I Learned 4 Things From my First Ludum Dare

I've done my first Ludum Dare Jam now, and actually my first game jam of any kind. Wow! I am so happy to have finally done this. It was a super rewarding experience and I want to share that, and my game, with as many people as will listen.

My game is Patient Out Of Time. It is an apocalyptic moody shooter about a doctor salvaging power sources from robots in the wasteland to keep the life support of his last patient running as long as possible. The hospital staff have all left, and they are the only two survivors. Keeping this man alive is all this one doctor has to keep him going.

It is a sad game, but it was also a lot of fun to make.

Here are some things I learned this time. I hope to learn more things the next Ludum Dare.

Little Steps Make Safe Steps

I didn't have time for broken builds or half-built code I needed to fight my way back out of just to get the game running again. Every change I made had to be broken down into tiny, discrete non-breaking changes. Every step of way had to be playable. This kept the game constantly in a "technically releaseable" state, which kept stress about finishing the game off my back.

Refactoring Can Be Treading Water

My habits as a developer tend towards building systems. Now, I get a lot of enjoyment out of this and preach the merits of systems as code design, but I'm trying to learn to cautiously apply this form of what is, some times, over thinking things. So, I did my best to permit myself to write "bad" code and move forward.

I didn't have a lot of assets, so as I added them one by one through the process I never built any kind of asset management. That's what old Calvin would have done. You know, to "clean it up". Instead, I just added what I needed to make the new thing work, because spending time to change big things would do two negative things:

First, it would violate the first rule: Little Steps Make Safe Steps. Refactoring is a great way to get lost in the weeds with a half-completed bit of work that'll take you hours just to get the feature set back to exactly where you started. No thank you.

Compromise When You Find a Dead End

A lot of problems we come up against as software developments make the little voice in our heads say "Oh, I know, I'll just..." and then, hours later, we're still struggling with all the pitfalls and unforeseen problems with what we thought would be a totally simple solution.

When you see this, don't forget that you can give up. And I mean that in a good way, because some times it just isn't worth it.

As an example from Patient Out Of Time, I wanted to make the robots chasing you avoid the problem of "clumping" too close, which was common since they all just headed straight towards you. I started experimenting and thinking about different kind of flocking algorithms and coordination between the robots. It was all turning pretty complicated!

Instead, I backed out of all that and just randomized all their speeds a little bit. Problem solved with one line.

Add a Little Bit Of Everything

I had 48 hours. Technically, I had 72 hours, because I'm doing the Jam and not the Compo. However, I do have to work on Monday! And I have a family, and I try to avoid burn out. So, really, my time to put into this was pretty limited. Still, watching the clock, I was sure to rotate my efforts between code and art and audio and design.

Evenly distributing the effort across the different pieces that make the title contributed to that "always releaseable" goal. I didn't wait until the very end to figure out sound. I iterated on my art and animations interlaced with feature tweaks and bug fixes. Everything grew up together.

This also meant I got practice and new experience with everything. I did some audio sample editing. I worked on my pixel art animation skills. My skills with the Love2D platform I've been using were improved a bit. Every muscle got a little exercise.

Have Fun

I highly recommend trying out Ludum Dare some time. If you do, don't take it too seriously. Have fun!