Playing with Shadows and Legos: an Interview with Chambara Lead Designer

I recently attended Denver University's "Arcade", a game expo for local developers and students. There I met Tommy Hoffman, a student and developer for Team OK, the BAFTA award winning studio based in Los Angeles. Their game, Chambara, is a stylish splitscreen fighting game that meshes the environment and mechanics in a unique manner: one player is black, the other white, and the mostly dichromatic environment hides and reveals them for exciting and tense multiplayer matches.

Tommy put me into contact with the Lead Designer of Chambara, Esteban Fajardo, for a talk about level design.

Esteban, you're a developer with Team OK. We're going to talk plenty about Chambara, but before we get there I'd like for you to walk us through a timeline of your game development background.

I’ve been surrounded by game development for much of my life. My father works on games and even teaches them at the University of Denver's Emergent Digital Practices program. As such, I’ve been surrounded by the tools and books about game development since I was very young. My very first game was a bat echo-location maze game I made for a science fair when I was ten, and since then I’ve been making small personal games. Chambara will be the first commercial release I work on.

Can you talk a bit about the role you've played in the development of Chambara?

My role on the project has been lead designer, so in addition to working on many of the map’s layouts I’ve worked on the gameplay, UI, and a mess of other things. Each of those has been a team effort, though, and I’m lucky to work alongside super talented people.

What draws you to level design? Was there a "Game That Made Me Love Level Design" that you remember playing or still continue to play today?

We’ll get into this more later, but I think it was LEGOs which actually led to my love of level design! I like physical, sculptural things. So my favorite game is 'We <3 Katamari,’ which also has that sense of tangibility, even though it is a digital game. I also think the early Katamari games have great level design - their careful construction melts away and they offer a lot of player freedom, as well as quietly hiding hundreds of funny stories in the environment.

Building upon that, as I'm sure you're aware, level design in games is a varied subject. Levels are used differently depending on the kind of game. They can be used in single player linear missions, or for open world environments, or for close-quarters, multiplayer engagements. While you may not have made an open world game (because who's got the time for that) what is it specifically about this kind of multiplayer, combat-oriented level design that you enjoy?

The best part of multiplayer arena level design is that you get to spend a lot of time in the levels. The goal is to make a map that players will want to come back to time and time again, rather than making a bunch of areas that players will race through and never look back. And since the game will be competitive, we have to take a lot of care with every detail. That’s a lot of pressure as a designer, but also really exciting!

Cool. Let's get down to how you design your maps. I've heard from Tommy that you use sticky notes to great effect. I've also seen in your team's pitch video that you use Lego bricks too. Tell us about these tools you use in the early stages of level design.

It can be a little difficult to explain, but each map is made of particular arrangements of color and shapes. I use LEGOs and sticky notes to build the tightest elements, and I use that module to build up a stage. For example, our stage Radio Tower’s central motif is three walls arranged in a triangle, with the interior space all the same color. That module is then repeated and reversed until we have the basic layout of the entire stage. This gives each stage a distinct character and gameplay style.

The benefit of using LEGOs or folding up sticky notes into origami is that the stage element can be conceptualized in three-dimensions. I like to draw the maps out on paper and mark all the sight-lines, but since the game exists in 3-D it’s good to build models with height as well.

Let's say you have an infinite supply of Lego and sticky notes. What do you build first?

I would build full models of our current stages, because how cool would it be to have those? I never constructed entire stages from those tools - just the basic color interactions, like "what players are visible against what parts of a structure?” so having a tiny Chambara would be cool.

I love Legos. I had to ask. We'll get back on subject: You've been developing maps for Chambara for two years now. What was it like making those first few maps? What do you do differently now given that you're more familiar with what works and what doesn’t?

In the beginning, making levels was really tough! Chambara has a very unique set of rules to ensure that players can stay oriented against the minimalist art style and that no player is ever entirely invisible. The rule book for how to make Chambara style levels took a while to develop, and continued to evolve over the two years we’ve been working on it. The core mechanic is the player’s ability to hide in plain sight, so we embraced open arenas, and tried not to obstruct sight lines across the map. The final stage we developed for the game, Mansion, has no obstructions whatsoever, and I consider that a sign that I had finally learned how to make levels Chambara-style. I think players will enjoy that stage when the game is released.

Can you tell us a bit more about the process you went through figuring out what did and didn't work? Lots of play-testing, right?

Playtesting is the most important part of understanding whether a level will work or not. After conceptualizing a layout we’d build the stage quickly using primitive objects, and then immediately test it. Our game is incredibly easy to greybox and test, and we’d often be playing a map and iterating it within an hour or two of conceptualizing it. This let us cycle through ideas quickly, and spend most of our time iterating on what we felt had a spark.

Are you familiar with the Halo CE map called Chiron TL-34? If not, it's a map that is composed entirely of small rooms with teleporters that connect them. It is disorienting, it is exciting, it is a beautiful mess. I've read that the designers still laugh about all the rules they broke when making that map. Did you ever make any maps for Chambara that were beautiful messes?

I’m not actually familiar with any of Halo’s maps, but that idea sounds like quite a mess!! One of our earliest stages, Neon City, was originally a messy workshop where I would just chuck in experimental arrangements and patterns of structures without any thought of how they related to each other, just to run around them before developing them into their own stages. But somehow that area became the most popular of our initial stages! So we embraced the clutter, and fleshed it out. People still tell me it is their favorite stage, but it was an accident!

The opposing players are light and dark, and those colors are also in the environment and hide the corresponding player. How do you make sure that the environment offers a fair distribution of both light and dark locations?

To understand the stage on a deeper level, we’d place dummy players in each hiding spot on the level and make sure that each color had an even number of hiding spots. We’d also save “heatmaps” of the stages by taking a bird’s eye view picture of each match and analyzing where the action was happening. This was easy to do, since players leave a colorful burst of feathers at the spot of the map where they are hit, and all we had to do was see where the most colorful areas of the maps were.

It must be easier for symmetrical maps, right? But what about asymmetric maps?

As a designer, I like to make symmetrical, precise, “perfect” maps with entirely thought out sight-lines and patterns. When I come up with a design for hiding-and-finding places that can be replicated like a fractal infinitely, I get very excited. But that is not the sort of thing which is fun for a player - a good map needs landmarks and distinct areas. In the end, it is better to have asymmetry, even if it can risk a “perfect” balance.

I heard from Tommy that the goal is to have each location on the level be both advantageous from one perspective, and vulnerable from another perspective. How do you go about implementing such a feat?

The golden rule for Chambara’s levels is that no three faces of a corner can be the same color. If one face of the corner is the opposite color, then the opposing player can see them from that angle. Using that, it is possible to diagram out what sight-lines a player is visible or invisible from. The trick is then to map out routes so that players can move from one hiding spot to another and have opportunities to see their opponent while doing so. This is easier said than done! Especially since all the walls and their colors need to line up, and changing the face of one surface will affect all the other surfaces it is connected with. Making levels is a big puzzle, but that makes it almost as much fun as playing the game!

Another big factor is player movement and how they maintain their bearings. What elements of level design do you leverage to make sure the player does not get lost or disoriented?

The best landmarks are detailed art props. However, players cannot hide against detailed surfaces, since the opposite color will reveal them. So the art team and the design team needed to be in close communication to make sure the art would have space to give the stages life and orientation, and so the design would have fun hiding spots.

I've seen too that vertical space is especially important in your maps. It seems like spawn points are above the typical engagement area. Tell us about the thought process behind this pattern and other ways you utilize vertical space in your levels.

We placed the spawn points high above the playing field so that players would have a chance to survey the map, orient themselves, and scope out their opponent before choosing to jump back into the action. I’m actually surprised more arena based first-person games don’t do this, we’re very proud to have implemented it!

I'm a software engineer by trade, but I do a lot of ceramics. I've noticed that when I work on a project in clay for a long period of time I begin to perceive the world around me differently, sometimes as if it's literally made of clay. I know. It's strange. Does something similar happen to you given that you spend so much time thinking about light and darkness in 3D space?

Coincidentally, I’m taking a ceramics class right now! So I totally know what you are talking about, after working in clay everything sort of feels like it has that texture, and you can imagine molding and shaping everything around you. Chambara’s play of light and shadow is bizarre and abstract enough that I don’t see it in real life as much, but I will draw arrangements based on crawling shadows if I see something interesting enough in real life.

What do you see yourself working on after Chambara? Still doing level design for combat-oriented games?

The whole team has learned so much from Chambara’s development, and all I can say is I’m so excited to take what we’ve learned and blow everyone away with our next thing. We’re just getting started.

What game have you been playing lately?

I’ve been playing a lot a Metal Gear Solid V’s online mode… it’s been out for a long time but I’m still having fun squeezing in matches when I have free time. I really like having an intricate character with deep controls, and it feel comfortable to just run around in that skin.

Final question: How excited are you for the return of Samurai Jack sometime this year?

I’m pumped to see more from Genndy Tartakovsky! He’s a master of pacing and movement, and obviously an inspiration.

Chambara is slated for release in 2016. Be sure to check out previous articles involving Chambara from Chris Baker and from Team OK's project lead, Kevin Wong.