2015 Chapter 11:
Food Scientists vs. Michelin Stars
On the advertising agency side of things, we have the “parable of the ice cream man” that we use to illustrate the old adage of “if all you have is a hammer, everything looks like a nail.”
Huh? Ice cream men? Food scientists? Hammers?
What the heck is Jason rambling on about now, you ask. Is he hungry? Demented? Just too damn hot in the unusual muggy summer heat that makes wearing headphones and writing a misery?
Maybe a bit of all of the above. But bear with me.
First, let’s talk about the advertising agency biz. Ad agencies rank down there with used car salesmen and Kirby door-to-door shills in many people’s eyes, despite some lingering
Mad Men-inspired-but-really-never-happened fun-times memes. And yeah, I understand. When many agencies make a living by charging ridiculous multiples for every intern’s time and have big headcounts that do nothing but justify their (oops, surprisingly) large bills, the reputation is deserved.
But one thing a (good) agency does well is
solve marketing problems, rather than
sell a specific service.
For example, there are any number of marketing specialists like “social media consultants” or “native content advertising specialists” or “pay-per-click advertising companies” out there. And they’re defined by their single specific slice of the marketing pie…social media or native advertising or pay-per-click or whatever. That’s what they do. And if you, as a company, invite them in the door, you know exactly what they’re going to try to sell you…either social media or native advertising or PPC.
And that’s how they’re like the ice cream man. The ice cream man drives around and sells ice cream. On a hot day, he’s a welcome relief. The perfect solution (well, that is if you aren’t paleo, or on a ketosis diet, or, well, hell, food is weird these days). But on a cold day, who wants it? Unfortunately, ice cream is what he has to sell, even on the cold days, the rainy days, and the times you really have exactly zero interest in ice cream.
Just like those marketing specialists. If you want (and need) what they’re selling, you’re in luck. If you don’t want or need it, you may end up buying something that does nothing for you.
Compare this to a full-service advertising agency or marketing agency. A (good) agency will sit down and listen to what you want to do, then offer ways to accomplish your goals. These ways may not be what you expected. They may not even be what you asked for. But a (good) agency will be able to pull from a variety of strategies and tactics that it has experience with, and recommend some that offer the highest probability of success.
So, you could say they sell ice cream. And tacos. And organic nuts. And fresh fruit. And lots of other things. But they’re not just selling ice cream. They’re selling what you need, rather than only what they have.
And that’s the parable of the ice cream man. I’ve used it tons of times…when a company comes asking for a social media program when they really need direct mail (yes, still relevant—depends on the audience) or when they want to change a perfectly good website that really just needs more traffic.
So what does this have to do with food scientists vs the cordon bleu? Read on.
The Insult That Backfired
Early on, soon after Schiit launched, the objectivist movement sniffed us out and began throwing various insults our way. “They must not use any test equipment, they just talk about how things sound.” “Those designs are huge, inefficient, and overpriced…probably unreliable, too.”
And my personal favorite, which went something like, “How can a tiny audiophile company imagine it could do better than the latest TI op-amp, they have a worldwide staff of scientists and engineers working on those designs?”
When I first read that, I was taken aback. Because, yep, it sounds completely insane. One or two engineers at a tiny company that started in a garage with miniscule resources, pitted against the combined might of, say, TI or Analog Devices? Most people would bet on the big guys.
But then I started thinking about it.
And I realized:
Frito-Lay probably has more food scientists working on the newest formula for Doritos than there are Michelin-starred chefs in the world…but which would you rather eat?
And…
Darden Restaurants and Taco Bell, Inc and McDonald’s probably have tons of food scientists stapled to their beancounters…but wouldn’t you rather have something fresh from a Cordon Bleu chef’s kitchen?
Both of those are perfect examples of where
single individuals trump
teams of scientists armed with the latest technology all the time.
Why?
Two reasons, I think:
- They actually give a crap about raising the bar on how something tastes (and have the knowledge to at least point in the direction of a solution.)
- They don’t have to worry about packagability, shelf life, merchantability, etc. (and yeah, I know, this is a big one.)
So, happy with turning the insult around (though I never said anything about it), I let it go.
Until today, when I realized this is a great way to describe what we’re doing with digital audio.
Digital, Michelin Stars, and Flavor Technicians
I’ve gone over this before, so I won’t go into insane depth again, but here’s the gist of what we’re trying to do with multibit digital audio playback:
preserve the original samples of the vast majority of music out there, in order to offer the best possible playback fidelity.
And yeah, I know, some people think we’re full of Schiit when we start talking about closed-form digital filters and how most DACs simply replace everything with a reduced-bit-depth approximation of the original. That’s fine. I’ll let them argue with Baldr (Mike Moffat) about that one.
But here it is: we
do have a stated purpose. And that stated purpose is to
make the most out of the ingredients we have.
Kinda like a top chef, huh?
(Or, well, maybe at least a good one. Throw me a bone here.)
Contrast this to the approach taken by delta-sigma DACs. The vast majority of music in this world (about 99.99%--do the math on the tracks available yourself) goes in at 16/44 and gets transformed into a multi-order noise-shaped 2- to 5-bit approximation of the original. Yes, even in some very exotic designs.
Which, if you think about it, isn’t that a lot like the food scientists at Frito-Lay and Taco Bell trying to preserve some shade of the original ingredients through latest-tech additives, preservatives, flavor enhancers and modifiers?
“So you’re saying that everyone else is wrong?” you ask, arms crossed.
Not at all.
What matters in food is in the tasting. And there are some true food artists working in the molecular gastronomy space. And then there’s personal preference, too. Sometimes it’s time for a bag of Doritos. And sometimes it’s time for yellowtail sashimi with lime and burdock.
If someone achieves the perfect result for your ears, don’t worry about this analogy. Sit back, listen, and enjoy. Hell, we make delta-sigma DACs too, and they sound plenty fine. And, especially in the case of $99-149 products, there won’t ever be a multibit option at that price range.
But, as Mike has said, we’re gonna be bringing multibit technology to our upgradable products…and then you have a choice. Stay delta-sigma, or go multibit. It’s entirely up to you.
More than DACs
Almost every day, someone asks why don’t we do a Bluetooth audio interface, or a Class D amp, or digital room correction, or any other number of nifty audio-scientist-style tricks that are as easy as picking up a chipset and (in some cases) doing some RF certifications.
Aside: which goes to show how simple and ubiquitous these massive technological solutions are. If you ride the teams-of-audio-scientists train, you do get quite an amazing toolkit. Easy wireless transmission of compressed audio, $5 100W amp chips, and automatic room correction for the price of a chipset, a license, and a microphone.
But every one of these solutions runs counter to our goals. Bluetooth audio is lossy and variable in terms of quality, depending on your source and how far away you are from the receiver. It’s taking those original samples, applying perceptual encoding-based compression, and decompressing on the other side to an approximation of the original. Yes, it can sound good, but it’s not us. Class D amps? Same thing. They take the analog waveform (or, in some cases, direct digital input) and use noise-shaping and filtering to approximate it at the output. Digital room correction? Again, everything’s being run through a number of very complex mathematical transformations—in this case, where the actual goal is to change the output in order to compensate for the room. Used correctly, this is impressive technology…but, like stability control on a car, sometimes it’s hard to tell where it actually goes off the rails until something bad happens.
“So you guys are grumpy old buttheads, forcing the hard way on everyone?”
No. Not at all. We’re explaining what we do, and why we do it. If what we’re doing sounds good to you, then there you go. If not, not. Depending on how many people think we’re crazy, we’ll either continue to succeed, or die on the vine.
But…there
is a method to our madness. There are
reasons for what we do, besides "let's make a product that hits the current hype du jour and makes big profit."
If this parable helps you understand what we’re shooting for in digital, then it’s done its job. Beyond that, it’s up to you to decide what your own personal preference is.
The Real Moral: Here’s to the Misfits
Okay, I’ve explained what we’re trying to do. But it’s important to remember that there are plenty of other companies out there that are pushing the limits of what we can do in audio…and that don’t have globe-spanning teams of scientists and engineers.
So here’s to everyone who tilt at the monoliths. Who don’t think a homogeneous, Lightning-cable-connected, DSP-corrected-to-the-eardrums solution based on big science and lowest-common-denominator parts assembled in an inexpensive corner of the globe is the One True Future Path. Who work with parts long obsolete…or create new technologies that elude even the giant engineering teams (remember, TI licensed Nelson Pass’ patent, not the other way around.) Who explore improbable ways to do things…and sometimes succeed. Who ignore the datasheets and reference designs and tweak it until they have something better. Who listen, and measure, and come back to do it again and again.
Because, otherwise, all we have is Doritos. And that’s a very, very sad place to be.
