Concept

This is the first post in a dev log about Spilled Mushrooms, a turn-based card solitaire puzzle game available exclusively for Playdate.

Principles

Every game starts with a concept. To find mine, I laid down some design principles that emphasize my favorite part of the Playdate: its extreme portability. In my opinion, games excel at hyper portability when they have:

  • Little to no narrative: lots of narrative means lots of important details that can be missed when playing in busy environments
  • Highly interruptible: games that require remembering a lot of context or that have real-time actions can also be difficult to focus on in busy environments
  • Quick play sessions: longer sessions are more difficult to weave into the small idle spots of daily life

It's important to note that these principles are't essential to have in Playdate games! These are just the principles I wanted to develop with. Many Playdate games are heavily narrative-driven or feature longer gameplay loops.

Using these as guides, I started making some concrete decisions:

  • The story will be purely thematic. Beyond trying to avoid players missing details, I also wanted to design a game that I could enjoy playing as the omniscient developer. If I wrote a story, I feel like I wouldn't be able to enjoy it properly. I also didn't have a particular message I wanted to convey or a particular set of emotions that I wanted players to experience. Most importantly, I didn't want to write a story or narrative. So I just decided not to.
  • The gameplay will be a turn-based card solitaire. Turn-based games are substantially easier to pick up and put down than real-time games, while the solitaire aspect will avoid the need to invest time in crafting a strategic deck. I use the genre of solitaire here to mean that each deal is random and you have a win condition with no opponent and no reactive system. This is in contrast with a lot of single-player turn-based card games like Hearthstone's PvE adventures, Inscryption, or Slay the Spire, where the opponent is computer-controlled, even if the gameplay may be asymmetric.
  • Challenges will be small enough to memorize, but large enough that they can't be brute forced. With a small puzzle, you inherently have a smaller gameplay context to focus on. I really aimed for challenges that could fit entirely on one screen and be played through in a matter of minutes. One of the difficulties with designing this type of puzzle is making sure that the player has enough options available to make the game tree balloon up, but not so many options that they're stuck in analysis paralysis for longer than I want the entire puzzle to take.
  • Challenges will be randomly / procedurally generated, but gameplay will be deterministic. I considered a design like Hearthstone's PvE adventures at first, where the challenge is designed but the gameplay has a substantial random element to it, but I was turned off by memories of spamming retry against Heroic Naxxramas bosses when I didn't get the exact opening hand I needed. With random / proc gen challenges that play out exactly the same each time, I also wouldn't need to handcraft a fixed number of challenges, which would provide a huge amount of replayability as long as the game is fun.

Prototyping

These are the four ground tenets I used for the game's design. With these in hand, I spent a sleepless night brainstorming. Even though the story isn't a major gameplay driver, I still wanted a concept to theme around. I remembered how Shigeru Miyamoto was inspired to make Pikmin by a line of marching ants, which gave me my own ant-spiration. I imagined using an army of ants setting out to conquer various regions by building an anthill in each of them. This formed the idea of playing cards from your hand to various areas.

I'll save the major influences for the next dev log, but a few of them led to the idea of an 8 card deck. It's small enough to memorize and could probably fit on one screen, but playing one card each turn would have at least 8! = 40320 orderings, which is too many to brute force. From there, I formed a gameplay loop: draw two cards, then play one and discard one. When you run out of cards to draw, shuffle the discard pile and repeat. When you run out of cards to shuffle, the game is over.

The next day I was able to take those ideas and turn them into a functional text-based prototype:

Prototype screenshot

The idea of building anthills turned into destroying anthills before the prototype was even playable. However, the gameplay was essentially the same as it is now: remove all of the resources from the field by the end of turn 7. Each turn, you pick one of two cards to play at one of three zones, and then the cards you've played will collect resources from the zone they're at. However, this was still very much a prototype and a lot of the modern game was still very much missing or in an early stage:

  • Inscryption's influence was so strong I was blatantly copying its cards. Several cards and regions have been renamed, eg: Stinkbug / Scavenger became Raccoon / Scrappy
  • Spells! These often felt quite miserable to use in practice as they usually had very little impact and occasionally didn't even do anything because you had no cards on the field to target. They were eventually replaced with similarly functioning animals: Growth became Moose / Inspiring, Nurture became Elephant / Nurturing, and Scoop became Hippo.
  • Every zone started with 10 resources, with traits like Abundant and Barren adjusting the starting amount rather than affecting animals played there. This meant that not every puzzle was solvable, and even when they were solvable, you almost never had the satisfaction of an exact solution like the modern game has.
  • Difficulty was selected by choosing the number of zones. You could play "easy" with two zones, "normal" with three zones, or "hard" with four zones. Easy was usually brainless, normal was occasionally solvable, and hard was effectively impossible.
  • Your deck was shuffled every attempt. That is, you could play the same deck, but keep retrying until you got a better shuffle. Your deck was also shuffled at the start of turn 4. This really didn't jive that well with the deterministic gameplay decision I had made earlier, so it was fairly shortlived.

Even though it was a prototype, it captured enough of the game's essence to validate that this could be a fun and engaging game, which is all that I needed to keep pushing forward with development.

I think there's always some good fun to be had from looking back at the earliest prototypes of something and seeing how far it's come from the original design. Nothing starts perfect.