2017, Chapter 17:
Where We’re Going
Okay. This chapter’s a bit harder than the last two. After all, it’s pretty easy to recap What We Did Right and What We Did Wrong, but looking forward…weeeellllllll…take what I said about there being no perfect companies and multiply it by 1000. Not only is there no perfect crystal ball out there, but there’s no crystal ball, period.
I mean, I can say there are some things that seem very certain to happen next year…and they might
not happen. I can say that there are other things that seem like they won’t happen next year…and they
might. You never know.
Because no company can predict the future.
I can say, “Yeah, we’re absolutely gonna have a turntable in 2018,” and then do a face-plant because the castings are junk. Or the die-caster goes out of business. Or the carbon fiber for the arms becomes unsourcable. Or a dozen other things.
And that’s why I keep saying, “It happens when it happens.”
It’s actually a bit easier to predict things that won’t happen, because our product development, testing, and production ramp-up cycle is now long enough that if it hasn’t started now (or soon), it ain’t happening in 2018.
“Wow, sounds like it takes a lot longer to come up with new products,” some might say.
And yes, it does. It’s no longer 3 months from concept to shipping, like it was in the garage days. Because in the garage days, we didn’t have significant testing during the prototype stage. We didn’t send stuff out to early listeners and get feedback. It’s good that we have these things is place now, because it means less chance of a surprise when we start making thousands of them.
Aside: and, as I’ve said many times before, it’s not like we sit on our hands, so there are a lot of products that are way down the development road. So, it’s entirely possible that a concept we’ve never hinted at could be in production within 3 months.
“But, like, a year to develop something new?” you ask.
Yes. Or longer. It’s not the same rodeo as when we were in the garage. In addition to much more robust testing and qualification, including assessments of possible production variation, we now have lots of products that have both analog and digital components. And firmware…some products needing two different sets of firmware, too. The products are more complex, so they take longer to develop.
Hell, take Jotunheim. That hung around here for four years before it became real. Or the turntable, which we’ve been farting around with for 5 years. Both of those are a little extreme, but yes, we have products that have been in development longer than a year, and that will (probably) ship in 2018.
Again, though:
it happens when it happens.
But, all that said, it’s probably easiest to say what probably won’t happen, than try to predict what will. So let’s start the map to Where We’re Going with some “this ain’t us” kinda stuff.
Paths Untaken: Stuff That (Probably) Won’t Happen
I know I said some scary stuff about consolidation and canceling lines and less products in the last chapter. But the fact of the matter is that saying is one thing, and doing is another. I went into some of the reasons you won’t be seeing wholesale slaughter of lines and a new ultra-lean approach in 2018, but it bears repeating.
So, let’s start with that. And then, let’s go down the list:
- You won’t be seeing wholesale slaughter of products. Fact is, most everything is selling at least pretty well, and killing stuff that is selling pretty well is not a great idea, as long as the product is mature and reliable. Of course, this doesn’t mean that everything will continue selling pretty well, and it doesn’t mean that we won’t be looking to consolidate (where it makes sense) and update (where it makes sense.)
- You won’t suddenly see huge changes in our basic approach. We’re going to continue to be a company focused on both innovative digital (Multibit, Gadget, USB interface) and discrete analog technologies, for the desktop and 2-channel market. We’re not suddenly gonna pivot to selling delta-sigma op-amp portable products primarily. And yeah, I know, not everyone thinks that our digital stuff is all that innovative, and not everyone thinks we have the best approach. That’s cool. But when a company offers a cheaper, better-measuring alternative (like Modi Uber vs Modi Multibit) for people who want to save money and have better measurements, and tells people straight up they can save $100 by choosing it, or save $249 by using the DAC in their phone, you know, well, that ain’t much to get outraged over. Sure, you can go on arguing how we’re trying to “trick” people into choosing multibit in your best Captain of the Debate Team manner, but remember, the Captain of the Debate team is now a lawyer or diplomat or someone getting paid $$$, while you’re arguing with random dudes on the internet. Sorry. Just had to get that off my chest. Bottom line, if you like what we’re doing, cool. If you don’t, move on. You’ll live longer.
- You won’t be seeing software-based stuff like streamers or DAPs. Yeah. Sorry. Not us. Allergic to that stuff. It’s a Chinese finger-puzzle. Stick your finger into the software stew and eventually it’ll eat you. You’ll become a software company. We don’t want to be a software company. We don’t want the support and update requirements. Sorry if you really want us to do something like that, but there are plenty of other options out there. And if they don’t work, or have some weird software glitch, you can contact the manufacturer of the streamer or DAP, rather than us.
- You won’t be seeing us spending tons of money on fancy packaging, offices, or anything else that isn’t the product. Yes. Our focus on putting the highest value in the product will continue. Your Schiit will not arrive in a box that cost more than the product inside. Nor will it come in a chassis that cost more than the components. Nor will we be building a spaceship-like campus to house our tens of thousands of employees in the near future (that’s a joke, guys. Laugh.)
- You won’t be seeing widespread product proliferation on the 2-channel side to match the headphone side. As someone said, “When I look at your 2-channel lineup, it makes sense. Two preamps. One power amp and one integrated amp. Three DACs. When I look at the headphone stuff…whooooooo, what the heck happened?” Time. Time happened. Time, and organic growth. We’ve learned some things. And that’ll be why you won’t be seeing a Vidar Uber and a Thor and two more preamps that take DAC cards and all sorts of crazy stuff. Sure, there may be a place in the line for another amp (eventually—what about a 25W class A amp in a Vidar chassis that does 200W into 4 ohms mono, just throwing that out there…), and sure, one of the products may need an update, but we’re not going crazy on products here.
So yeah. No huge surprises here. Or at least there shouldn’t be. That is, of course, unless the year itself is surprising. When the rules change, you need to change.
Or you can change the rules yourself.
Signposts: Analog IP Edition
Now that we’ve Kevorked some of the speculation about what we might do, let’s zoom out. Let’s go big-picture and look at some of the behind-the-scenes stuff that will determine our future…both in 2018 and beyond.
I’m talking analog IP. Or, to spell out the fancy terms,
intellectual property. As in, unique stuff that we do that gives us an edge. Or at least stuff we think that gives us an edge.
“Wait a sec,” some of the more engineering-minded are saying. “Analog IP? What can you do that’s really unique? Most analog stuff was hammered out a long, long time ago.”
Yes. It was. But that doesn’t mean there aren’t unique ways of doing things, even on the analog side. Remember, the basis of Mike’s “megacomboburrito” digital filter was a 1917 paper. And one of the things I got to do in 2017 was to spend a lot longer on basic research. Which means we’re now developing an analog IP portfolio that will figure increasingly in our products in the future.
To understand why this is important, we need to zoom out again. Let’s look at the analog technologies we’ve implemented since inception, and compare them to now, and where we’re going.
When we started Schiit, we were using fairly tried-and-true approaches to analog design. Class A amps with current sources, triode amps with White cathode follower output stages, Circlotron-style topology in the balanced products, and Lin-style power amp topology in Magni.
Now, there’s nothing wrong with this. Circlotrons, in particular, are an interesting approach to solving the problem of mismatched output devices (since they use only N-channel, or only P-channel, devices). And we did employ an interesting sliding-bias “Dynamically Adaptive” stage in Lyr, which is kinda like a White cathode follower on steroids. Ish.
Today, we’re deploying new ideas. Ideas like Jotunheim’s Pivot Point topology, which is an interesting (and, as far as I know, unique) current-feedback approach to offering balanced and single-ended in, and balanced and single-ended out, from a single gain stage. (Circlotrons have to use summers to get SE output.) And, in Magni 3 and Loki, we have a new current-feedback single-ended stage that is simple, elegant, and measures well.
Also, I kinda snuck in a new type of topology with Vali 2: a combobulated tube-transistor gain stage that doesn’t need interstage coupling caps to span different rails, and also which operates in current mode, allowing for higher plate voltage from a given power supply. Again, I have never seen this used elsewhere. I never named it, but it does an elegant job of combining the two devices. Maybe Coherence™ or something like that?
Seems like we’re going down a current-feedback path? Yes, we are. In my development, it seems to provide the best combination of simplicity and performance. And, I think it sounds good.
Aside: And yes, I know, some people want “real no-feedback,” which has lots of gotchas in terms of output impedance and is not applicable everywhere I’d like to use it.
In the future, the new analog IP will kick into high gear. Yes, we’ve already started deploying some new ideas, but that’s just the start. Because a whole lot of what I did last year was to step back and do a lot of “what if” style research. I had time to ask, “What are some of the biggest problems we want to solve,” and work on answering those questions. And, in the last couple of months alone, we’ve had some big breakthroughs. Or at least I think we have. I could be fooling myself.
Because of this, 2018 will see the introduction of two or three new analog technologies, including some that you may never have seen before. The entire foundation of our analog approach will start to shift towards these new technologies (where appropriate, and, of course, based on what you guys think of them.)
What kind of new analog IP am I talking about? One of these new approaches solves a problem common to all Class AB amplifiers. One of them allows us to “program” output stages for square-law operation—or, in other words, make things sound more like tubes. One of them is the “holy grail” true differential I/O stage that I’ve been searching for, for a long, long time.
Now, I know this might sound grandiose, or even a little crazy (especially the “sounds more like tubes” part, where I’m careful to use the modifier “more like” rather than “just like” and where I totally understand that some people may not want things that sound like tubes, but we also have to disambiguate people’s ideas of what tubes sound like, because most of the tube stuff we do isn’t the soft, syrupy, overly colored presentation that some people associate with tubes.
But, when you do a prototype (one of many different prototypes of an upcoming product, we do a lot of a/b/c/d stuff these days) that you listen to, and you go, “Huh, that sounds kinda tubey,” and you give it to Mike and he says, “Huh, that sounds kinda tubey,” and you give it to one of our trusted listeners and he says, “Wow, that sounds kinda tubey,” and when you finally think to measure it, and you go, “Holy schiit, that distortion profile looks exactly like it’s coming off a square-law device,” you know, maybe there’s something there. And what I found, in testing another kind of analog tech, a way to tweak an output stage to sound more like a tube—while running cool and not using any kind of exotic parts. (Hell, the output stage of this very tubey prototype runs 8mA bias—damn near off.) Whether or not that’s a good thing is entirely up to you, of course.
So when are all this new analog IP going to show up? Patience. Again, we’re not rushing into this. And it’ll happen when it happens.
Wayfinding: Digital IP Edition
Mike’s got a long history of changing the rules in digital. And he’s going even farther down that path with Gadget technology. He’s also got a few other surprises up his sleeve, including a new UAC2 USB input that’s all our own (and will start on the lower-cost side of the equation, as it has benefits in terms of cost and size.)
All of this stuff has been mentioned, as has the fact that we’re working on new platforms to power even more processing-intense versions of the Gadget. So, I bet at least some of you are asking, “So when are we gonna see this stuff?”
Well, I’m not going to speculate. It’s best to ask Mike about matters digital. But I can say that he, Ivana, and Dave are actively developing some very exciting digital IP that will factor into what we’re doing in 2018.
And, at the same time, Dave is helping me extend the domain of our “intelligent” amplifier management, as seen in Vidar and Ragnarok. As I mentioned before, Vidar took a step back from Ragnarok’s full management in order to enable more standard testing. But there’s more than one way to do full management, and I think we now have a failsafe management technique that allows for some cool stuff (literally) like output stage shutdown to save power. There’s a lot more development and testing needed for this, though, so don’t be surprised if nothing happens with this until beyond 2018.
Direct Paths
“So, after all this blather, what can you guarantee we’ll see in 2018?” someone asks. Okay. Fine. Leaving aside major disasters, I can give you a couple of certainties. Or near-certainties. Things like:
- Sol. Barring any major surprises, you’ll see a turntable from us in 2018. We’re in the final process of tweaking the casting and ordering parts. We’re working hard to keep the price very low, especially when you consider features like the 12” carbon-fiber arm, all cast-aluminum construction, true unipivot design, completely separate motor pod, and real-time VTA adjustment (something we added recently.) I’m not going to pick a quarter, but we’re shooting for the first half of the year. I’m also not going to speculate on how well we’ll be able to stay in stock. Depending on demand, this may be a crazy, crazy product to keep up with. Or we’ll have lots of Christmas presents. Who knows.
- The Gadget. Also barring any major platform hiccups, we’ll have a sellable Gadget in 2018, and we’ll have one that performs even better than the one we had at RMAF. It may include additional features. But it’s pretty certain that we’ll have something to sell, in some form, in 2018. Beyond that, we’ll see how it goes.
Want more? I can’t give you direct product names and introduction dates, but I can absolutely say that we’re going to continue to improve both our internal and external processes. This is going to include:
- Additional investment in R&D. It’s time we bring in some tools, including a benchtop CNC and a 3D printer, so we can accelerate product development, and communicate better with our vendors.
- Additional process streamlining. It’s time we invest in some tooled parts to eliminate some of the painful hand-work that we still have to do with some of our products, especially the more complex ones. We’ll also be investing in custom parts to accomplish much of the same thing, and moving more products to more efficient manufacturing methods. We've also bought a whole bunch of Avermetrics analyzers for production-level testing.
- Additional work on guides. We need to help people decide what to get…and to use it effectively once they get it. Our text-based guides are a start, but we can do much more, and we will.
Aaaanndddd…beyond that, I can’t say very much. Will there be product replacements and updates? Probably. Will there be additional products, like that Class A amp I teased you with? Maybe. But maybe not. Like I said there’s no perfect crystal ball.
Alluring Detours
Now, of course, there are plenty of things we
want to do, in addition to what we
have to do. Number one on the list is opening one or more additional Schiitrs, based on the performance of the first one.
Let’s face it. The first Schiitr was a lark. It was a “hell with it, see what we can do,” kind of thing. The fact that it did so well, in such an inconvenient location, is a signpost to the future. So yes, we’ll be actively looking at opening at least one more Schiitr, and perhaps more, as the performance is proven.
Where? I don’t know. Alone or in cooperation with someone else? I don’t know. Ask Tyler. He’s gonna be the mastermind of this one.
Beyond that, holy crap, there are tons of detours we’d love to take. It’d be great to have a “full” speaker power amp lineup, with a giant powerful behemoth and maybe even bigger monoblocks added. But that’s a little grandiose. And premature. And maybe not us at all. It’d be great to have a whole line of Lokis, from tiny to gigantic. It would be great to have a whole line of Gadgets.
But all of these alluring detours have one thing in common: the “it’d be great” part of the idea ignores the reality: If we had a ton of speaker power amps, we’d have tons of confusion. Plus shipping big amps is a real bear (hell, I knew one company that had to go to crating—as in, wood crating—for their larger power amps). If we had a ton of Lokis, then there’d be confusion about which one to buy. And people asking for different types of EQ that we don’t even offer. And if we had a ton of Gadgets, again, the first question would be “Which one?” And for all of those products, “it’d be great” gets us what, 8-10 more products to deal with? Meaning 8-10 more things to try to keep in stock? Which makes operations even more challenging, and our out-of-stock situations potentially even worse.
Yeah. It’d be great.
Until you have to live it.
So excuse us if we choose a limited number of fairly straightforward paths next year. Hell, the turntable alone is crazy enough. Isn’t it?
We’ll see. Here’s looking forward to a great 2018!
