Devtober 2021 Postmortem

November 1, 2021 · game dev

Over the past few months I have been working on being more consistent and form regular habits around my side projects, interests, and hobbies. This ended up lining up perfectly with #devtober, which is a month-long challenge to spend every day of October working on a game.

So for October I set out with the goal of finishing the Nintendo Switch port of my game Double Turn.

Admittedly I didn't make it through the entire month doing game development every single day. I spent only 24 of the 31 days of October working on game dev. Considering I already only have time to work on game dev part time, having a day off here and there is healthy so I'm totally happy with 24 days. Honestly if I could hit 24 days every month that'd be amazing.

My approach

My number one goal was consistency and my number two goal was momentum. The biggest thing I've struggled with doing game dev part time in the past has been being really streaky in my work habits. I have 2-3 really hot weeks where I'm working nearly every day for 1-3 hours, then have 2 months until I work on stuff again. This is especially true working on Double Turn which I've worked on off and on for so long it can sometimes feel like opening up that project is a huge task in itself.

So I made sure I my days looked like one of the following:

  • Complete a small task
  • Contribute to a larger task for 30 minutes

That's it! I also tried to keep my time working under 1 hour a day, so I could more feasibly get keep the momentum going, while still having time in my day for other things. Most days I worked for less than 30 minutes and knocked out a small bug or task. Probably only 4 of the days I worked for more than 1 hour.

What went wrong

The main thing that went wrong I didn't really solve until the last week. While most of the work I was doing was very small in scope and I had a well organized backlog of bugs and tasks to do, some of the issues I ran into took a few hours to investigate or come up with really good solutions. I kept losing context on where I was and there were a few moments where it felt like I had ping-ponged between two similar bugs for a few days fixing one while re-breaking the other.

My first gut reaction to this was I had to find the time for longer sessions lasting 2+ hours to do those tasks. That worked but it went against my 1 hour max policy. I knew for me that was only sustainable occasionally, so I shouldn't rely on that otherwise I'd burn out or get back into streaky habits.

Therefore, towards the end of the month I started doing a few tricks to make splitting up that work easier. First, taking notes as code comments exactly where I left off the previous day, little breadcrumbs to help me find my way back quicker. The second was using my 30 minutes to just do planning and debugging. This seems almost obvious in retrospect but it was so easy to fall into the trap of chugging along without stopping to think. A lot of my other tasks were much simpler and easy to knock out, I got used to that and expected everything to be solved with a half hour and some elbow grease.

What went right

The biggest thing that went well was that my momentum never really slowed. While I did skip the occasional day it was normally because I had other plans or just wanted a break! The next day I kept it going like nothing changed.

Additionally, because I was working on the game a little every day it was a lot easier to keep track of what I was working on and what needed to get done next. It forced me to keep a clean backlog with upcoming work ready to go in small actionable tasks.

Finally, I had a lot easier time forcing myself out of slumps, becuase I never had time for any! Imposter syndrome can hit me pretty hard and I've been working on Double Turn off and on for nearly 5 years. Every long stretch of time I don't work on it, it gets harder to get back into it. Bugs seem further and further out of my ability to fix. Features seem more and more difficult to make. When I had to just grab a task and finish it in a half hour, I knew I had little time for my silly brain to overthink it.

Takeaways

I think going forward I am going to try to keep this 30 minutes a day structure going. I think there will definitely be times where I need to spend a bit more time on something chewy, but what I've noticed this month is that's generally all I need for most things and it helps me stay fresh and the ball rolling. Given that Double Turn is an existing game with different needs and an established backlog of tasks, I am excited to try this approach with a new prototype to see how it transfers to something brand new.

I also didn't fully hit my goal of finishing the port, but I got really close. Working on it consistenty nearly every day was the focus I needed to get it done and I plan on rolling with this momentum to do the remaining work.

Double Turn running on the Nintendo Switch
Double Turn running on the Nintendo Switch
Kenny Goff
Kenny Goff © 2011-2020