Student Council Showdown Devlog & Postmortem

WEEK 1 (10/8 - 10/14): PLANNING

I've committed to making a choice game, something I've never tried. However, due to the scope being only 3 weeks long, I want to create a system where the choices aren't complete preordained. This will hopefully lighten the workload and lend itself to a shorter experience. An initial idea was to have a card game where the player can collect cards to be used as words in a conversation, but that ended up feeling too limiting, since either the cards would have to be super vague so as to fit in any conversation, or extremely contextual, only being able to be used a little of the time. Instead, I have decided on a school debate, since the question and answer format lends itself much better to these choices. I've outlined the debate systems in a design doc, but it basically boils down to the player being asked questions that are related to certain topics. I've also written out 16 questions and 16 preferences that correspond to various possible answers to those questions. The 8 cliques of the school will each have randomized preferences, each tied to possible answers to the questions. The player must appease all 8 cliques in order to win the election.

WEEK 2 (10/15 - 10/21): ASSET CREATION

This week I worked on the visuals of the project. The aesthetic I'm going for is that of a fighting game to relate the debate to a brawl. As such, I drew inspiration for Street Fighter for the mise en scene and River City Girls for the sprite art. I started with the two debaters, which actually turned out to be the easy part. I went on to do the 8 cliques, which needed both idle and cheering animations. I want to rely on high school stereotypes while also not perpetuating the negative aspects of them. To differentiate them, I've given them each unique color schemes that should help them stand out. The cheering animations took me longer than I'd like to admit, but that's mainly just because I'm a perfectionist, and I'm happy with how they turned out. I finished with the environmental art as well as UI graphics. After importing all the visuals into Unity and arranging the scene, it inspired me to write a piece of music for background. Since the art of the game is reminiscent of 8-bit games, I used a NES soundfont for the melody, but to make more of a modern revisionist chiptune soundtrack, combined it with trap beats from a drum machine.

WEEK 3 (10/22 - 10/28): PROGRAMMING

Programming this project, as with all, proved a lot more difficult than I expected it to be. The biggest hurdle came when I realized the best thing for me to use for keeping track of so many variables (such as preferences, topics, and scores) was arrays, which has generally not been my strong suit in coding. However, when I started to get to understand them better, it let me drastically lessen the number of variables in my project. That, paired with my decision to abstract each question function rather than write out an entire function for each of the 16 possible questions (plus rebuttals), allowed me to cut over 1000 a 3000+ line debate manager script. However, since so much of this project is dialogue-based, there is still a lot of code that is taken up by strings. Dialogue, as I've learned, can quickly get out of hand. Luckily there wasn't too much combinatorial explosion since I kept the questions and answers fairly abstract, but it still was a lot to deal with. I also split my game manager into 3 different scripts because my main script was become unwieldy. Luckily, by using abstracted functions and arrays, once it fit, it all fell into place.

PLAYTEST FEEDBACK

The visuals are a good indication of how the crowd responds to the player's answers, but with so much of the information being withheld from the player, they find themselves making the decisions based off what they think sounds cool/funny, not to appease the crowd. One major component of this is the polling feature, which the playtester thought was an interesting concept but because it was so limited and didn't allow for tracking changes in approval across multiple rounds, ended up feeling extraneous. Thus, while the playtester was happy with their outcome, they said "I don't know how I won that." I wanted to keep some amount of information to the player to make them think more, but it seems I kept so much that they thought less.

POSTMORTEM

There are numerous features I would want to add to this project if I continue to work on it. The good news is, because of the way I coded the questions, it's fairly easy to upscale and add more possible questions. The calculation of the player's final score is  very rudimentary in its current state (simply adding together each cliques approval); I have a ton of variables that I never ended up using that would measure the consistency of the player's message and the adeptness of they're debating. This would require a lot more math and playtesting to get the numbers just right. Polling would be an easy fix; it just needs to give more information, such as the preferences of the cliques polled and the ability to keep track of cliques polled across multiple rounds. Overall, though, I'm extremely impressed with what I was able to achieve and learned a lot, especially with arrays.