The Postmortem

Beginner's Maze is the first game I developed and published on my own. It is a tiny adventure puzzle made with the Bitsy engine.

In the present devlog I'll share some insights about my creative process and learnings I got from this experience.

Ideation

I chose Bitsy on purpose, to limit my options and keep everything (concept, storytelling, design) as simple as possible in order to quickly deliver the game.  I gave myself a week to complete it and submit it to a game jam. But, as expected, I struggled with the simplicity rule. 

My original intention was to build an interactive fiction. I wrote some dialogs, sketched some scenes but I didn't storyboard the whole thing. I quickly jumped to Bitsy and got lost designing rooms, adding animations and conditional dialogs. The lack of core definitions soon got me stuck. I did a couple of prototypes but none convinced me. I lost some momentum there and stopped working on the game for 1 or 2 days. The interactive fiction was too complex for that moment so I decided to throw it away and start all over.

Making a maze puzzle was in my backlog of future game ideas. A puzzle would allow me to be less strict with the storytelling and less attached to details. 

I started sketching maze patterns and soon realized it would be trickier than expected. I wanted the maze to be real, with a solution path and correct room transitions when exploring it. A quick Internet search lead me to a handy online tool called Maze Generator. It provides many configuration parameters to create a maze of various shapes, sizes and complexity.

I used one of the generated designs as the base of my game. I split it into 4 sections, each representing one room in Bitsy. I used placeholders to mark the connections between rooms and the final exit. From that moment on, everything went smoothly. In just one day I built the whole game, which included two prototypes and the final version.

Prototypes and Final Version

Below is a summary of the iterations the game went through.

Prototypes 1 & 2

• Constructed the maze in full.
• Used placeholders to mark the exits.
• Added a few interactive objects.

Final version 1

• Added Start and End screens.
• Changed business logic: interact with objects instead of collecting them.
• Changed the color palette.

Final version 2

• Changed the design of the Start and End screens.
• Changed contextual texts and name of the game.

Conclusions

What worked

• Using a maze generator. Originality wasn't a priority in this area so I went to the quickest solution.
• Keeping the maze fairly small. My original design had 9 sections instead of 4.
• Using placeholders to mark the exits between rooms. This made it super easy to add and test the connections.
• Sticking to a concept that I liked (despite all the back and forth I mentioned in the beginning paragraphs).

What didn't work

• Researching too much. I spent a considerable amount of time playing other Bitsy games, which was valuable for me so as to to know the capabilities of the engine and the creative solutions used by other developers. But this also made me a doubt a lot about my current 
• Changing the concept frequently. Related to the first item.
• Not following the "keep it simple" rule.

Lessons learned

• Sketch the screens and storyline before diving in the implementation.
• Don't waste time on details.