I wrote the following retrospective as a blog post on my personal website after I submitted my prototype of Routine Patrol as my entry for Pirate Studio Game Jam 15. I figured I should post it here as well.

This was written before I received any feedback from anybody. I didn’t get very much feedback, but the feedback I did get was very enlightening.

What went well

I built a mostly functional game with Godot for the first time, despite not having used it for a project before. I was productive with Godot despite my unfamiliarity with it, and more productive than I would have been with other approaches I had used previously.

I was able to control project scope and maintain focus on the idea I came up with.

Deliberately not spending every second of the jam implementing something gave me space to consider a larger vision that resulted in a roadmap for the next few months, if I want to stick with this project.

What did not go well

I did only work about 5 days of the 14 that made up the jam. That is an underestimate; I thought about the game every day, and if I pull the commit logs I’m sure I committed on at least 7 days. But I did the heavy lifting on the weekends. That’s okay, but I needed to do more in the middle of the week.

Compiling Aseprite was a bit of an annoyance in the middle of the jam.

I did not complete the entire functional scope that I had planned for the jam. In particular, I missed out on implementing intruders and alerts; two features that I think would be vital for actually realizing the “anti-stealth game” concept.

I fumbled around quite a bit on the first Saturday with assets.

What would I change

Find a way to commit something useful daily, no matter what. Or work less at my day job.

Settle on the toolset in advance of the jam. I was enamored with the idea of having no preconceived notion of what I would do; if I instead committed to using all of the tools I had used in advance, I could have saved quite a bit of time.

Make the game scope even smaller than I did this time. The potential of the idea’s mechanics should be should be obvious in builds after the first full day of implementation.

Pre-bake generic assets, like tile sets, prior to the start of the jam. That way you don’t have to mess with colliders and such.