I'm a Grown UP !! Dreamy Sweetie The Devil Slayer (2023): Following Quest Chains Through the Mist of Design

Toby Alden on the merits of creating a game from an vague and elaborate conceit

Find the developer elsewhere: (Twitter, Blog, Cohost)

I'm a Grown UP !! Dreamy Sweetie The Devil Slayer (what a mouthful!) is a side-scrolling roguelike about completing a convoluted quest chain to kill Satan. The player starts in Sweetie’s bedroom, and can enter the dream world by going to sleep in their bed. Dying results in waking up and having to start again, unless you've reached a dream within a dream... but we’ll get to that later.

In his book A Theory of Fun for Game Design, Raph Koster talks about “core loops”, the second-to-second actions that you build your bigger game loop out of. For instance, in Super Mario 64, the core loops are Mario’s movements: running, jumping, diving... all the ways he can be set into motion before inertia brings him to rest again. These atomic units of gameplay become the building blocks for larger loops, such as traversing worlds and collecting stars.

While many games repurpose this concept to sinister ends by engineering compulsive behaviors (loops by their nature), the original idea is a useful one, and it’s how I tend to make games. I start by making a little guy move around in interesting ways, then give them an interesting world to move around in, and finally give them some interesting things to do in that world. This has worked pretty well for me, and I like having a guarantee that the core loop of movement is fun, giving me a solid foundation of kinetic joy to build on top of.

The downside of making games this way is it’s easy to get to the part where you have a little guy moving around in interesting ways and then become daunted by the prospect of creating a dozen bespoke challenges for them. This happens to me often.

In my mind, the biggest hurdle of making games is that they’re a lot of work, and sustaining your enthusiasm throughout that work can be very difficult. The anticipatory exhaustion described above is a big contributing factor, and could be rephrased as a form of fatigue that happens when it feels like you’ve made all the creative decisions already, and left your future self only busywork. To be certain, the task of creating a dozen bespoke challenges will entail creative decisions, but they’ll be of a secondary importance to the foundational decisions you made creating the core loop - and it’s too easy to fixate on the “dozen” part.

Put another way, creative decisions are like the breadcrumbs along the path of game development that give you energy to keep going, and the flaw with this “build up loops” method is you front load all the biggest breadcrumbs. No decision you make after making your little guy move around is going to have as big an impact on the moment-to-moment gameplay, which makes it easy to succumb to that feeling of exhaustion.

All of these thoughts were percolating in my mind when I started thinking about making Sweetie Dreamy, and culminated in a stoned notes app session in the bath that produced this:

Nearly all of this ended up in the game as is! But at the time, I didn’t really know how most of it would work in practice. I didn’t know how you’d defeat Satan; I didn’t know how pot recursion would work; I didn’t know what dream recursion was, other than something that sounded cool. I just described a fake video game that I’d have liked to exist and then tried to make it real.

And as a result, this game was really fun to work on the entire time! I was constantly having to invent answers to questions that would come up when I tried to form these vague ideas into an actual game. It felt risky and exciting; my loose outline acted as a trellis, but the game grew organically around it. Problems often acted as solutions to each other; for instance, the question of how to defeat Satan was solved by the question of what purpose dream recursion served, as I hid a Satan-slaying sword at the deepest level of Sweetie’s subconscious. I felt in conversation with the game as it grew, rather than its impersonal architect.

Like I said earlier, I do think this approach is riskier! It takes longer to see if your ideas work, and it lacks that “core loop” fun foundation. But once you’ve made a few games, the iterative “build up loops” approach can become rote, and a process that should bring joy and opportunity for surprise becomes an exercise in making something that’s simply functional. Once the fundamentals of game development are in your comfort zone, you have to wander off again to continue growing. This approach - the loose quest chain, a GameFAQ for a fictional game, a narrative you’d like to experience - is the closest thing to a map for those wanderings I’ve found. It’s got a destination and a path, but enough smudges and torn parts to ensure you get lost, have a memorable journey, and tell a good story afterward.

Design bounty: The world is ready for a new age of Sweetielikes - games made of surprises to discover and share, whose purpose is hidden, that move constantly forward and nest like matryoshkas. 

Consider taking the Volume 1 reader survey!