I discovered Killer Queen inside a dark, stuffy IndieCade tent pitched on a hot black parking lot. What seemed at first to be a typical pair of four-player arcade cabinets turned out to be one custom-built 10-player beast of a machine that pits two teams of antfolk against one another in surreal Joust-esque combat.

With multiple win conditions, character classes and harvestable resources, Killer Queen -- nominated for Excellence in Design and honorably mentioned for the Saemus McNally Grand Prize in this year's IGF -- is trickier than it looks.

When I finally stumbled across a second Queen machine months later, as part of a GDC exhibit, I invited friends to play without teling them anything about the game's depth and complexity. I wanted to win, of course, but I also wanted them to laugh and shout as they discovered the game for themselves.

Here, Killer Queen co-creators Joshua DeBonis and Nikita Mikros explain why they're fascinated by local multiplayer game design, what other devs can learn from their work in the field, and how Killer Queen grew out of a field game played with spools of string.

What's your background in making games?

We’ve both done a good deal of freelancing and contract work for various studios around NYC. We met each other while we were each doing work for Gamelab in the mid-2000s. We’ve been running our own indie studios, Tiny Mantis/SMASHWORX and Sortasoft LLC for years, but recently decided to join forces and start a new supergroup studio together called BumbleBear Games, LLC. Our focus is on local multiplayer arcade games, which we love making and playing.

What value do you see in local multiplayer arcade games, and what design advice would you give other developers based on your experience?

Almost every day we hear from new people who tell me about friends they made through playing Killer Queen. Local multiplayer games are truly social. Because Killer Queen is only available at certain places, you have to make a distinct effort to go there and meet people.

Our design advice for other developers making arcade games is to think about, and to iterate, the entire experience, from when players first see the marquee until the end of a game. That entire experience needs to be thoughtful, and it works differently than most other types of games.

What development tools did you use to build Killer Queen?

The software is built in Unity. We also used Tiled as a level editor. For hardware, we have an Ubuntu Linux computer. Each cabinet contains an I-PAC for input and a Pac-Drive to control the lights.

How much time have you spent working on the game?

We made the field game that inspired it in 2011, and then started making the arcade game in 2012. We’re still putting some of the final touches on it.

How did you come up with the concept?

We were working on a new field game for the Come Out & Play festival, and wanted to explore something with a complex system.

We specifically chose to base it off the social behavior of ants. Originally players would run around a field and leave “scent trails” made with string to hidden caches of “food”. The string ended up too messy and complicated so we cut it but we kept the title Killer Queen.

Why did you feel compelled to explore complex system design, and what lessons have you learned in your years working on KQ?

The collaborations that we did prior to Killer Queen, (Pigeon Pinata Pummel and Pitfall Live at the Tank) focused on spectacle and the visceral experience. The rules of these were interesting, but only in the simplest ways. We wanted to expand ourselves, keeping the spectacle but adding greater depth to the gameplay.

We've learned that complexity in the overall structure of the game doesn't prevent casual play, as long as the controls and moment-to-moment interaction is simple and intuitive.

I've only played Killer Queen cabinets at IndieCade and GDC; why choose to build this game in an arcade cabinet, and what interesting challenges did it present?

Our original setup used 12 NES controllers and a projector. It was really fun, but the game was unreliable and unwieldy. It become clear that it was necessary to have a more controlled environment if the game were to be to able stand on its own without us being there to maintain it all night. Additionally, the game was created as a game for public spaces. We felt arcades might be a viable space for this type of gameplay.

Have you played any of the other IGF finalists? Any games you've particularly enjoyed?

We haven’t played very many of the other finalists, but we’re really looking forward to playing them all. We think Icebound is really interesting, both narratively and experientially. How do you Do It? is cute--we love that it’s so short and sweet. Rooftop Cop is a fascinating exploration of games and their messages.