After being inspired by the most recent Game Devs of Color Expo, I wanted to get back into development despite having been away from learning it for many years. One piece of advice echoed by many of the indie developers and speakers there, was to take part in game jams even if you were a beginner, so I decided to do exactly that.
Immediately after the conference, I started looking for upcoming jams, and Virtual Turtle Games’ 2026 Cozy Fall Jam caught my eye. It seemed straightforward enough; 3 days to make a cozy game with the themes “Fall” and “Roots”. Surely I could make that happen even with my limited skill set.
In fact, to ensure I could *actually* make it happen, I kept the project small in scope. Most of my original plan for the game made it to the end of the process, and I only made a few aesthetic/visual cuts that didn’t affect the core gameplay. Features like falling leaves, dynamic weather, and day-to-night transitions that started off as “nice-to-haves” ended up as “I don’t really need it for this project” when I realized how much work would go into figuring it out. I didn’t have a lot of time to spare so, unfortunately, they had to go.
Another important decision I made early on was to abandon Blender when it became obvious that I would spend too much time struggling with tools (and trying to build custom models) rather than on building the game. It didn’t help that the program kept crashing on me before I even had the chance to attempt a single model, so asset development will have to happen another time. Switching to free asset packs gave me enough time to focus on level design, atmosphere, audio, and creating an ending I was happy with.
I spent way too much time navigating Unreal’s Blueprints, UI systems, packaging requirements, and engine-specific quirks. As frustrating as it was, in a way, this turned out to be a good thing because I learned a lot about how different systems intersected. I also gained newfound love for the foliage brush (after all the hiccups, I definitely did not have the time to place a bunch of individual assets across the map).
Working on this project also allowed me to practice important development principles: commit before doing anything risky, time-box any extra features, pivot early when something isn’t working, and most importantly, assume that things will take longer than expected. Nearly everything took between one and a half to two times longer than I originally estimated.
All in all, I think this project was a success. My biggest win was that I completed the full development cycle, from concept to published game, in a single weekend. There are areas that I want to improve in next (specifically my Blueprint fluency), and gave me the confidence I needed to reassure myself that I can take an idea, manage the scope, adapt when things go wrong, and still deliver a finished game.
Leave a Reply