interface designed explicitly for the computer screen? Has there been any knowledge gained by humankind in the last fifty years that could refine or augment the processes hidden inside this magical box, and could we realize those insights digitally?1 Instead of spending all kinds of time and money and effort figuring out how to perfectly model a rare British NOS 10k resistor, gingerly plucked from a console used to mix every #1 hit song ever, we should be figuring out exactly what this hardware does to make sounds that pass through it sound better, and then use our digital chops to at least try and beat the pants off the hardware.2

Luckily, there are already tons of people out there doing this: Guys like Sean Costello, who recognizes the hard work the good people at Lexicon did in the late ’70s—but then sets out to beat them at their own game, instead of wasting his time painstakingly reverse engineering a relic that costs more than my Honda Civic. Or Steven Massey, who knows his sounds and his EE, and who devotes himself to coming up with newer, better, and more creative ways of tackling old problems, creating legions of diehard fans in the process. (While you’re at it, do yourself a favor and read through his über-insightful blog posts.) Or Kush Audio, which didn’t just digitally mirror their own UBK Fatso in software, but re-imagined it in their UBK-1. Or SoundRadix, who realized the world didn’t need another analog EQ clone, and came up with SurferEQ—a simple but revolutionary EQ concept that for some reason took until 2011 to be invented. The list goes on.

I love my analog boxes for what they do. And I also love mixing with, say, Slate VCC. But I don’t really care if the perfectly distressed knob is set to a cheeky reference to a mixing console that’s twice as old as I am—it could say “Yummy McAwesomeSauce” for all I care. I’ll never do a shootout, and even if I could, it would be a pointless exercise. All I know is that it makes my mixes sound better in a subtle and pleasing way. And in the words of the Navin R. Johnson, “that’s all I need.”

Update: Nice (but short) article by Clive Thompson at Wired, who also argues the case for thinking outside the skeuomorphic box:

…skeuomorphs are hobbling innovation by lashing designers to metaphors of the past. Unless we start weaning ourselves off these defunct models, we will fail to produce digital tools that harness what computers do best.

  1. I have a hunch about why people often don’t ask these questions: In the case of tape, it’s because a high quality tape machine—or vintage console, or whatever—is pretty much a make-everything-sound-better machine. But if someone released a plain looking plug-in with one knob called “The Make Everything Sound Better Machine,” it would sell like…what’s the opposite of hotcakes? Many people just wouldn’t take it seriously. But make it look and act like a very heavy, expensive, and sentimental machine, and suddenly it starts sounding like gold. Yes, examples like Analog Channel and Crane Song Phoenix and Avid HEAT exist, and people use them to great effect. But often those tools aren’t necessarily seen as state of the art—it’s products like the A800 that grab all the attention and accolades.
  2. Just once, I would love to see a promo video that says, “You know that vintage Studer an army of Pro Tools monkeys couldn’t pry from your cold, dead hands? Well, it sounds okay. Our plugin does the same kinds of things to your sound that your machine does, but it does it way better. A little bit of juice, a little bit of crunch, a little bit of squash…you know, the good stuff. Whatever you do, just don’t call it ‘tape,’ okay?”