Schiit Happened: The Story of the World's Most Improbable Start-Up
Dec 22, 2015 at 5:03 PM Post #9,241 of 170,003
  Some words from Mike on Audiostream: http://www.audiostream.com/content/qa-mike-moffat-schiit-audio
 
I'll have the closing-of-the-year chapter here in a couple of days.

This includes Mike's elaboration on my explanation of why multibit+oversampling is likely to sound better than NOS (non-oversampling):
 
" All of our converters at $600 and above are multibit and multirate. By multirate, we mean oversampling. I do not believe in making non oversampling (NOS) converters. The required NOS analog “brick wall” filter has multiple poles and is very prone to ringing and horrendous phase shift."
 
Dec 22, 2015 at 6:13 PM Post #9,243 of 170,003
  Curious friend says "why do you need a separate digital analog converter"
schitt head says " because the ones in computers sound like crap"
Curious friend says " I think my computer sounds fine"
schitt head says " I'm happy for you but in the end of days your laptop will be just that, a laptop. My Yggy can be transformed into control systems for fighter jets and submarines and one badass zombie tracking and elimination device"

time to get rid of curious friend! 
 
Dec 22, 2015 at 6:33 PM Post #9,245 of 170,003
  Curious friend says "why do you need a separate digital analog converter"
schitt head says " because the ones in computers sound like crap"
Curious friend says " I think my computer sounds fine"
schitt head says " I'm happy for you but in the end of days your laptop will be just that, a laptop. My Yggy can be transformed into control systems for fighter jets and submarines and one badass zombie tracking and elimination device"


Gosh, that could have been Matt, Jodi, or Daniel. All listen to headphones at work. I mean, maybe Clay or Nathan, doubtful, Clay has already expressed that his hearing is too damaged to let him care. hmmm, who could that be? At least it wasn't me!
 
Dec 22, 2015 at 8:20 PM Post #9,248 of 170,003
I wonder if Mike has given any thought of ISIS getting a hold of his Manhattan Project technology?
rolleyes.gif

they (ISIS) toooooo stupid and would not understand Schiit 
deadhorse.gif

 
Dec 22, 2015 at 8:39 PM Post #9,249 of 170,003
The new gadget is a SYS with left-right channels' balance control knob for those with difficult speaker placements or active speakers. I suggested it a while ago, and it will be known as SYS IVER.

Just hoping, and just saying.
 
Dec 22, 2015 at 8:57 PM Post #9,250 of 170,003
The new gadget is a SYS with left-right channels' balance control knob for those with difficult speaker placements or active speakers. I suggested it a while ago, and it will be known as SYS IVER.

Just hoping, and just saying.


Mike's quote in the interview said the product was entirely in the digital realm
 
Dec 23, 2015 at 12:29 AM Post #9,253 of 170,003
  Mr. tonykaz!
 
You keep forgetting to push the typewriter buttons before clicking the post button.
 
I'm totally curious to know what you forgot to type!
 
wink_face.gif

Or is it possible that Mr tonykaz is a beta tester of the new Schiit product and his previous 2 posts are the side effect... 
biggrin.gif
tongue.gif

 
Dec 23, 2015 at 10:31 AM Post #9,255 of 170,003
2015 Chapter 20:
Looking Forward to 2016
 
So it’s the end of another year. A good time to look back at what went right, what went wrong…and what might be coming next year.
 


Aside: And yes, I know, Mike has been teasing you about The Gadget, but that’s really early dev stuff right now—think Theta Frankenstein stage, if you remember back to the earliest days of high-end digital. I can’t say much more about that, other than to reiterate it is like nothing else ever done in digital…and that I’m sure it’ll generate its fair bit of controversy when it arrives (is it right, is it wrong, is it fair, is it really “better,” etc.)

 

But I’m getting ahead of myself, with an aside only two sentences in to the year’s recap. So let’s turn around and look at what we did…and what was good and bad with that.
 
Most obviously, we introduced several new or updated products. Much less than last year, though, as I promised. Mike and I met after-the-fact of 2014 and asked ourselves, “Why did we do so many products? Why so many updates?” Hell, there seemed to be products we updated for no good reason, when we looked back on them. And, like I said last year, it was just too much. It strained our capacity for growth, and caused any number of production glitches, hitches, delays, and challenges.
 
Or, in numbers, last year we had 12 product intros. This year, we had 6:
 
  • Yggdrasil (final-friggin-ly!)
  • Gungnir Multibit
  • Bifrost Multibit
  • Mjolnir 2
  • Vali 2
  • LISST
 
One new DAC, two significant DAC updates, two significant amp updates, and a new idea: solid-state tubes. Not too bad, pacing-wise. If we averaged that, it would be a new or updated product every two months. Still maybe a little faster than we’d like, but is LISST really a product unto itself (it seems to be—they are selling like crazy) and are updates really new products? The answer to both is “well, probably not.” LISST are built for us start to finish by our production partner in Simi Valley. Updates—especially DAC updates—don’t change the chassis significantly, so a big part of the “oh craps” that we run into in production are eliminated.
 
So, if you toss the updates and the LISST the only new-new product was Yggdrasil. By that metric, we had only one product introduction. Positively lazy!
 
Except…maybe not so much.
 
 
Surprise the First: Multibit Profusion
 
As I mentioned before, Mike handing me the Gungnir Multibit before the Yggy was actually shipping was a huge surprise. I’d figured we’d be able to do something with Gungnir…you know, sometime far in the future. Instead, we were staring at the ability to do something with it, like, right now.
 
But should we? We actually talked it through. Like most of our talks about product strategy, it went something like this:
 
“You know, most companies would milk the Yggy for at least a year before introducing a downmarket variant,” I told Mike.
 
“Yeah, but we aren’t most companies,” he told me.
 
And that was that.
 
Gungnir Multibit followed Yggy by only about 3 months. Bifrost Multibit followed that by 5 months. While the world was still struggling to digest the bizarre, non-DSD, 21-bit Yggdrasil (see next section below), we’d sucker-punched them with an entire line of multibit DACs, starting at about ¼ the Yggy’s price.
 
Why is this significant? Two reasons:
 
  • As mentioned, we didn’t drag this out two years, spacing the products at “audiophiley appropriate” times. Instead, we said, “Hey, let’s get the maximum multibit benefit out there at all feasible price points, as rapidly as practical.” Suddenly, a whole lot of stuff out there starts looking overpriced.
  • All the multibit platforms we use are entirely new, and not based on moribund or NOS DAC chips. This means we aren’t limited on how many we can make, and it means we’re bringing entirely new things to the market. Suddenly, off-the-shelf audio DACs don’t seem so interesting.
 
And well, actually, let’s add a third reason: despite the huge R&D effort needed for Yggdrasil, we didn’t start with a hogged-outta-solid-aluminum, gold-plated monstrosity that cost as much as a car…even though we certainly could have. That set the bar for all of our multibit DACs, on down the chain.
 
No, wait, let’s add a fourth: because our DACs are upgradable, 2/3 of the multibit DAC line is available as upgrades to people who already own our DACs. This is a huge win for everyone—saving even more cost.
 
So yeah, DAC-wise, the end of 2015 is dramatically different than the end of 2014. In 2014, we had zero multibit DACs. Now we have three.
 
Holy moly, what a change.
 
 
Surprise the Second: Multibit Reception
 
When we introduced Yggdrasil, I wondered what the world, currently drunk on dreams of 32/384 and 8X DSD, would think about this bizarre, 21-bit, no-DSD DAC. I knew what it sounded like, and I knew that 99.9% of digital music is 16/44, but, you know, what would everyone think?
 
Turns out I shouldn’t have worried. The biggest problem was actually keeping up with demand. We broke the world supply chain for AD5791s (or at least it seemed like we did). We had to move up runs. We had to plead with metal suppliers to deliver early. We had to set up an entirely new, separate line to keep making Yggys.
 
What did people think about it being 21 bits? Except for a very few questions, nobody seemed to care or notice. All they cared about was the sound.
 
How about the “crippling” lack of DSD (as some opined)? A non-issue. Even the most ardent DSD fans are beginning to realize that the Sony vaults are not opening wide, and the format will remain a tiny niche.
 


Aside: yes, for some who have large investments in SACDs that they then have ripped, maybe multibit is not ideal. We get that. It’s cool. We can’t please everyone.

 

And beyond popular response (which is really what matters), what happened? Amazing stuff:
 
  1. A blockbuster review of Yggy in Computer Audiophile
  2. Product of the Year from Computer Audiophile
  3. Product of the Year from Headphone.Guru
  4. Gungnir Multibit in the Rolling Stone Holiday Gift Guide
  5. A big interview of Mike Moffat at Audiostream
 
So yeah. Reception. I think it’s been fine.
 
 
Surprise the Third: Big Changes to Analog
 
Another surprise, for those who have been watching our prior “second gen” products, is how much the analog side of things changed. Both Mjolnir 2 and Vali 2 are significantly different than their predecessors—far more than a few new features, tweaked cosmetics, or fractionally higher power output.
 
Mjolnir 2 went from a solid-state amp to a tube amp. It also became the first amp you could choose to run solid-state or with tubes. It was literally nothing like its predecessor, save for one transformer, and the use of a circlotron topology.
 
And, introduced with Mjolnir 2, LISST also allowed Lyr 2 and Lyr owners to switch to solid-state if they felt like it. It’s a small product, but it allows us to offer something that nobody else does—the ability to run several different amps as either tube hybrids or all solid-state.
 
And let’s not forget that Mjolnir 2 added something I bet you thought you’d never see: front-mounted switches. (Yeah, I’m being a bit jocular here, so shoot me.)
 
Vali 2 saw similar huge changes—literally, the output device size and the overall size of the chassis are almost literally the only things unchanged. It’s a monumentally more flexible amp than the original Vali. Heck, it even allows tube rolling. Hell, it even takes LISST!
 
The point is: these are huge changes, not a quick feature-add or incremental update. Expect to see this trend set the tone for future next-gen products—or perhaps the next-gen products will be even more radically changed.
 
Bottom line, on the analog side, we’re not standing still, either.
 
 
Right and Wrong in 2015
 
2015 was boring. And boring years are very, very good years. Excitement, in manufacturing, is rarely excitement in the positive sense. Excitement usually means things are breaking, stuff is late, someone messed up, and you are boned.
 
Boring is what happens when most things go right.
 
So let’s start with right:
 
In-stock intros. A couple of years ago, you would have fallen out of your chair if we introduced a new product and it was orderable. In stock? Inconceivable! And yet, this year, 5/6 of the products we introduced were in stock when we announced them. As in, we announce, you order, you get shipping notice 15 minutes later. You know, like a real company. Maybe even better than a real company (I’m looking at you, Apple…come on, “Available November?” Sheesh.
 


Aside: and yes, I know, we were 15000 years late on Yggdrasil, but hey, the 5 products that followed it were all in stock on intro. And not for microseconds.

 

Significantly less problems with new products. No, we were not perfect, but there were no embarrassing “total recall” moments like we had with Asgard 2 a few years ago and Fulla last year. Other than a few transformers coming off boards on early Yggdrasils (which could not have been foreseen, since they were the same size and mounting as we’ve used in literally tens of thousands of other products—and quickly addressed in production, and everyone affected getting free return/exchanges) and a very bizarre glitch with the Bifrost Multibit in a small percentage of systems (so small, we had to bring back some units because we couldn’t replicate the problem—and we cannot get the returns to replicate it all the time, meaning it’s system-dependent), the products are solid.
 
Better stocking. Although we’re still not perfect, we’re much better at keeping products in stock. Yes, we were caught out a bit by Mani demand, and we’ve had some struggles keeping up with the multibit DACs, but long, grueling out-of-stock statuses are significantly reduced.
 
Lower cost parts. Thanks to Alex bringing on new distribution partners—ironically enough, found when we were breaking the world supply chain for AD5791s—we’ve been able to reduce costs, while using the same high-quality parts.
 
Higher quality parts. A new metal vendor, and a couple of intense meetings with the current one, and our chassis are now looking better than ever—even the inexpensive chassis like Vali 2.
 
Increased operational efficiency. A new hire, Tyler, has been putting into place the systems we need to operate as a real—er, I mean, much larger, company. At the same time, Alex has been pulling out the stops getting everyone cross-trained and finding what they are best at doing. As a result, we cruised through this holiday season without having to add staff.
 
What about wrong?
 
Sure, we do lots of wrong things. Every company does. This year, it wasn’t too bad, though. Most of it was functional stuff:
 
Overbuying. We went a little crazy at the end of summer when it came time to get stocked up for the end of the year. As a result, our overall parts cost went through the roof. It looks like it’s a gamble that has largely paid off, this holiday season, but it just as easily could have been very bad. As an entirely self-funded company with no bank lines or receivables financing, we are extraordinarily free of pressure to perform at some arbitrary metric, but we also need to keep an eye on cash flow.
 
Bad inventory visibility. With so many parts spread between us and our production partners (the assembly house in Simi Valley and vendors in the San Fernando Valley), keeping up with what we got and what we need started outstripping the ad-hoc systems we were using. It also contributed to the overbuying spree. As a result, we’ve been putting in a real inventory management system, and recently started using it. Yeah, I know, basic stuff.
 
More metal problems. What’s a year without screwy metal? Well, so far, we wouldn’t know, because every year has had screwy metal. This time, it was a new vendor literally making thousands of bad parts due to no oversight (at their firm). They got to eat it…but we had to rush through several replacement runs. The good news is that we are working much more closely with the vendors now, and I believe they understand, more than ever, what we need.
 
But this is nit-picking. It was a great, great year on virtually all fronts.
 
 
What About 2016?
 
Yeah, yeah. I know, 2015 is old news. So what will you see from us in 2016…that’s what you really want to know, right?
 
And that’s tough. I can’t exactly outline our product plan in detail. Because, well, you know, that’s talking before you actually have something to sell, which we’re trying very hard not to do.
 
And, frankly, because we’re not 100% sure of what we’ll introduce.
 
So let’s go broad-brush, and let me see what I can talk to you about.
 
Advertising. Yeah, big deal. But it’s a fact. You will be seeing more of us—in Stereophile, at Computer Audiophile, and maybe a few other places. Not a lot, nothing over the top, but it’s time for the 2-channel world to know a bit more about the Multibit DACS. Also because….
 
2-channel. I’m certain you’ll also see the first dedicated 2-channel products from us in 2016. As in preamps and amps. How many products, and when remain to be seen. But you will see some things from us. I hope. And they’ll be very, very cool. I think. At least in part because…
 
New topology. It’s not often you see radically different stuff on the analog side of things, but I think we’re ready to unveil something I’ve been working on for a couple of years…a new, inherently balanced, current-feedback topology that is not a circlotron, and is not supersymmetry. This is by far the best-performing topology we have ever come up with, both in terms of measurements and (in our opinion) sonics. This topology can stretch from the 2-channel world to the headphone world pretty easily.
 
Manhattan Project. I’ve said all I can about that, but I do believe it’s a 2016 product. However, it might not be.
 
Another secret. We have another shocker for 2016, which I can say even less about…but I can confirm it is NOT digital.
 
Improved distribution. We're hoping to extend our "direct partner" model--as with Electromod--next year. Also, I hope to have some 100V product for Japan, as we have some resources to work natively there. These are grand experiments...we'll see how they do.
 
More book. I’ll keep writing, if you’ll keep reading. I may be doing some fiction again as well.
 
Portable? Still not real excited on this one. I know it’s in vogue, and a lot of people want them, but to do it right, it will be huge, hot, and heavy. Wow, that sounded weird. I’m still trying to convince myself we can do something different and meaningful.
 
And, of course, we reserve the right to come up with a surprise or two…like we did with Vali 2 at the end of the year. And, after all, the Manhattan Project wasn’t even a glimmer in Mike’s eye at this time last year—it was totally out of left field this year.
 
Thank you again for all the support, comments, and suggestions! We’re listening…and we’ll continue to try making the most fun, high-value stuff out there!
 
Schiit Audio Stay updated on Schiit Audio at their sponsor profile on Head-Fi.
 
https://www.facebook.com/Schiit/ http://www.schiit.com/

Users who are viewing this thread

Back
Top