Tools & Tech

A Social Network was written in Phaser, a javascript game development framework. I used MAMP to host the game locally during development. I wrote all of my code using github’s Atom. Art assets were designed and turned into sprite sheets using Pyxel Edit, picked up for cheap in a humble bundle. The rather pitiful SFX I scrounged up were generated using BFXR. LICEcap was the real MVP for incredibly easy gif capture and recording to share on twitter.

Game Design

The theme for Ludum Dare 38 was “A Small World”. The first place my mind went (and many others’ if you look at the game results) was to create a small planet. The other angle that I felt was common was the social implications of the theme. I had a brief image of a social media creeping game, but settled on modeling the connections of a growing social network amongst people. The appeal of a creating a small planet with a nice starscape lived on as a decoration element.

What Went Well?

Sharing progress a lot in Slack, Twitter, Discord helped me stay motivated. Sharing my progress along the way got me some small tips and thoughts about the project. Seeing other people interested in my game the game helped me keep up the motivation to finish the weekend with a product. Being able to record gigs easily was a big boon. Sharing is so much easier when it's so intuitive and painless! The people at Itch.io wrote an article about other easy gif creation tools.

I made it a point to have something super minimal running by first night (stayed up too late though). In addition, I had chosen an artstyle and had a mockup to work off.

Since this was my first solo game jam, I wanted to keep my scope incredibly small. Thanks to this, I managed to finish my MVP by the afternoon on Saturday. This gave plenty of time to polish the game. (You'll see I didn't use this time very well... I talk about that in the next section!)

Concept Art!

Very basic MVP

Waking up early on Saturday and getting it of the apartment to hang out at a coffee shop was great. It really helped to not just sit at home in front of my monitor for the entire jam. Getting outside and working in a different place helped me out with imagination and creativity! Especially since my theme was social, being out with other people around worked out nicely.

What Went Wrong?

I did not come up with a good solid game mechanic. My theme interpretations were not anything special and didn't have a central mechanic to revolve around. In the past when I've done jams with other people, it was easier for us to collectively arrive at fun ideas. It was much harder without being able to lean on others. Creativity is a skill that I need to practice more. Raph Koster has some interesting points on the subject: https://www.youtube.com/watch?v=zyVTxGpEO30

Creating SFX and music is hard for me. VERY hard. I knew this and left the second half of Sunday to work on the subject. But this was nowhere near enough time to actually learn how to create decent audio assets. I ended up hitting randomize a hundred thousand times on BFXR and ditching the thought of having a sound track. This continues the recurring theme of completing a solo game jam requiring you to have knowledge of skills outside of your forte.

The core idea of my game sounded interesting, but didn’t pan out very well. I did not have a clear goal for gameplay in mind which made it suffer. On top of this. my implementation ended up being a bit buggy and annoying to play with. I ended up adding some sandbox features like skipping between levels and adding endless amounts of people to try and make it a little interesting.

I left myself plenty of time to sleep this time around, but I ended up not sleeping at hours similar to how I normally do. This had me waking up too early in the morning after staying up late. It's not just getting enough sleep that matters, you’ve got to consider your normal sleep schedule too!

Conclusion

Working alone is harder than working on a team. You really find out what your personal weak points are. I didn’t have much appreciation for developing skills outside of my wheelhouse. Its very easy to discard other skills when you’re working with talented people who can handle those departments, but a solo jam is a whole different animal. To succeed at a solo game jam like Ludum Dare you really need to practice all facets of game dev! You can play the "finished" product in your browser here https://law5guy.itch.io/a-social-network.

Friday

The theme is "The More You Have, the Worse it Gets"! Of the other top themes, this was not one of the ones I was very excited about. I spent the most of Friday night brainstorming to try and come up with a game idea. This took me hours.

After that I got a boilerplate project set up and running here.

Saturday

Later in the day I managed to get Draggable components working and had some frustrating fights with the CSSTransitionGroup component. I tweeted what I managed to get done on Day #2.

I gave up on getting the new react-transition-group library in favor of the old react-addons-css-transition-group library. I figured that I didn't want to spend too much of my time trying to figure out an animation library when I already knew how one worked.

My final sin of the day was storing React components in the state of my React app. Fortunately, I've now learned why this is a terrible idea and won't ever have to make that mistake again!

Sunday

The final day of Ludum Dare! Today I actually had most of the day to wrap up my project unlike my busy day yesterday.

The best debugger ever made is a good night's sleep.

I saw the above on Twitter and it is truly the best debugging advice I've ever received. For whatever reason all of yesterday's problems with my React state and undefined variables became clear. Today I ended up creating the vast majority of the game's features, spawners, consumers, and the hdd/storage space mechanic.

In Progress GIFs:

My final piece of experimentation was deploying my first React app. Create React App comes with a built in yarn build script which creates your build files in the /build directory.

I did a quick static deploy of React using Surge first off. This worked pretty well for sharing my app quickly, but didn't allow me to have a URL that made sense (without configuring a custom URL).

I wanted to get everything running on Itch.io because its a popular platform for Ludum Dare and my other projects are hosted there. Create React App expects to be deployed as a static server in the root directory. However, Itch doesn't host your projects in this manner. I had to modify my package.json file with the value "homepage": "." to make the build process be dynamic.

The final piece of deploying my project to Itch.io was to make use of their command line interface deployment tool Butler. It's a really cool tool that makes it much more painless to deploy a patch to your project. Instead of zipping a bunch of files, navigating to the site, and uploading your files you just type butler push build law5guy/infinite-desktop-icons:html.

Play the finished product in your browser here.