Levis Umbra
Levis Umbra is an Action Stealth Game. Infiltrate an underground facility and avoid the machines at all costs. Find a way to go unnoticed as you penetrate deep into the core, armed only with a hacking device and your wit.
Main Role: Level Design, Design Lead
Secondary Roles: Environment / Lighting Artist, Game Design
Timeframe: 16 Weeks
Platform: PC
Team size: 14
Software used:
3DSMax
Jira
Unreal Engine 4
Perforce
G-Suite
Main Responsibilities
--------------------------------------------------------------------------------​
-
Full ownership of the last third of the game, from conception to release
-
Level Design Documentation giving an overview of the entire player experience throughout the level.
-
Visual benchmark created to guide the art team towards a coherent direction.
-
All the dressing is done for the last level including light, props, and VFX.
-
Creating supportive assets to help keep the main building blocks visually interesting.
-
Creating the trailer for the game that was submitted and featured in the unreal student showcase.
-
Level Design related tools to help us speed up and ease the level creation process (and are widely spread amongst my peers)
-
Taking care of the design team and updating the vision holders as the design lead.
-
Creating the backlog for the design team and keeping up Scrum as the Scrum master.
Pre-Production Phase​
Research and concepting
After having done the concepting, the first thing I did was competitor analysis on Styx and splinter cell. After exploring our reference games I started sketching out moments to get an idea of the type of gameplay that we wanted to create. I also wanted to sketch gameplay scenarios, closely related to our user stories, to get visual reference for what we wanted our players to feel like.
Reference and build language
Since we had no artists during Pre-production (other then for the character), I combined the work I was going to do for metrics and build language into a small example scene to concept the look and feel of the game as well. I experimented a lot with lighting and volumetric fog to get a strong and convincing mood going, and prepare the setting for artists that would join later on in the process.
Planning and documentation
After concepting dozens of different gameplay setups I picked a few that looked the best on paper and fit well together, and I combined them into one full level concept. For the planning I made a Molecular design and put it into our level design document together with a tension graph, level design pillars and user stories.
Blockout
Blockout
I started the blockout by recreating my sketches and stringing them together in one level. Going from there it was very important to our level design to work with verticality since our limited character movement had only one additional mechanic, which is climbing. During the blockout we didn't have a chance to properly playtests just yet due to the lack of mechanics, but it helped us greatly get an overview of what would work and what wouldn't.
Whitebox
Since the end result of the project was supposed to be a vertical slice, we mostly iterated based on internal playtesting. Every setup has it's own quantity, but the overall minimum is at four iterations per setup. During the white boxing process I've done a lot of cuts, replacements and additions to the content all based on feedback from the team, peers and teachers.
Blockout progression
After having done the concepting, the first thing I did was competitor analysis on Styx and splinter cell. After exploring our reference games I started sketching out moments to get an idea of the type of gameplay that we wanted to create. I also wanted to sketch gameplay scenarios, closely related to our user stories, to get visual reference for what we wanted our players to feel like
Gameplay sections
Walkway
The walkway acts as a connector and resting moment after a difficulty spike in the level before. It being a fairly long hallway without any enemy interaction, I had to find some ways to make the walkthrough a little more exiting. I made two cheap custom events, interconnected it with a different gameplay moments (which you can only see) and a reward for going slightly off the path.
Drilling site
Drilling site has two sections; a small traversal section to make time for narrative voice lines, and the first stealth setup of the level. Once the player enters the room he gets a vista of the main subject of the level which is the drilling tower. They make their way to the bottom of the tower where they get challenged by a more open section, in which they need to make their way past two robots.
Debris
Debris is the railway that supports the drill tower by taking care of all the rubble and side-produce it makes. The setup is within a fairly tight mining tunnel that features a small flank route that can help the player sneak past the two patrolling Robots. To get through the level, the player has to open a door by hacking into a control panel that get's guarded by the patrol cycle of the robots, creating an interesting timing challenge.
Storage
The storage room is where all the equipment used to keep the drill tower active is stored, and my favorite setup in the level. The entire setup is enclosed in one small room filled with crates and one single enemy. The player has to sneak over, under and through the stored goods all whilst going around a guarding robot.
Roadblock
Roadblock is the part that used to be connected to Debris as part of the railways system. in this setup players get to play around with verticality a lot, since they have to find their way on top of the platform. Like Drilling site, this is a more open area in which the player has more room to be creative and come up with their own solution.
Option 1:
After hacking the panel and opening the door to the next part of the level, you need to make your way there. This option is to sneak past the pitfall and the drilling site back to the door.
Option 2:
After hacking the panel and opening the door to the next part of the level, you need to make your way there. This option is make your way up to a scaffolding above, sneak past the enemy patrolling there and drop down onto the door.
Intersection
Intersection is one of the few tunnels that's still operational after the robots took over. In contrary to most of my setups, this one has the additional routes and is very open, but the player is also very vulnerable everywhere they go. The level has only few save spots where they can recollect their thoughts and plan their next move, making it much more challenging.
Walkway
Walkway is the last setup of the level and is a bit of a cooling down setup. The scaffolding this level consists of leads directly to the top of the drill where the player ends the game, as patrolled by only one enemy. In this setup the player has to tail the enemy and is often very close him behind boxes or closed of rails.
Environment art
Going into the project we realized that we had a severe lack of environment artists. Even though I wasn't versed in anything related to environment art, I had some base knowledge in 3DSMax and the will to learn more, so I offered up a chunk of my time to help out the art team. Below is the work I did to help them out:
​
Visual Benchmark
Through a combination of assets by our artists, an asset pack and my selfmade assets I made a visual benchmark to visualize what the final dressing of a level should look like.
Creating this scene has helped the team a lot in getting a visual representation of what our game was going to look like and gave us a ton of motivation. for my Personally it helped a lot with getting confidence in my artistic skills.
Prop art
Since I had basic knowledge of 3DSMax I naturally flowed over into doing a bunch of supportive props for the game. I made a total of 10 Assets for the game that we're used in my level, and the first level which is the only level that received an official art pass.
Due to my lack of knowledge regarding textures I mostly used textures from Quixel, unreal store asset packs and modified them to fit the setting and style.
Level dressing
After making the visual benchmark I got fairly confident that I could dress the levels myself without needing much interference from the art team. I've spend time setting up lighting presets, outsourcing crate assets, and actually dressing my level. I managed to dress up to 80% of my final level and the other work I did was used throughout all three levels.
Automated screenshot tool
"During this project I scripted an automated screenshot tool in the EU4 Blueprints. The tool automates the process of consistently making progress updates, automatically renames the files to the cameras name and has the option to take screenshots of all cameras in scene or a custom select few. The tool is very well received and wide spread amongst the level designers in my school."
​
Full playthrough of the level
Personal Post-Mortem
My stance in the project:
As a level designer I didn't have an incredibly strong presence in the final product, but as an overall developer I was still very much present. My flexibility and will to learn has helped out the team in many other regions and by being design lead I've helped the team greatly by preparing their sprints so they could keep working.
​
My biggest mistakes:
During the entire process, from sketching until whitebox, I somehow convinced myself that because my level is the last of three, I had to create these big, complicated setups to have a hard enough to count as the last level. What I realize now is that a gameplay moment doesn't have to be set up in a very complicated way to be effective and difficult.
​
My best learnings:
Complex for me is NOT parallel to complex for the player. Difficulty comes from knowing your metrics and cleverly manipulating the map(and in this case enemy patrolling cycles) to fit the difficulty you're going for.
