Trapped
This project was created in a group of 15 developers consisting of programmers, 3D modelers, level designers, concept artists, a game designer, and a producer. The game was created in Unity and is a 3D maze rouge-like game, where you navigate through a maze on a cube fighting guards and increasing your stats to get even stronger with multiple levels of varying difficulty.
During this project, I was one out of the three Level Designers and was in charge of designing three levels with varying sizes. These levels were divided into three sections: small, medium and large. Our game designer emphasized on the idea that each cube will have the same maze pattern on each side made by each of the programmers, meaning that we were tasked to make a total of three maze patterns and repeat them six times. The catch was, different enemies, power-ups, pick-ups and switch placements would be different on each side giving the player variety on each side, while keeping the patterns the same. The player would advance to the next level if they push every switch on each side, besides the top side, and must navigate each side. With our tasks laid ahead of us, each level designer was given one level from each size, and tasked to design, annotate, block out, and implement them into the game. The player would go through the three sizes and the game will randomly select one of the three iterations of the levels we made, meaning the player can either get all three levels of one level designer, get one level from each of the designers, or any other mixture of levels.
My plan was to start from the smallest and finish with the largest, as I thought it would be a good way to get an idea of what would and wouldn't work without having to make a lot of changes if I were to start on a larger level. Knowing that every side would be the same on the cube, I knew that symmetrical design would be the most effective way for implementing the levels later on and would look the most maze like. With the symmetrical design in mind, I designed only 1/4th of a side and then duplicated three times to fill in the rest of the side to get a perfectly symmetrical pattern. I also made sure that the player had some room for movement, as the small levels would be the first levels and I wanted them to provide challenge but be lenient with the amount of room the player could maneuver around. This design would change for the medium and large levels, as they are suppose to be the harder levels. After approval from the game designer and the producer, I moved on to add in the varying enemies and power-ups on each side. For the first level, I focused on enemies approaching the player once they enter the room, to give the player time to react and plan accordingly. However, this became problematic in practice when I realized that the players, hypothetically, could enter from behind, giving them no time to react or plan if they entered on the "wrong" side. This idea was scratched and instead, I put the enemies closer towards the middle or closer to the edges of the map, to evenly place the enemies if they were to enter from any side. After approval of the finished annotated map, the next step was blocking out the level in Maya.
For blocking out in Maya, we used modular wall pieces created by the 3D art team members and would place these pieces on an already sized cube. This way, each level was exactly the same size and us not having to worry about other level designers making slight differences in terms of size. Similar to how I annotated the level, I blocked out only 1/4th of a side and would then duplicate and fill the rest in, then duplicating the sides and adjusting them accordingly to match on each side. The process went smoothly by having the pieces fit perfectly in almost any variation I put them in. Once complete, I exported the object out and was ready to implement them into Unity.
For implementation, the only task given to us level designers was to to add in some assets also created by the 3D art team, and let the programmers implement any scripts and enemies that would be needed to make the level function. They were told to look at our annotated maps as reference. Implementing the assets took longer than originally anticipated due to multiple changes for the character during this time. Our game designer wanted the assets to be different on each side, similar to the varying enemy and power-up placements. When the character changed size to be much bigger, they would clip and interact with some of the assets that were not supposed to be touched or interacted with, thus revisions of the assets were to be made. When adding the assets, I tried to differentiate each side by making the tubes a unique pattern on each side, this making it clearer to the player which side they were on to add with the different enemy placements.
While waiting for the programmers to make the level functional, as they were tasked with, I moved on to the next sized level, medium sized. For this level, I wanted to design not as free-flowing as the previous level to make it more difficult. I made the flow of the level to be like a figure-8, making the player having to make turns left to right and vice versa, and the pattern turned out almost like an hourglass. This level is not as symmetrical as the previous one, but it still has the symmetrical exits being in the center on on each edge of the pattern. Enemy placement was more lenient than the previous level, as the level was not perfectly symmetrical, it allowed enemies to be placed in many more different spots that would not cause frustration or a feeling of unfairness. The medium levels also receive another enemy, a guard who follows and attacks the player if spotted, which allowed a great amount of through into the enemy placement with the bigger space to work with in the medium level. With these extra obstacles and unique enemy and power-up placements, the game designer and producer were pleased and approved of blocking out and implementation. These next steps were very similar to the previous level of duplicating the sides to be as effective as possible and then adding assets with unique patterns to help differentiate each side. Once finished, it would be sent to the programmers for functionality.
At this point, we were getting feedback on our most recent build and there were issues with some levels being far to easy and player clipping out of the level. The level designers were tasked to play test our first levels, as they were the only ones functioning at the time, and discover the reason for the clipping issue and to fix it. This took a good chunk of the sprint, as we also were consulting the programmers who also worked on the levels to try and see if the issue had something to do with the scripts. After multiple tries of testing, I discovered that the player was clipping through due to how the colliders on the walls were being implemented. The programmers used a mesh collider, which is not as reliable as a box collider. Another reason I found was that the "speed" upgrade made the player so fast that they would easily clip through the mesh collider for the same reason stated before. Knowing what the issue was, I went back into Maya and reorganized how the walls were grouped together and combined. To make implementing the box colliders as simple as possible, I combined and separated the wall pieces to be only squares and rectangles instead of just combining every wall together as one object. This reorganization was able to add the box colliders in Unity easily and would also automatically overlap each other, preventing wall clipping, due to how they were combined, which worked out great.
After fixing the colliders, I was tasked with designing my last level, the large level. After seeing the feedback about the levels being too easy, I decided to make the large level more difficult than I was originally planning. Once I observed how the enemy AI acts, I decided to make the pattern have less free room and make them single lines with multiple paths to allow players to out maunever the enemies, but would still have the fear of being sandwiched. I also wanted the player to defeat enemies more rather than just avoid them, which was one of the reason the levels were easy. In response, I placed far more turrets that would be place in areas that would cause the player to either find a different path to the button or something to be cautious when slipping through the bullets. With the large amount of paths, the difficulty of the enemy and turret enemies, and the large size of the level, the game designer and producer were extremely pleased and really liked how it looked. Once approved, I moved on to blocking out the way I did to fix the colliders and then added my assets in the same way I did for the previous levels.
Once I finished the assets in the last level, I was making plans on adjusting my small level and medium level to be similar to my large level and make the pattern harder by making the paths thinner and enemy placement more prevalent. However, I was tasked to assist the other level designers due to how fast and well polished my levels were and helped the other level designers with their levels. While still assisting the level designers, our producer and game designer made a major change to the assets of our levels and told us to remove our previous assets in favor for newer one. These assets matched the themes and original vision of the game designer, but were made by the 3D art team and the level designers were to simply implement them into Unity and duplicate them on each side. While I believe that the new assets were better than the old one, I believed that the new assets with the old ones would look as good if not better. After discussing with the two, it was decided that it would be simpler if we just used the new assets to avoid clashing aesthetics. After putting in the new assets and changing their colors to make it even easier to distinguish each side, I continued assisting the other level designers until the game was published to itch.io.
Working on this project was enjoyable and was great experience for designing levels in a collaborative way with other developers. I have learned a lot, especially with designing when working in a group and development on certain aspects of the game take longer than expected or change at what seems random points. I am extremely grateful for the opportunity to work on this unique project and thoroughly enjoyed working with the team.
