Skip to main content

Level Design Blog 1

Blog 1: 8/20 - 9/3


In class, I worked on completing the randomly-selected blockouts. I worked mostly with the included modeling mode tools, although I also spent some time trying out the BSP brushes on the cruise ship.



Outside of class, I worked more with modeling mode and landscape mode and practiced lighting my scenes. One of the first blockouts I tried was of a playground. When I was a kid, my neighbors and sister and I would visit the playground near our house all the time, so after my mom told us it had been completely bulldozed a few days after the semester started, I decided to immortalize it in Unreal. 

I used photos I had taken, along with Google Earth satellite images, to try to reconstruct the playground as a basic whitebox level:





I found that I struggled with more complicated shapes like the slides, so I used placeholders like spiral staircases for them, and I skipped the railings around the playground entirely. Once I was done, I used Merge Actors to turn the playground into a single mesh. In the future, I might bring the mesh into  Blender or Maya to practice modeling and some simple textures.

I also brought the playground into a second level and used it to practice with more complicated landscapes and levels. I wanted to work on setting up levels with intentional paths created for the player, so I created a landscape with a lot of hills and valleys and set the playground on a hill in the middle, along with some tall, simple trees, to act as a goal point for a potential player. From there, I decided on a starting point for the level, and I tried to whitebox the level in such a way that it would motivate the player to get from the start to the playground. 


I set up the first building so that the player would be led to a small doorway, which would act as a pinch point, framing the playground and the trees through the additional cylinder "trees" I put around the doorway:


From there, the player would have to navigate down through the "trees" to the next building, briefly losing sight of the goal point along the way:


The playground comes back into view at the end of this bridge:


From there, the player has to pass through a cluster of buildings, losing sight of the end point again, and then across a ravine:



I didn't finish the level, so in the future, I would like to go back and add more to the end of the level to make it more satisfying. I enjoyed working in modeling mode, though, and I felt like I learned a lot, especially about landscape creation and about making more complicated shapes in Unreal. 

In between my whiteboxing practice, I tried using some of the megascans from Quixel Bridge to make a couple of simple scenes. I used most of the same assets for these two scenes. 



In both scenes, I found that the lighting played almost as big of a part in making the scenes look good as the actual assets and framing did. I spent a lot of time trying to perfect the feel of the blue glow in the first scene and its contrast against the sunlight.

In the second scene, I wanted the sunlight to help tell the story, so I set it up so that the sunlight would fall only over the chair that is still standing. I ended up adding a lot more lights, including some very dim blue-toned lights around the fallen chair and the more shadowed parts of the scene. Getting the lights and shadows to fall exactly how I wanted them to took more work than I expected.

I had a lot of fun with these two images, and in the future, I would like to work more with set dressing like this. If I were to spend more time on these scenes, I would try to work more with the actual assets; I think changing some of the textures, especially on the foliage, would give it a more realistic feel, and although I did tweak the exposure and contrast slightly with a Post Process Volume, I could have spent more time on that as well.

The last thing I practiced was more lighting on a basic whitebox. After playing some Baldur's Gate 3, I decided to try making a simple level based on the spider cave near the start of the game: 


The cave is essentially a group of small platforms with spiderweb bridges running between them. In my version, I started the player in a little room on a large platform, from which they can see a set of additional green-lit platforms:


I built all of the platforms in this level using Unreal's cubegrid tool and connected the platforms with longer rectangular meshes. Since I was originally planning to focus entirely on the lighting, I wanted to make the structure of the level fast, so that I could get to the lighting more quickly.

It isn't until the player reaches the other platforms that they can see the goal point of the level, which is actually set into the pillar just below the starting platform:


I wanted to try to use lighting to convey the relative safety of each of the areas: the yellow-lit areas are the start and the end of the level, and I wanted to try to use a warmer light to show that they were safer. The darker, green-lit platforms are very exposed, and might be more dangerous. I used a combination of point lights and spotlights to highlight the platforms and add more ambient light to the area.

I wasn't incredibly happy with how this level turned out--although I liked how the lighting turned out, I didn't think the cumulative feel of the whiteboxed level completely fit what I was going for. In the future, I might try to redo the same concept, but with the addition of more structures and a stronger hierarchy of platforms.

Looking more closely at the differences in setup between my level and the spider cave, I think the cave works much better partially because of its structure. The spider cave is set up with multiple levels of platforms and bridges, all of which form a spiral shape leading into the deeper parts of the cave, which aren't accessible to a player. The platforms are all out in the open, but there's a finite feel to them, unlike mine, which is just floating in the void. 

The multiple levels of platforms also give both the player and the spider enemies more places to fight from. There's some strategy around where you decide to place your characters, and although I was picturing this level as some kind of arena when I made it, I ended up turning it more into a circuit that the player would be just trying to get around to get back to safety. 

In the future, I'd like to remake this whitebox with a stronger idea of how I want the final level to function. I started out wanting to focus completely on lighting, and it was definitely good practice in that sense, but design-wise the level felt weak. 

All in all, I was able to get much more comfortable with a lot of different tools in Unreal over the past few weeks, and I already feel like I am able to work faster and more confidently than I was when I began my playground level. I look forward to spending more time on more complex levels and on polishing the levels I started over the past few weeks. 

Comments

Popular posts from this blog

Level Design Blog 20

Blog 20: 4/1-4/8 This week, I finished the mock design test. As I was working through it, I found that I had a lot of uncertainties about the process; although it made working on the assignment more difficult, it was probably very good experience for future job applications. The biggest hurdle for me during this project was using Autodesk Maya, which may be my new least favorite program. The program kept crashing for seemingly no reason as I was modeling the scene, which became increasingly more common as the level became more complex. I tried to mitigate some of the issues by staying on the FIEA VPN, saving frequently, and merging meshes and deleting history as I worked, but even with that there were probably dozens of Maya crashes, and once where my entire laptop bluescreened. Most likely, it's an issue with my specific machine, and not with Maya as a whole, but it made the whole process much more tedious either way. The other challenge I faced with Maya was that I had only ever ...

7/10/25: Fortnite Manor Progress Update

7/10/25 This week, I began work on my custom level and game mode in UEFN. Although I had developed several of the main mechanics in my metrics gym, I ended up overhauling them almost entirely, both for aesthetic and practical reasons. Although I had created a custom AI to go with the beacon guards, I found that using the built-in guard AI would be more fleshed out and would integrate more easily with other devices. I also used the built-in set of devices for building and assigning teams, which I was able to complete fairly easily.  Waiting room for players where team selection takes place.     Visually, I stuck with a palette of pieces from several different Fortnite packs: the majority of the building elements, like walls and floors, came from the Grand Glacier Hotel, while the plants and other decorative elements came from several garden sets and the Hexylvania pack.  I started with a rough blockout of the rooms using generic wall pieces, which I then assigned them...

7/15/25: Fortnite Manor Final Product

7/15/25 This week, I worked on wrapping up my Dead By Daylight-inspired Fortnite level and game mode. Much of the level progress this week was on the flow of the level and the mechanics involved. One of the most significant changes I made was adding functional gates to the level through Barrier devices and buttons to ready teams. I initially struggled a bit with understanding devices like the Stat Counter, but once I figured them out, getting the player gates set up was a breeze.  A mid-week update to the front gates. Like last week, each team initially spawns in a different location: Hunters spawn together at the front gate, and Fugitives spawn separately in either side of the attic. Each team spawn location has a button to "ready" their team; when both teams are ready, the barriers around each spawn area drop, the AI guards spawn in, and teams are able to begin moving through the manor.  Other than adding additional props for scenery, most of the remaining changes were to l...