Chapter 12:
Schiit Goes Evil?
No story about Schiit would be complete without talking about NwAvGuy and the infamous Asgard Incident. Although the latter sounds like a neat title for a modern dark fantasy/conspiracy movie, it wasn’t a joking matter at the time.
If you want to look at this from a business perspective, this is about how you handle adversity. I’ve said it before, but it bears repeating: even if a Bernanke helicopter drops $100 million in your lap with a note saying, “startup capital for your company,” business is a series of challenges. Some of these are gonna be company-changing. This one was.
But first, the scorecard: we’re about 13 months into Schiit at this time. July of 2011. We still have the same three products:
- Asgard
- Valhalla
- Lyr
- And a prototype of the Bifrost (no chassis)
In terms of human years, we’re still a baby company. Still working in the garage, still putting everything back in to make more products and develop new ones, still with the 1.2 person crew (if you count 20% of my time, 30% of Eddie’s time, and about 70% of Rina’s time.)
So, we’re cruising along, with Eddie building stuff, me testing, and Rina shipping, and Mike and I working on tweaking Bifrost so we could get it shipping (originally planned for July, but hey, you know, we’re late on nearly everything), when we get an email.
This email has a video pasted into it, showing an AKG K701 driver flexing. With the email, there’s a short message, asking if this is normal behavior when you turn an Asgard on and off.
Also, although I didn’t know it at the time, the emailer also posted the same video on Head-Fi, asking “what causes this?” The whole thread is still available for your browsing:
http://www.head-fi.org/t/562736/what-causes-this-amp-related
It’s an insane day, so I send off a quick response:
Yep, it’s normal. Thinking at the time,
Yep, of course it’s normal, many amps have transients when you turn them on and off.
Anyway, back to the email…
The buyer wasn’t thrilled by my response, so I offered to let the buyer return the amp for a refund, even though the sale was of a B-stock product, which we don’t normally offer a 15-day return on.
Eventually, he took me up on the offer—but not before the Schiit hit the fan on Head-Fi.
Enter NwAvGuy
On the weekend after our buyer posted the video on Head-Fi, NwAvGuy appeared, pronouncing Schiit all a bunch of knuckledragging morons building dangerous amplifiers. Of course, I’m being a bit hyperbolic, but here’s the actual first paragraph:
“My professional engineering opinion, having done both amplifier and speaker development, is this is potentially harmful and a serious flaw with the Asgard. The Schiit amps are clearly not designed for good objective performance. And when designers have other goals in mind there are often some serious side effects. NuForce doesn't seem to care the uDac-2 has serious channel balance problems for example. They claim that bad channel balance was necessary for the best sound at the price. The Asgard may have been designed with similarly misguided, or sloppy, priorities.”
Of course, Head-Fi went nuts.
I was lucky enough to see this post shortly after it happened, so I jumped in, citing the turn-on transient as a relatively minor one (which is what I remembered from the engineering notebook—in retrospect, I should have just grabbed an Asgard and tested it—but it was a weekend, I was tired, blah blah), defending our design against his assumptions about us putting all the cost in the chassis, and bringing up some reasons why we didn’t want to use relays (most of all, contact degradation from outgassing, which NwAvGuy dismissed out of hand, although it is clearly present on the data sheet.)
The war escalated as a Head-Fi moderator stepped in to defend us (he owned an Asgard), and others weighed in. Mike Moffat even posted (as baldr) an emotional tract, calling out NwAvGuy as a coward and a bully, hiding behind anonymity with unknown agendas. Perhaps a bit over-the-top, but I think it’s important to say that engineers are no less passionate about their products than an artist or performer. Attack the product, attack the person.
On Monday, more data came to light. The Head-Fi moderator tested his Asgard and found that the DC offset was more like 1-2 volts, not 150mV.
I confirmed this measurement.
Crap. What now?
The notes on the prototype didn’t jive. So I tested a few more units, then went back and tested the prototype. All the same—an order of magnitude higher than I remembered.
The explanation for the discrepancy between what was in our notebooks? I don’t know. Maybe I wrote the numbers down wrong. Maybe I was measuring the wrong thing. The bottom line: never assume the old data is right. I shouldn’t have. And I shouldn't have dismissed the question out of hand.
But, then again, I had no reason to suspect we had a problem. We’d tested about 1000 Asgards by this time, all through the full on-off cycle, through a single pair of Sennheiser HD650s. The 650s were fine. No problems. Bottom line on this: don’t trust anecdotal results. And don’t assume everyone has HD650s.
An aside: to this day, nobody has ever blown up a headphone with Asgard, at least to our knowledge. Lyr, yes—which is why we added the relay mute to it before the whole Asgard Incident.
After this, I went back on Head-Fi, apologized for my misrecollection, and said we’d:
- Warn people about the need to unplug and re-plug headphones in the owner’s manual
- Investigate adding a relay mute to Asgard
- Refund anyone who had ever bought an Asgard, from the start, if they were disturbed about its performance
The third item—an unconditional refund on every Asgard—wouldn’t have killed the company, but it would have come very close.
So, why did I offer it? Simple. I was done.
I was making no money from Schiit, Centric was going great guns, and along comes an anonymous blogger and calls into question our competence and reason for existing. What other crazies would be coming out of the woodwork, I wondered? When would the next attack come?
Maybe this whole audio thing wasn’t such a hot idea, I thought.
Maybe better to pull the plug completely.
But, I sat down with Mike over dinner at a local Korean BBQ (I still remember going there, but I don’t remember the food at all, or even if I ate—I was really, really upset). And Mike talked some sense into me.
“This isn’t the end of the world. We have Bifrost coming,” he said.
“If we ever finish it,” I moaned.
“It’ll be done. And when it’s done, we’ll laugh about ever considering quitting.”
I sighed. “That’s easy for you to say.”
“No. It isn’t.” Mike sat back. “Theta had its share of attacks over the years. We were reported for FCC noncompliance by a competitor, and the FCC came in and shut us down. Hell, our offices were broken into, and our engineering computers were stolen. Someone was trying to get our digital filter code. The thing is, these days, the hounds are invisible. You can’t touch them. They can drop in out of nowhere, say whatever they like, and have no repercussions.”
“So what do we do?” I asked.
“We be better than them,” Mike said. “Add that relay. Kill the current run. Don’t sell another Asgard without it. And offer to update everyone’s current Asgard.”
“The whole current run?” I asked. That was a pretty big investment.
“Yes. The whole current run of headphone-killing amps.”
“But they aren’t headphone-killing—“ I began
“It doesn’t matter that they are or aren’t. What matters is that everyone thinks they are. Or at least enough people to matter. So, we go above and beyond. And make it good.”
I nodded. Mike was right.
The next day, I called the boardhouse and had them scrap the current run of Asgards. Then, in a 16-hour fit of engineering, prototyped a relay mute, added it on to the Asgard PCB, and ordered new boards, rush, from the boardhouse.
While I did this, the thread on Head-Fi grew and grew. Supporters, detractors, people bringing up other amplifiers that needed headphones unplugged, replugged, etc, etc.
Finally I was done. That was when I posted this to Head-Fi:
“That said, we now understand that the precautions common to ultra-high-end (where it's well-known that turning on your multi-kilobuck amp before you turn on your multi-kilobuck preamp may involve having a very bad day) simply won't fly with inexpensive gear, so we're making the changes necessary to have our stuff be as user-friendly as possible.
Asgards will now ship with the same relay mute as Lyr, when we are back in stock. We will also offer a retrofit relay mute for Asgard and Lyr, for customers who want the convenience, and install it on any current owner's amp for free.”
An aside: This offer continues to be in force for all Asgards and Lyrs. Because, even though we’ve alerted the owners in our database, there are (a) second- and third-hand products out there with owners we don’t know about, (b) many owners who didn’t want the relay, and (c) our database is never 100% accurate—people move, etc.
But the biggest thing that happened in this thread was that NwAvGuy got banned.
Now, the reason for his ban wasn’t his criticism of our products, but that isn’t what mattered. Because the ban happened during the Asgard Incident, and because of what NwAvGuy posted on his blog afterwards, he’ll be forever connected to us, rather than the other manufacturers and DIYers whose products he’s criticized.
Oh, I don’t believe you, you might be thinking.
Head-Fi protects its sponsors, that’s why he got banned, end of story.
Nope. If Head-Fi protected its sponsors, that thread simply wouldn’t exist—and it certainly wouldn’t exist now, almost three years later. If Head-Fi protected its sponsors, lots of negative stuff simply wouldn’t happen. Think back on all the controversy you’ve seen here. Are the threads still there? Yes.
And—consider this—Jude contacted me, bought our products, and talked about them before we were ever sponsors. This is something Head-Fi does very well—uncovering new, interesting products and getting them out in the world, even if the companies that make them are not sponsors.
Coda: What We Did Wrong…and Right
What we did wrong:
- Relying on memory of a measurement that was incorrect
- Minimizing a valid question because we’re busy/tired/etc
- Not immediately checking the performance of the amplifier
What we did right:
- Throwing away the current run
- Redesigning and shipping ones with relays
- Recalling and retrofitting Asgards and Lyrs, free
NwAvGuy: The Good and the Bad
“So, you guys really, really hate NwAvGuy,” you might be saying.
The reality is more complex. NwAvGuy did a lot of good for the industry, including raising awareness about output impedance matching and the importance of measurements. And, as with the Asgard Incident, he helped make Schiit Audio a stronger company. For those things, I’m thankful. And, if he ever surfaces again, I’d buy him a beer.
But, he also brought a lot of absolutism to the fore, like the oft-stated idea that an amp (or DAC) with good measurements is audibly transparent, and cannot be improved upon. The “us vs them” mentality of objective and subjective, audiophile and engineer—the objective-subjective divide widened considerably during, and after, NwAvGuy.
He also assumed a lot, without confirmation. Like our “expensive” chassis. Yes, they look expensive, but they are not. Like our “design by ear” philosophy. In actuality, it’s more “confirm by ear.” The speculation that we don’t have, or know how to use, test equipment. The reality is that we have better equipment than the vaunted DScope.
And the bigger reality is: NwAvGuy, by his own admission, never touched a Schiit product.
Also on the good side, he helped kick the inexpensive headphone/DAC world into high gear with the open-source O2 and ODAC. Although neither of these designs are like anything we’d make, they’re very popular.
And that’s why you made Magni and Modi, you’re thinking.
Actually, no. Modi is why we made Magni, and Modi actually appeared before the ODAC, at least in prototype form (see the upcoming chapter, DAC in a Toilet Paper Roll.) I doubt if we would have done Magni much differently if the O2 wasn’t around (more on that in another future chapter.)
I guess the biggest difference is in one of philosophy. We have a “live and let live” attitude at Schiit. We don’t think we know it all, and we don’t believe that our answers are always the best ones. We know how much work it takes to bring something to market, and we salute every company out there. It was only after NwAvGuy, though, that we enshrined our basic principles here:
schiit.com/about/principles
So, if someone else can come along, kick us in the pants, and help us make things better, let me say in advance: thank you!
Where Did He Go?
Despite the joking about “Schiit had him offed,” the reality is we have no idea who he is, or why he disappeared. But let’s have a little fun and speculate, because some of these ideas I haven’t seen in other places:
1.Muzzled by a retainer. As an audio consulting engineer, NwAvGuy’s retainers would have heart palpitations if they saw him attacking other companies. If they found out about it, they may have said, “please stop doing that,” without the please, and with a threat of contract termination.
2.Hidden in plain sight. Maybe he’s now working in the industry, and knows the ramifications of attacking other companies (in short, lawyers, lawsuits, expenses, bad stuff all around, see Apple and Samsung.)
3.Gone to the subjective darkside. Maybe he heard some gear that was incredibly magical, but measured like crap. Maybe that rocked his world enough that he’ll next be reviewing at The Absolute Sound or 6Moons.
Coda 2: Business Lessons
So, what did we learn from all of this? I can joke and say, "Well, we learned to put relays in everything," but that's not entirely true.
We did learn a lot on the engineering side, and it did push us to put one of the most advanced, analog-computer-style protection system in Mjolnir, and perhaps the most advanced protection system ever in Ragnarok, fully microprocessor-controlled for all operational and fault states, running proprietary algorithms that both manage thermal runaway and conditions that may cause de-biasing of the output stage (such as when playing loud, compressed music.)
But most of what we learned was on the business side. It reminded us that we don't have all the answers, that we can and do make mistakes, and that we have to stand up, admit them, and make it right when we do.
Because no business, no matter how great the engineers, no matter how skilled the production team is, no matter how solid the logistics guys are, no matter how enlightened the management is, is infallible.
You screw up. Bad things happen. And you make them good.
