The Art of Headlander
Lee Petty, Project Lead at Double Fine Productions, talks about making lasers cool again with their new game Headlander.
I’ve been a huge fan of Double Fine ever since their first game, Psychonauts, blew my tiny teenage mind. So I was pretty excited when they started using Niice while developing their latest game, Headlander.
Lee Petty was the project lead on Headlander. We had a chat about how to keep a creative team moving in the right direction, designing snarky doors, and making lasers cool again.
Chris: When I look at all the games you folks have done, no two of them look alike. You don’t have a house style. What is it about the culture at Double Fine that enables that?
Lee: When a game like Headlander comes out, some reviewers usually say “this is unlike anything they’ve ever done”, because in their mind they’re expecting something like Psychonauts. But when you actually go and look at our catalogue you realise they’re all quite a bit different; both in terms of looks and sensibilities. Some of them are tactical, some of them more action, some of them more ‘puzzle-y’... I think it’s partly just down to the fact that our games have been led by different people, and Tim (Schafer) encourages people to bring their own sensibilities to the project.
“We’re a creative-led studio, not a genre-led studio.”
Tim always says we’re a creative-led studio, not a genre-led studio. A lot of studios are based around making a certain type of game, and there are a lot of advantages to that; you get good at making those types of game, your tools & technology are best suited for it and it makes it easier to do sequels. But there is also something nice about approaching things fresh each time, so that’s generally the approach we take.
For me personally, I guess I look for certain sensibilities. Headlander isn’t generally a laugh-out-loud sort of game, but there’s quirkiness to it and for me that’s in service of personality. Personality is a very generic term, but it’s something I think that all the Double Fine games have or try to have anyways. A lot of bigger games try to avoid personality at any cost. It’s like light beer: you can sell a lot of it but it doesn’t have any particular personality.
In terms of our creative process, in all projects there’s that early stage where we’re trying out different things, trying to find the most appealing element. For me it’s trying to find a visual style that fits the themes of the game as much as possible. So, for example, early on in Headlander we had a more ‘cartoony’ art style, and it was appealing and in some ways had better readability, but it didn’t feel right to me. We kept trying different things — we’d do really rough models, or rough coloured things, and get them in the game to look at with the game camera. Most of my inspiration was coming from 70s science fiction films & literature, and if you look at the artwork and the film sensibility of the era, they weren’t particularly trying to be stylised. They were, but they weren’t focused on it as a point of style. So I think when I saw the cartoony stuff, I thought “No, this is really goofy”, whereas I felt the 70s stuff was more absurd than goofy.
But it took itself quite seriously at the same time.
Yes, it did, but the premise was somewhat ridiculous and I liked that. I found that tone, that sort of clash, interesting. So for this game it felt better having the character look a little more straight, but being in an absurd world. Serious things are going on, but you’ve also got a lot of really ridiculous references and strangeness. That felt more in line to me for this game anyways.
So it’s more about absurdity rather than slapstick.
Yes.
For Headlander, which came first: the mechanic, the story or the world?
The mechanic, world and look tend to all come at the same time for me. For this game, it was just basically a high-level concept of, “You’re just a head in this world”, and the framework of the world followed from that: “How does that make sense? Oh its robots.”, which led to a lot of the 70s themes coming out. The actual story and characters came later, but that world building and high-level mechanic idea usually come at the same time for me.
I did Stacking as well and that was a little different. I knew I wanted it to be Russian dolls and I knew I really wanted a diorama world influence. In the first of the prototypes there was no Blackmore family, there’s no chimney sweep, there was no world’s smallest doll. That all came a little bit later, but the basic visual idea of this stop-motion, diorama world and the ability to stack up and stack down dolls was all kind of simultaneous.
Doesn’t Tim work similarly to that? I read that Grim Fandango started with the idea of mixing the aesthetic of Day of the Dead and Film Noir, and then figuring out a story within that world?
Yes, I think he works a little differently on each project, but I think it’s about creating a context and then putting things in it and letting that organically evolve to some extent.
When we did Broken Age, Tim just had a very high level idea about two worlds that had a connection with a boy and girl. One of them involved monster sacrifice and the other one involved a hermetically sealed future environment. That’s all he had, he had no other stories and didn’t know who the characters were, and even that was pretty loose. So we did a lot of drawings on stories and ideas within that very loose framework and then things started coming out of that. That was a little looser or a little less defined for him than say, Brutal Legend, because Brutal Legend was built around the idea of a roadie having this epic adventure. Then the world back story came out much later after this initial idea.
“I just wanted to have lots of lasers and stuff.”
Tim doesn’t tend to come up with mechanics first, he tends to come up with a world idea and character. I tend to start with one core mechanic, maybe not all the game. In Headlander it was two things: I had an image of launching your head and flying around like a lunar module, and I had a phrase, “Making lasers cool again”. Everything’s bullets nowadays and in 70s just a simple laser was magical, so I just wanted to have lots of lasers and stuff. Those were my two early things.
Which I guess led to the geometric puzzles. It reminded me of this old snooker game we used to play in school, where you had to find the right angle to bounce stuff off.
[laughs] Yes, there is that kind of angle to it. There were times where the game was a bit more puzzley and a bit less combat. And then it ended up being more combat, a little puzzley. It kind of went back and forth, and a lot of that comes out when you are developing the game. You have a set time to develop the game and you go after the things that are feeling right in the moment. One thing that separates games from other forms of art, like movies, or say a novel, is that the games have so little editorial time. You basically don’t see the final product together and have it be playable until right at the very end. And then your ability to sculpt it is always limited compared to say, film, where they over shoot a bit and they make the whole thing in editing. Also they have all the tools they need for the most part nowadays, we don’t have that in games.
“One thing that separates games from other forms of art is that they have so little editorial time.”
The cost of even throwing away the equivalent of a scene in games is sometimes not an option, because games are reviewed, judged and bought by value proposition. No one cares if a movie is an hour and 35 minutes versus two hours and 20 minutes. There’s a pretty big range there. But in games, at a certain price point it’s got to have a certain amount of gameplay. That’s been changing in recent years though. You have games like Abzu,which came out recently, and that’s only about two hours but it’s been selling pretty well.
Do you think it’s because of how young the industry still is? In the early days of Film, when it was still a new technology it cost an absolute fortune to do anything, so they maybe struggled with the same issues back in the early days of the medium and just got more and more efficient at it?
Maybe, although Film isn’t intrinsically tied to technology in the same way. The basic idea of shooting scenes hasn’t changed much. Yes, they’re great now, but it’s pretty much the same view, just better. Games on the other hand have changed quite a bit, and there’s still a bit of just figuring out what games are.
Games were seen as a child’s toy for a long time, and therefore reviewed as a gift proposition. The culture of looking at games through that ‘toy’ lens hasn’t really changed, so I think that hasn’t helped games. You see a lot more short form games now, but they tend to not make any money. They’re mainly games done by individual creators who are having an artistically fulfilling experience, but as an industry it’s hard. There are some exceptions nowadays though, which is awesome, because there didn’t used to be any.
Was that a concern with Headlander? Were there any aspects that you would have liked to do for artistic reasons but had to cut to make it more feasible commercially?
Not too much. We’re in business to make games, and we want to make more unique games. So to some extent I always look at it from the mindset of commercial viability, and that probably weighed heavily in on my thought about action early on. Some of our puzzle games are sort of classic adventure games, and I think they’re well executed, but sometimes we have trouble finding an audience because people don’t want to read or sit through that many cutscenes and whatnot. So my thought was “can we apply some of our sensibilities to something a little more action focused, so that you’re going to be immersed in the aesthetic, but if you don’t want to dip too deeply into the story or talk to the side characters, you don’t have to?”.
“Can we apply some of our sensibilities to something a little more action focused?”
So that was certainly a consideration, but I don’t feel like we made too many compromises there. Mainly, our compromises came from trying to make a game around the size and budget of Headlander, which is difficult for us to do. Like, I want to put in all these unique looking rooms and, well, that costs a lot more than a generic grungy room. But then I’m like, “Well we’ve got the characters in there, and some of them might have to have unique dialogue”, but as much unique dialogue as I have there’s still a lot of repetition in it, because most action games just don’t fill their rooms with that many characters, right? To some extent it only hurts us to have those elements, but we do it because it’s something we care about.
How many of you were involved in developing the game’s aesthetic?
Ultimately, the whole team contributes to the overall game, including the aesthetic, and some people touch on that more than others, depending on their role. Fr example, our visual effects artist did pretty much all of the effects, so the connection of the visual effects to the overall 70s aesthetic was something that person worked on a lot. When I started the team off, we started off with a few concept artists and a couple of programmers. I told them about the influences and the high level of what I was after, showed them some drawings and paintings I had done, and then they each explored different ideas. Then, ultimately, it becomes a curation process to take what’s all moving in the right direction. That then gets built & realised in 3D, and mechanics, and iterated on. It was a team effort.
In this particular project I went in saying, “I want to make a science fiction game, but I don’t know what type. I’m specifically interested in this weird era of science fiction and here are some of my major influences, and some of the artists”, and then they’d go to find other artists, or would have a different spin on it, and that would be incorporated. There’s always that line, when you’re leading a project, to make sure you’re providing enough direction to get people started with a core flavour, but also giving everyone enough freedom to add, clarify and improve as it’s developed. Otherwise, it’s not much fun.
“There’s always that line, when you’re leading a project, between giving enough direction and giving enough freedom”
How do you structure that? How do you ensure that everyone is having an input, but at the same time it’s not going crazy?
It’s pretty much just the culture of the studio, it’s been there for a while. Part of it is that anyone’s allowed to comment on anything. With a smaller team there’s not a ton of hierarchy, but most people understand that the Project Lead has the final decision. I don’t feel like I made a lot of controversial decisions, I sold them on this high level vision, and when they would contribute and I didn’t feel it was quite right, I’d say why I didn’t feel it was quite right. Sometimes there were things I wasn’t 100% sure of, that someone else was really passionate about, so we’d put it in the game and let it percolate for a while, then maybe we’d change our minds, or the context changed, or it got pushed and pulled a bit.
Just giving people timely feedback is a big one. A lot of project leads get so busy they don’t, and then people wait for days or weeks and they get frustrated and just do their own thing. Then there’s clear communication; If you’ve got a particular thing that’s important to you, communicate that. And if you can even express it in a higher level idea, that helps too. One thing we’d say on Headlander was, “70s science fiction is equal parts existential and stupid”, and I like both those things. So when we looked at a design, we’d be asking “Does this speak to the technology of the time, or some idea of identity?” but also “Where’s the really stupid comic element?”.
We tended to have people developing similar areas at the same time. We’d use the weekly team meetings—where we play through the area and discuss it as a group — to let people see how it’s progressing. This meant people were throwing ideas about earlier rather than later. Then there was this whole library of ’70s films we watched. The whole company, actually. We just showed them at lunch over two days and people would see those. So they got inspired to see some of the context.
That’s the thing about the art style. It’s very familiar, but still very unique. You’re meshing a lot of things together. What was the hardest part to make work visually?
We used a side view presentation, which is nice because you have a little more control over the composition than you normally do in games with the cameras moving all over the place. That was good, because it flattened the space a bit. I also used a field of view on the camera that was pretty flat, which kind of reminded me of some of the art from that era that was pretty flat. There wasn’t a lot of deep perspective type stuff.
But because we wanted to have it be so colourful, (and also use colours as part of the game mechanic), readability could be an issue. For example, you could lose track of your head in a firefight. That was one of the most challenging things: balancing the aesthetics with readability of the gameplay.
One of the things we implemented was depth-based colour grading in the game. The camera was set up with four layers of Z-Depth colour grading. Each of those layers could have their own complete post-processing, and they blend between each other based on depth from the camera. This allowed me to frame certain layers in certain ways. So if we had a really blown up background, rather than that being generally lit like games tend to be, which would blow the character out, the player is constrained to a certain slice of the Z-plane. I could colour grade that differently, so that you’d read against that rendered background. But when they went in to a darker room, I could flip it. That’s just one simple example.
Then there are certain things like the door colours or laser colours, which were important outside of the crazy colour schemes. I could exclude them from post processing so they would always stay pure, and try not to put too many similar colours right around them, so that people could distinguish those. That’s actually where the logic of the door came from. For most people the difference between red, orange and yellow isn’t super dramatic, and for colourblind people it’s even harder. So we had to have auditory feedback for the doors, then when someone used the wrong colour that’s a way we could tell them without putting up hint text all the time. But so long as I’m doing that, I’m going to give it a personality. As an artist I was naturally snarky about it, I just wanted to say “That’s not orange that’s yellow.” So the personality and puns just grew out from that function.
The door puns were worth it all. Worth every penny.
[laughs] Yes, the doors are good.
In terms of managing a creative team, are there any things you’d do differently in your next project? What do you feel you did well?
On every project you ship, you’re always going to incrementally improve on things, though I think we did a pretty good job overall.
Getting people playing the game and getting constant feedback about the game was useful. We had Slack channels for the different levels of the game, and they’re not closed channels so anyone could join. People can see what other teams are doing, and there’s also a lot of mobility between teams as projects ramp-up and ramp-down. When it comes to images Slack is kind of like this unruly West; good channeling is okay but using it for feedback about an image doesn’t really work. Niice helped with that.
We found Niice boards to be an easy way to keep stuff for a particular topic all in one place. It enables the level designers and world artists to easily get a snapshot of what a level feels like. I put together colour bars and overall maps in Niice, creative direction-type things, then show them to the artists and give them feedback on the concepts they’re generating. The concepts we’re happy with are kept there too, so it’s an easy-to-reference receptacle for all those things that are visual in nature.
“Niice enables our level designers and artists to get a snapshot of what a level feels like.”
I think I probably could have done a bit more brainstorming with the team. I’m not a giant fan of brainstorming generally; it’s fun, but straining through 10,000 ideas to try and pull a thread is a lot more work than say presenting a rough thing and getting feedback on it, revising & iterating. But sometimes brainstorming is helpful just to get people engaged. We did do some of that, but maybe we could have done a bit more. Also, as a group we probably should have played more games in the same genre to get people excited about where the gameplay could go. We watched a lot of reference, but we didn’t play anything else besides our own game as a group much. Next time I’d probably do that a bit more. Sometimes it’s a little out of your control; when you start a project you don’t have a lot of time, so you start with the things that you feel most sure about and get moving.
Most people, I think, were pretty happy working on the project. It was a fun team and everyone got along well. Part of that was because we always tried to meet as a group, more than as subsets of the group. That dynamic doesn’t work if you have a large team, but we never had more than 15 people on the project at one time, so everyone was able to get together. I didn’t have any discipline-specific meetings for the most part; we had some separate channels so that if I needed to coordinate build process with the programmers, it didn’t need to be a whole group conversation, but anyone could still join that conversation. It was just a way to limit people’s noise. In bigger teams you can’t involve everyone in every meeting, so it is still helpful to have discipline specific meetings. But with these smaller things, as much as possible, we try and keep it not discipline based. It’s more like, “You’re a team member, you have a primary job that you do and primary skill that you do, but you get to have an opinion on this whole thing you’re making.” ▯
Headlander is available on Steam and the Playstation Store.
You can follow Lee on Twitter @LeePetty, and Niice @NiiceApp.
