Back in February of this year, a Texas jury awarded half a billion dollars to ZeniMax in the long-running Zenimax vs. Facebook case. Half a billion bucks is a lot of money, even for these titans. Zenimax is of course the parent company of Bethesda, of Skyrim and Fallout 4 fame. Facebook owns Oculus, the former darling of the VR headset scene. So while in court this was a fight between Zenimax and Facebook, to the gaming community this was a fight between “Bethesda” and “Oculus”.
I don’t really care about the petty slap-fight between these two gargantuan companies as they bicker over a pile of money neither one of them needs or knows what to do with, but there are some really interesting side-arguments going on here about source code and VR.
I know Zenimax is overbearing, petty, and not above bullying people with lawyers. I know making inaccurate statements is a good way to piss people off, and it’s really easy to do that in a case this complex. Back in 2010 I wrote a column that explained that you could sue anyone for anything as long as you could afford the lawyers. This is still true, and Zenimax can now afford another half a billion dollars’ worth of lawyers. That’s a lot more lawyers than I can affordZero. so I’m hoping that hanging the word “opinion” over this series and filling it with very obvious embellishments for comedic effect will act as a totem to ward off the Zenimax legal shock troopers.
This article (or series of articles) is my opinion and analysis, which is based on imperfect knowledge of complex events and shaped by personal biases. The parts that aren’t obviously opinion are often extrapolation on my part. I am not a lawyer, I don’t have any special inside knowledge of this case, and I have no expertise in corporate contracts except that I’ve signed a few and always felt uneasy about it afterwards.
To be honest, this entire case is a fractal. From a distance it looks like Zenimax is suing Facebook over VR, but then you zoom in to see the details and it’s an argument over trademarks, and if you zoom in on that it’s a he said / she said argument over source code, but then maybe it’s also about a contract, or maybe deep down it’s about a couple of engineers and a trade show demo? Or maybe a Kickstarter campaign? It’s actually kind of all of those things, it just depends on how deep you want to dig and how much patience you have for the fussy details.
I’ve read numerous news stories about this case and I still feel like I’m missing context. I’m not sure if two news stories are contradicting each other or if they’re simply reporting different fragments of the same larger story. It’s like a game of telephone where the message changes a little each time, except I can’t even tell what order the messages moved in, or where the gossip chain started.
I even asked a lawyer for help, and he was able to point me in the direction of more documentation and clarify many points for me. As a result I’ve gone over some legal briefs like this one. They turned out to be surprisingly readable. Having signed my share of legal documents, I know how impenetrable legalese can get. I assumed court briefs – documents aimed specifically at lawyers and other experts – would be even more baffling than the mortgages and NDAs I’ve dealt with, which are ostensibly intended to be read and understood by laypeople. But as it turns out, some of this court stuff can get pretty informal and it’s not particularly dense with jargon.
As helpful as this was, it only served to show just how much is lost in translation when trying to cram a complex disagreement into a typical news story. I felt like the more I read about the case, the less I knew. I just kept zooming in on the fractal, trying to find the final level of detail that would give me some kind of handle on the thing. I’ve put in several days of work in trying to write this series, but I’m still doubtful I’ve gotten all the facts exactly right. I’ve done what I could, but there are limits to what I can accomplish with the limited information and expertise available to me.
One final note is that doing this research has helped me to understand the Zenimax position a little better. I was pretty firmly against them based on what I’d read after the verdict, but now that I’ve read the details I find their position less objectionable. I wouldn’t go so far as to say I’d side with them in the disagreement, but their arguments now make a lot more sense.
Except for their arguments on source code. Their arguments on source code would be hilarious if they weren’t infuriating. But we’ll talk about that later.
I’ve been pretty public about my professional admiration for John Carmack, former head programmer of id Software. He was the creative force behind the technology that drove the various incarnations of DOOM and Quake. He’s not as big a force these days, but only because the technology has grown to the point that no one person can single-handedly exert industry-wide pressure the way Carmack did in the 90s. In a field this big, it was pretty amazing that one person was able to stay so far ahead of so many rivals for so long. His technology wasn’t the only thing that made their games a success – id Software had lots of other brilliant creative people in their ranks – but there’s no denying his graphics engines were pushing the limits of the medium, selling faster computers, and making id Software an industry darling among both fans and critics.
Back in 2009, Zenimax bought id Software. In the deal they got the stuff they were interested in: Some trademarks, a back catalog of classic games, and some technology. But as they were unpacking the cardboard boxes of intellectual property and trying to figure out what the hell a Commander Keen was, they realized they’d gotten something else along with the deal: A middle-aged genius software inventor with a resume that makes him sound like the Thomas Edison of videogames.
“But Shamus! Edison wasn’t as great an inventor as people give him credit for. He just made commercial versions of a lot of ideas that were already floating around the scientific world. Also, Tesla was way cooler!”
The comparison is more apt than you might realize. A lot of Carmack’s greatest accomplishments were built on tricks and ideas that mathematicians had been playing around with for years. Carmack didn’t invent binary space partitioning, a key component of the DOOM engine. That idea had been around since the late 60s. Carmack was just the first person to make it work on then-modern hardware and figure out how it could be used to make a game.(And then make it really fast.)
Although there is one really important difference between Edison and Carmack: Edison had over 1,000 patents to his name, while Carmack is a big believer in the hacker ethic, which takes a dim view on efforts to “own” and control ideas using things like patents.
Like Carmack’s previous bosses, nobody at Zenimax was really sure what to do with him. You can’t make him invent something, after all. Still, better to employ him than to have him go work for a rival company. Also, he still had the ears of hardcore graphics nerds, and thus he had some sway with hardware manufacturers, and that’s never a bad thing to have around the office. It’s not clear if they gave him specific work to do, but they did allow him to spend at least some of his time on his own projects.
Apparently what Carmack really wanted to be working on was Virtual Reality. There wasn’t a lot of really interesting breakthrough work to be done making graphics engines in 2009. If you magically invented some clever technique that doubled graphics performance, what would that get you? Visually, that’s maybe the difference between running Crysis on “high” graphics settings and running the game on “ultra”. That’s nice, but that’s not the sort of jaw-dropping revolution like DOOM or Quake brought to the industry. Those old games made people say things like, “I never knew this was possible!” But at this point in the tech tree the innovations were limited to stuff like smoothing the edges off of jaggy pixels and adding another lighting pass. That makes people say things like, “Yeah. That does look a little nicer” as they squint at the screen. It’s just not the same.
Which means it’s really hard to impress people these days. Games look pretty good already and even a huge gain in performance will result in small gains in visual quality. But VR? That was one area where you still might be able to blow some minds.
The Long Troubled Evolution of VR
People had been “working on” VR for decades. It’s tough to pinpoint where it all started because the early definition of VR was pretty muddled. To someone in the 1970s, any modern first-person game would qualify as “VR”. In 1978 MIT created Aspen Movie Map, which involved taking photographs of Aspen Colorado and allowing the user to navigate through them. That qualified as VR by the standards of the day, but it was more like a precursor to Google Street View.
Eventually people settled on the idea that “VR” would mean using a computer to render an enveloping 3D environment, with distinct views created for each eye to create a sense of depth perception. We tried it in the 80s, but the CRT based displays of the day were heavy, flickering, and hampered by heavy cables. It was basically the last part of the computer you’d want to strap to your face.
By the time Carmack stepped up to the plate, VR had been “five years away” for a quarter century. People would build VR headsets, realize the problem was harder to solve than they first anticipated, and the effort would die. Then we’d get a fresh round of breakthroughs in rendering, displays, and circuit miniaturization, and people would take another crack at it. Eventually people would run into a new hurdle and it would die again.
It turns out VR is a really hard problem to solve. You need extremely high resolution screens, because once they’re strapped to your face those tiny pixels are going to look gigantic and blocky. You also need excellent refresh speedsIf you’ve ever pressed your face up against an old television screen you’ve probably noticed that the seemingly steady television image is actually very strobe-like.. But you also need the screens to be small and lightweight because they’re hanging off the front of your face. Also, you don’t want a heavy bundle of cables tugging on your head, so the device can’t use too much power. And the device needs to be cheap enough for the average consumer.
So we need screens that are high-definition and lightweight and small and power efficient and cheap? Good luck with that.
Oh, and that’s just what we need for the screens. We have to overcome similar challenges with lenses, rendering technology, processing power, user input, and head tracking. Everything needed to be custom-made, cutting-edge, small, and yet somehow still remain cheap enough for middle-class consumers.
In 2012 Carmack thought the technology was close enough for another go. He began working on his own headset in the mad scientist wing of the Zenimax basement labyrinthIt’s right across from the bestiary, which I’m pretty sure just contains an ice troll, a super mutant, and ten thousand cliff racers.. As part of the project he was also reading various enthusiast sites to see what everyone else was doing. The hacker ethic is strong in groups like this, and it mostly boils down to a bunch of engineers hammering away at a problem, comparing notes, and maybe occasionally smugly showing off when someone gets a breakthrough. At this point Carmack discovered then-20 year old Palmer Luckey, who had cobbled together a headset from consumer parts.
This is a complicated story, which means it’s going to take some time to unpack it all. This series is going to be five parts long. Next week we’ll talk about the technology Carmack and Luckey built together.
 If you’ve ever pressed your face up against an old television screen you’ve probably noticed that the seemingly steady television image is actually very strobe-like.
 It’s right across from the bestiary, which I’m pretty sure just contains an ice troll, a super mutant, and ten thousand cliff racers.
Zenimax vs. Facebook
This series explores the troubled history of VR and the strange lawsuit between Zenimax publishing and Facebook.
A look at the main Borderlands games. What works, what doesn't, and where the series can go from here.
Diablo III Retrospective
We were so upset by the server problems and real money auction that we overlooked just how terrible everything else is.
id Software Coding Style
When the source code for Doom 3 was released, we got a look at some of the style conventions used by the developers. Here I analyze this style and explain what it all means.
Raytracing is coming. Slowly. Eventually. What is it and what will it mean for game development?