The New Colossus: Building Wolfenstein II atop a million small decisions

Jens Matthies loves to grow things.

Tomatoes, strawberries, grapes, onions, brightly-colored peppers; you name it and, if you could grow it in a house (or on a balcony) in Sweden, he's probably done it. 

Some years his garden lays fallow, when his work at Uppsala-based studio MachineGames gets too busy. This is one of those years.

Matthies serves as creative director on the studio's upcoming game Wolfenstein II: The New Colossus,  a reprise of the role he played in the development of the game's 2014 predecessor Wolfenstein: The New Order. It was MachineGames' debut, and its salvation -- the studio reached a deal with Bethesda Softworks to make a new Wolfenstein game just weeks before Matthies expected to lose his apartment.

The New Colossus debuts in late October; a few months later, Matthies  will mark his 20th year in the game industry.

"My secret goal is I'm going to lobby to be a senior creative director," he says, with a self-deprecating laugh, during a recent chat with Gamasutra. "If I get that through, that will be my celebration...it's been a good run."

And there's probably a ways to run yet, if MachineGames has its way; Matthies noted in a recent E3 interview that the studio has always envisioned its Wolfenstein games as part of a trilogy.

Even if that doesn't come to pass, MachineGames will soon have a second Wolfenstein under its belt, bringing it in line with the likes of Muse Games or id Software. Both helped define the franchise (with Castle Wolfenstein and Wolfenstein 3D, respectively) and both are credited with two big Wolfenstein games apiece.

MachineGames' spin on Wolfenstein won critical acclaim for its characters and storytelling, which is something of a rarity for both the series and big-budget first-person shooter games in general.

A few years ago Matthies suggested to Gamasutra this was in part because the studio had focused on making protagonist B.J. Blazkowicz feel like someone you could easily relate to, rather than an empty vessel you could easily inhabit.

Matthies says he and his team learned a lot about how to best tell stories in first-person from that game and its standalone expansion, The Old Blood. But more than that, he says working on the Wolfenstein games have taught him a very important lesson: make decisions early. As many as you can, as early as possible.

"You need to make really strong decisions...and stick to them"

"Making a game is only about making decisions," he told me. "You have to make a million decisions, and once you've made all of them, the game is done. And it's incredibly hard to make those decisions, and it's especially hard in the beginning, because you know so little."

This advice is frustratingly broad and simple. Certainly it's quite apt if, like Matthies, you're working in a lead role on a big-budget game, but it's also applicable to lone indies and just about everyone in between. In a sense, life itself can be boiled down to a long line of poorly-informed decisions, many of them incredibly hard.

"No," clarified Matthies. "The worst thing for a project is changing your decisions and sort of careening in different directions. You need to make really strong decisions early, and stick to them, and keep making the decisions when you have to. Because it's also incredibly easy not to make a decision."

He offered the following example for fellow devs: say you need to create a character, early on. A sidekick, maybe, or an informant who sends the player on missions. You need to name this person, but since it's still early days you just call your character Bob. Bob One. You'll fix it later.

But now that decision is made, and that name ripples through your pipeline. The character model is called Bob. In the scripting, it's Bob. In the casting sheet, there's Bob. Finally, six months or a year or two years later, when the time comes to actually name the character, there's a tide pushing against you. It's a tide of Bobs. 

"So now you're stuck with this confusion point where if you're looking under the hood of the game, you have a character called Bob," Matthies concluded. "But as a player, it's called Johnny Evanson or whatever. So it takes a little bit more time to actually come up with a name if you decide what a character is going to be early on, but it's worth taking that time. And that goes for all of the decisions. Like, everything, it's very important just to make a good decision early and stick to it."

This is a silly spur-of-the-moment example to illustrate something Matthies evidently takes very seriously: the notion that every decision you make, even the ones you make by deciding not to make a decision, influence your game in ways you can and can't see. 

This applies to pretty much every game dev, whether you're working for yourself (on your own project or someone else's) or a larger team. If you're working with a team, Matthies' point is especially salient because you influence everyone else's ability to get things done.

"Every person on the team, when the project starts, they're like, what are we doing? And in order for them to work at peak efficiency, they have to know that," he said. "So that means you have to decide everything. How is this going to work? How is that going to work? Everybody's blocked, basically, until you make the decision. And once you make the decision, things get moving."

So, okay. It's all well and good to say "decide everything, as early as possible", but in practice what does that actually look like? For Matthies, it looks like two specific things that should be done for every game development project: prioritize decisions that are likely to be contentious, and decisions about features that will take a long time to implement.

The former might not be applicable if you work alone or with a very harmonious team, but it's a good reminder if you (like Matthies) are used to working within big groups with multiple teams.

How an early decision echoes through MachinesGames' Wolfenstein 

"So like, I might pitch my story idea to the gameplay team and they might hate it, because they want to do this gameplay thing, and I might hate that. So the easy thing to do in that situation is to say 'oh fuck this, I'll go do something else for now,'" Matthies said. "Then that decision doesn't get made. Then two years later, you pay the consequences of that, because it didn't get sorted out and people haven't been working on the problem."

As an example of why he takes the latter recommendation so seriously, he points to the decision to design The New Order with a "dual timelines" narrative that sees the game change in subtle ways based on a choice the player makes early on.

"The thing is, we knew that if we're gonna do two timelines, and we're thinking about this as a trilogy, we are going to double up a shitload of work. And it's going to compound, over the years," recalled Matthies. "The consequences, in terms of development, are grave. Like, they are really severe. And they were unknowable, at the point in time where we had to make this decision."

Without yet understanding the full scope or cost of development, the conversation inside the studio was evidently not an easy one ("we knew we were going to be paying for it for a long time"). Matthies said an agreement was eventually reached to implement dual timelines, very early in development, and now that decision is reverberating through the studio's work on the sequel.

Which is in itself an interesting challenge to MachineGames, since this is the studio's opportunity to build upon its established tools and assets rather than putting something together from scratch.

"A real, true, proper sequel has never happened before, for us. Not only that, but when we were at Starbreeze we never owned anything; we had to build everything from scratch for every new game," Matthies said. 

"We never had the opportunity to iterate on something that we had. Like, all the animations had to be rerecorded for locomotion, or whatever. And there's some good stuff about that, for sure, you can learn from your mistakes and do it a little bit better every time. But it's a very long walk just to reach parity with what you did before."

Coming full circle

Now MachineGames is working with id tech 6 to create The New Colossus, building upon what it did with id tech 5 in The New Order. Working directly with id has brought Matthies' life as a game developer full circle, since he got his start in the industry modding an id game: the original Quake.

"Back then it was all about playing Quake, modifying Quake," Matthies said. 

"This was of course before the time of blogs, but there was something called .plan files. I was reading all of them, hanging on every word, and once I started making games at Starbreeze, the first time I went to E3, I saw [longtime id staffer] Tim Willits. I think they were working on Quake 3 Arena, and they had some sort of stage thing set up and there was people playing Quake 3. And I was like 'holy fuck, that's Tim Willits!'"

In doing a bit of research, it appears there was a Quake 3 Arena tournament hosted at E3 in 2000. Professional player Johnathan 'Fatal1ty' Wendel seems to have been the winner, netting himself the top prize of a Suzuki motorcycle then valued at ~$11,000. When I asked Matthies about whether he saw Fatal1ty there, his face went blank.

"I don't know. I don't know what that is," he said, adding that he only knew Tim Willits thanks to his .plan files.

"Because they were my heroes I knew them all. I knew everybody that worked at id, I read their .plan files, I knew what they looked like and everything," he continued. "So once I actually saw Tim Willits at E3, that was like 'holy shit, that's him!' That's a celebrity. Right? Of course, as devs, we were making levels at the time, with similar tools. So these were like our gods, you know. If we could make a level as good as Tim Willits makes levels, that would be like [makes a chef's kiss gesture] we have reached our peak."

Matthies grew up playing and making games for systems like the Amstrad and the Atari ST. He joined an Atari demo group, went to conventions and got really into 2D graphics, then basically said goodbye to it all when he saw Wolfenstein 3D running on a PC.

"Of course PC, nobody took seriously. And then Wolfenstein 3D came along and I was like, 'what the fuck is this?!' It blew my mind so totally. And of course then Doom came and I was...I was over," said Matthies. "That's why those platforms, Amiga and Atari, died; because once people saw Doom, the game was up. You can't go back to 2D when you're playing Doom. You see somebody with a PC, and they're playing Doom, it doesn't matter how many colors your thing has; it's not Doom!"

Playing those games convinced Matthies that virtual worlds are best experienced from a first-person perspective, so the player can feel like they're deeply embedded within the games they're exploring. 

Doing things outside of games makes you a better game maker

Matthies says he gardens for the opposite reason: he appreciates having something to do that takes him away from games, away from the projects he's working on and the problems the team is dealing with. His time spent practicing mixed martial arts benefits his workflow in similar ways; it breaks it, forcing him to come back to a problem later with different eyes.

"I think it's incredibly important to be able to focus on something that's not the game. I think if you're a good developer, then the passion of your life is the game. And that's hugely important, because that's the only way to make a good game," Matthies explained. 

"But the downside of that is that you're constantly thinking about it, and you get so incredibly wrapped up in it that there's no time for your brain to sort of calm down and create some distance to the work. Because that's hugely important too."

"It's so hard, but to have some sort of mechanism by which you can distance yourself from your work, to at least attempt to neutrally evaluate it, and make good decisions on what to change and what to not change. And I think any kind of side interest that you're able to fully focus on, that takes you away from that for a few moments or a few hours, is incredibly important. And I think growing stuff is one of those. But I recommend having several, so you can go to that place in your mind when you need to."

After The New Colossus ships, Matthies has a new project: he's finally been able to put money down on a piece of land, a place where he can build his own greenhouse. More importantly, it's room enough to accomplish a dream he's nurtured for a long time: to plant something that will outlive him. To plant trees. Big, beautiful, colossal trees.