In order to keep track of progress and tasks, our team decided to use Trello. This helped with organization, tracking, and reporting progress and work.

After FireLight was pitched, we had only one week to organize our initial ideas and designs. Because our team was small and I was serving as both the level designer and environment artist, I needed to quickly establish the foundation for our level layout. Our original plan was for FireLight to function as a semi–open world game built around a single, expansive level. This level would be divided into distinct areas, each with its own purpose and atmosphere. For the first quarter, we focused primarily on the central section of the map while intentionally designing entrances to future areas that we could develop over the summer and into the fall quarter. 

After getting the base ideas down on paper, we decided to gather some base references to define the style of our game so we could begin work. Each team member had their own mood boards that we then after combined and refined. The image on the left shows the environment mood boards myself and the other environment artist collaborated on.

Our professor suggested shifting the project into a level-based game with semi-platformer elements. We agreed this direction would strengthen the experience, and also decided to make the game isometric. With the isometric camera in mind, I had to strategically place each platform so the character would always remain visible. I began with a quick sketch and then translated it into a blockout in Unreal Engine 5. 

After receiving additional feedback from our professor and classmates, we realized the environment felt too flat and needed more elements to make it more engaging. In response, I added more bridges, platforms, and mountainous areas to encourage freer exploration.

At this point, I was rushing—we had only one week left before Beta Day, an event at SCAD where senior projects are showcased for playtesting, debugging, and feedback. Our team, however, had been assembled due to an awkward sequencing change in our major’s requirements, so our seven-week game was being presented alongside projects that had already been in development for over a quarter and a half.

I quickly set dressed the environment, and we packaged the game just in time for Beta Day.

After Beta Day, I decided it would be more efficient to completely redesign our environment rather than try to fix the existing one, especially with only two weeks left in the quarter. With the team’s approval, I created a new level with a fresh environment. This redesign introduced a clear path up the mountains and a single route back down, establishing stronger directionality. I also created new tree assets to better match the intended style. Given the limited time to both model new trees and build an entirely new environment, I used Blender’s tree plugin to generate the base tree models. Around this time, our team also agreed it would be best to remove the isometric camera angle. 

After the summer break, we were placed in a class with a new professor. His guidance was to refocus our core gameplay loop and make significant changes to the environment. This essentially led our group to rebuild the game. Reflecting on my earlier work, I realized I had been prioritizing learning stylized asset creation rather than designing the environment around our game’s mechanics and applying solid game and level design principles. Now that I was the sole environment artist, I wanted to begin by creating a new, more detailed mood board—one that genuinely centered on our gameplay and mechanics. 

Over the summer, I decided it would be a good idea to learn proper techniques for creating stylized work. I realized that a major part of stylized asset creation involves ZBrush and using generators in Adobe Substance 3D Painter. With that in mind, I set out to study both programs. When the quarter began, I moved in with a fellow group member whose main program is ZBrush, and she helped me deepen my understanding even further. 

A fellow group member took on the initial level design for the quarter. Once that was complete, it was my role to bring the environment to life by adding all the foliage and materials. Another teammate created some props, which I also integrated into the overall design. 

I began by creating several stylized foliage which would be incorporated onto ground foliage assets, wall foliage assets, and the trees in ZBrush and Maya. I then also began learning how to create certain materials. One material I learned to create is a fake lighting shader that detects the direction of Unreal Engine’s directional light and generates simulated light and shadow directly on the model, without relying on a physical light source. This approach helped with optimization, and it also gives full control over the effect. Some features include the ability to adjust the strength, color, and the distance the light and shadow travel across the surface. 

Creating a dynamic water material was a challenging task. The original water texture was too pixelated and lacked the movement and flow I wanted. It also didn’t use depth fade, and each water layer had its own set of controls, which made the material instance unnecessarily confusing.

For the new version, I studied Fabian López Arosa’s free water material and broke it down to understand how it worked. I didn’t want to copy it directly, so I modified and rearranged several nodes, experimenting until I achieved a result I was satisfied with. I then rebuilt my version in our project.

After completing the first level in just two weeks, we decided to create a second level—this one serving as the endgame area where the player finds Grandma. We chose a swamp theme for this level. To reinforce the narrative, I designed the gameplay so the main character chops down thorny vines instead of trees, since the swamp trees would be too wet to use as firewood.

I envisioned the swamp as a series of small islands connected by bridges, reflecting how real swamps often consist of patches of land separated by water. This layout also naturally increases difficulty as players leave the safety of their fire. The lighting for this level needed to contrast with the forest to create visual variety while still feeling consistent with the style and world of our game.