Banding Together
Made for Game Off 2022
Postmortem
For Game Off 2022, I chose to have my theme, a cliche phrase, randomly selected by a friend of mine. The phrase I was given was "Banding Together to Beat the Odds". So, I decided to go about making a game with an AI partner to "band together" with, and for "beat the odds" I decided to take it literally and make enemies out of odd numbers. I wasn't able to find a team in time, so this was a solo project.
Restrictions
Game Off 2022 Theme:​
-
All projects had to adhere to the game jam's theme, which this year was "cliche phrases".
​
What Went Right

Spline-Based AI
This took up the bulk of my time while making the game. I wanted to base my AI partner off of the companion Zelda from The Legend of Zelda: Spirit Tracks. I tried looking for any documentation or anything I could use as a reference point but was unable to find anything to point me in the right direction. So, I spent about a week of just trial and error trying to get the AI to follow the spline. And then it still took more time to ensure the spline mesh worked properly and that the AI didn't get stuck or try to move to unintended locations. I'm glad I stuck with it though, because I think I was able to replicate that draw to move feel from Zelda, which was my main goal.
Odd Enemies
As part of my theme, I wanted to make enemies based off of odd numbers. I tried to have each enemy interact with both the player and their AI companion in different ways. I also tried to incorporate their general shape into their behavior. The only number who's shape I had trouble with incorporating was 5, which is why I think its area is probably the weakest. I think all of them achieved their goal of having the player and companion work together to beat them.

Moving Camera with Mouse
Before this functionality was added, the player couldn't move the camera at all, and it just felt bad. I thought about some different kinds of camera systems for top-down games but decided to go with a system normally found in RTS games, with the camera moving when the cursor is at the edge. It felt a lot better to play after it was added.
What Went Wrong

Color Path Switching
The green switches in enemy 9's area were originally going to have multiple purposes beyond opening and closing yellow doors like they do now, with one of them being to switch the colors of the color paths. This would have allowed certain areas to be locked and unlocked to either the player or the partner. Unfortunately, I couldn't get it to work quite right in time for submission, with it allowing the player and partner out of bounds or softlocking them, so I had to scrap the idea altogether.
Grab
Originally, the player and their AI partner were going to be able to grab objects to move them around, activate buttons, and distract enemies. Unfortunately, due to time, I had to scrap the idea completely for the partner, and was unable to complete it in time for the player.


Cursor
This was probably my biggest mistake of the entire project. I didn't notice it at the time I made it, and since I spent so much time on the partner's AI, the consequences of it weren't discovered until other parts of the game were being made. The cursor is part of the player. Because of that, any object that needs to detect if the player is overlapping or if the player is near or anything like that, has to cast to the player and check if the component is the cursor. And because I had so little time left, and I didn't want to spend that time fixing the partner's AI, I had to leave it as it was. It's a pretty big mistake and one I wish I had realized a lot sooner.
Conclusion
Although l was short on time for basically the entire project (I actually joined a few days late and of course there was Thanksgiving). I really enjoyed working on it. I delved deeper into Unreal's AI behavior tree system, learning more about what and what not to do. I completed my first project in Unreal 5, which wasn't too different from 4. And I got to stretch my Designer skills while still having a roadmap to follow with the 30-day time limit and my cliche phrase. I especially enjoyed working on the AI partner's spline-based movement. Learning to do something like that without any help was, mind-numbingly frustrating at times yes, but also lot of fun when I started figuring things out, and it let me replicate one of my favorite mechanics from my personal favorite Zelda game.