YettiBebbis: Puppet in a Cult has players investigate a cult with a puppet controller

The 2023 Game Developers Conference will once again feature Alt.Ctrl.GDC, an exhibition dedicated to games that use alternative control schemes and interactions in new, exciting, and clever ways. Ahead of GDC 2023, Game Developer will be talking to the developers of each of the games that have been selected for the showcase.

YettiBebbis: Puppet in a Cult will have players trying to blend in with the dances and rituals of a cult, using a puppet controller to mimic the motions of the creatures around them.

Game Developer sat down with the team behind this intriguing puppet controller to talk about the thoughts that went into designing a game that really made use of the puppet and puppeteering, what drew them to create such a cute cult, and how the complexities of puppeteering lead to several vital design changes and possibilities.

What’s your name, and what was your role on this project?

Tingyu Yan: Artist and designer.

Yile Xu: Control mechanics co-designer and programmer.

Rei Yamada: Character artist.

Yuan Suo: Producer and game design as well as control mechanics design.

Jiacheng Qiu: Game system programmer and designer.

How do you describe your innovative controller to someone who’s completely unfamiliar with it?

Suo: Imagine you are controlling a physical puppet using your fingertips. It is exactly like that, but with a virtual figure. When you lift up a finger, the arm of the puppet moves accordingly. To make it realistic, we attached appropriate weights at the end of the wires so the experience feels authentic.

What's your background in making games?

Yan: I have a background in industrial design and interface design. My past projects include interactive installations, toys, and board games.

Xu: I have a background in level design, programming, and physical computing. My past projects include video game levels, generative arts, and alternative game controllers.

Yamada: Ever since I joined the graduate program [at the] Entertainment Technology Center, I have been involved in making games as a 3D artist for the past few months.

Suo: I have made several games since graduating from college. I have one game that is well-reviewed on Steam, and I also have experience working on level design for mobile game companies. Right now, I am mostly making games in VR or interactive experiences such as this.

Qiu: I have been making 3D games since I was in undergraduate university, and was mainly focusing on programming and music composing.

What development tools did you use to build YettiBebbis: Puppet In A Cult?

Xu: On the programming side, we used Unity as our game engine and Phidgets as the physics computing platform.

Qiu: We were mainly using Unity Game Engine to create the game. Some professional tools to help development include the Phidget platform for retrieving inputs from our controllers, Adobe AU for sound design, and Unity built-in Shadergraph for visual effect creation.

What physical materials did you use to make it?

Xu: Sensors-wise, we used distance sensors for local gestures and acceleration sensors for global movements.

Suo: in terms of controller materials, we are using high-density cotton thread as the puppeteering string, wood blocks to build a structure that supports the strings, finger glove for finger protection, and weight blocks at the end of the wires to make the experience feel more authentic.

What inspired the creation of YettiBebbis: Puppet In A Cult? A game about using puppetry to blend in with a cult?

Yan: We nailed down the idea of exploring the physical interaction behind puppeteering first. We spent a lot of time on brainstorming and researching to find a story that could bring this puppeteering experience into a digital form with a sense of risk.

Suo: I’ve always been interested in exploring physical senses and I had this idea of making a puppeteering game during brainstorm sessions for previous games. The cult idea is more of a story/design choice based on the game control.

What challenges did you face in turning puppetry into a game? Into turning the gestures of puppetry into a controller?

Xu: One challenge is to determine the hardware platform (Arduino? Phidgets?) and sensor (which type of distance sensor do we want, IR? Supersonic? TOF?). The other challenge was smoothing the input data for character control and correspondent movements.

Suo: The challenge is the mechanics itself. If you are using fingers to control a puppet through wires, the best case is the puppet reacts exactly depending on the distance your finger travels. In the beginning days, we were thinking about making a puppet combat game where lifting a finger up would make the puppet perform an action such as a heavy punch. This is an interesting game play/control, but it loses the sense of puppeteering because the distance your finger travels does not matter to the game, necessarily. In this case, the action of lifting is similar to pressing a button.

Eventually, we realized that the authenticity of control and how the game puppet reacts to your control is the most important part of the mechanics. Because of that, we decided to make a game that is all about doing simple gestures so that the lifting of fingers makes perfect sense.

What thoughts went into creating gameplay around what the puppet could do? What drew you to the idea of mimicking cult rituals?

Yamada: In the ideation process, we were interested in the silliness of having the creatures do the same movements while players struggle to follow them through. We all wanted to create a fun game with a unique theme, which somehow led to the idea of combining uncommon themes together. We had more ideas about what the puppet could do, but we had to cut down and simplify the capabilities of the puppet due to the time constraints.

Suo: Because we decided that the best case for this game was to have the puppet react to player control, we started brainstorming game ideas where it required the player to control the behavior of the puppet relatively freely. We thought about dancing and other similar themes, but eventually we all laughed at the idea of "What if you’re in a cult and you have to copy exactly what the NPCs do?" So, we went on with this idea and made a cult-mimicking game.

What thoughts went into balancing the game? Into making it challenging enough that players had to work the puppetry controls properly without making it too hard for someone likely trying to work a puppet for the first time?

Suo: The game is firstly designed to be a single player game where the player has to control the left controller for the puppet body position and the right controller for the limbs and head gestures. We found out during larger playtest sessions that this is challenging for most of the players simply because they have to learn two things in a short amount of time. So, we decided to have two players play this game as a co-op experience where one player only controls the body position (left control), and the other person controls the gestures (right control).

This is a lot easier and more friendly to most players because one can focus on learning one specific mechanic at a time. We also made the game very forgiving in the beginning so players can pick up their mistakes. Having two players playing this worked out very well in the new playtests, and is likely going to be how we set up this game at Alt.Ctrl.GDC. However, we do expect GDC players that desire more challenges, so playing solo is also encouraged.

Qiu: Balancing was a pretty large problem for us as we are using an intuitive, but still new, input system. While building the main mechanics, we developers have tested the gameplay tons of times and understand all the mechanics behind it, making the original prototype really difficult...

But, as we don’t want to frustrate players, we have tried to iteratively make it easier by first adding a tutorial stage (which is the first socializing stage right after calibration), aiming to let players try to mimic the moves of the NPCs, and get used to moving around.

We also adjusted the level of difficulty for each move separately. For example, the duration before the system thinks the player is suspicious is one second for regular hand waving and two seconds for jumping, which is because jumping requires larger movement on the left hand than just moving fingers. The percent of suspicion charged each second was adjusted multiple times based on first-timers’ feedback. Also, during our public playtesting we found out that somehow having two players control the character might be easier, and based on that we now support having either single player or two players cooperate through the game.

The cult looks pretty cute. Why did you make the cultists into adorable creatures? What do you feel that added to the experience?

Yamada: The word "cult" generally gives an impression of mystery and eeriness. When I think of the word "cult," I somewhat associate it with the image of the dead and mortals dancing in circles in fear of death and madness. Believing that juxtaposition creates something unexpected or new, having creepy and cute at the same time brought a unique worldview to our game that is not entirely scary nor bubbly.

Has building a game around a unique controller taught you anything unexpected about game design?

Suo: Of course! In my opinion, I’ve learned quite a lot in reiterating the controller versions to make sure that the guests can learn it quickly by themselves. We also needed to think about the operation side of things because when hundreds of people are trying to play this game, the wires can be easily broken. So, optimizing the design to make sure the experience can run as long as possible before maintenance is also cool to learn.