Level design takes time. What if you don’t have it?

Prize:

5th place in the official NEW GROUNDS mapping contest.

Steam Workshop :: Training Grounds

Responsibilities

• Level Design
• Set Dressing

Development Time: 9 weeks

Team Size: 3

Tools

Killing Floor 2 SDK

Project specifics:

Time was against us. We entered the competition late (it was from September till April) in December. Since I played Killing Floor 2 for only 6 hours I had to get familiar with the game and all its mechanics. I played until the 20-hour mark and then I allocated my time to learning the SDK tool which is based on UE3. My short playtime, intermediate skills at the game and lack of time for playtesting meant that hardcore players were an audience out of my reach. Two “advanced tactics” YouTube tutorials were not enough to fully explain how advanced players approach the highest difficulty and how maps support their playstyle (Psst… They do it by giving them areas suitable for a holdout).

Killing Floor 2's levels are created using a specific system of “building blocks/pieces” which meant more time to get comfortable with the level design process. Since the “pieces” had to be aligned in specific ways, I initially spent a lot of time on blocking out a space. The macro blockout stage was taking too long and I was not willing to make huge changes since that would mean re-aligning everything. I decided to not follow the system until I figure out the general scale and shape of the map. As I was gathering reference pictures, brainstorming concepts and figuring out focal points I had to constantly explore what assets were in the game and how we can re-use them in our map without the set dressing being repetitive or too familiar. This made the process even slower.

The process

Set dressing wasn’t a separate stage. I was doing some of it as part of the micro blockout. We didn’t have time or people for many custom assets so the layout had to work with what was already in the game. Next stage was the micro blockout where I had to pick the specific assets, create the layout of each playable space, create cover, balance the line of sight and adjust the connectors (corridors and stairways between all rooms). A lot of connectors had to be changed to create better flow.

A spiral stairway didn’t work well for combat so I added a shortcut for players to escape (orange line). This wasn’t enough so I had to re-design the whole area. I kept the shortcut idea to give players more fun options.

Initially the exit to the right would put players in a corridor that only lead to the right. It lead to areas that 2 other doors were already leading to. I changed it so players could go left so they would have more options to kite the zeds.

A fellow level designer from my university gave me the feedback that there isn’t enough elevation variety. So I used a bunch of assets to create some:

The bad movement flow of the 1st iteration of the city area is why I decided to drastically change that segment of the map. I didn’t like this area when I finished with the macro blockout and after spending some time on trying to figure out how to fix it I decided to look for other assets to use for the city area. Once I found smaller buildings I decided to use them and figure out a fun layout.

Yellow lines highlight the paths players can use.

Polish

When all the design was done and the map was ready for playtesting we still had to place the path nodes for the AI path finding and I needed to create custom collision volumes around the meshes so players and Zeds won’t get stuck.

We planned time for playtesting but the unexpected technical issues the solutions of which couldn’t be found in the documentation took that time away. Luckily the modding community is very welcoming and when I turned to them for help, they always responded.

While waiting for help I didn’t waste my time. I started to polish the visuals. I played around with the lighting in a few areas.

Zeds can spawn from this hole in the panel. I added smoke particles and red light to make it look more dangerous. The red light is placed high and it’s angled towards the snow in the playable area. When enemies come out their shadows are exaggerated and are the first thing the player sees creating a more mysterious and scarier experience.

The main problems of the map were that it’s too small and some areas were a bit claustrophobic. I should have made more spaces with plenty of room for players to maneuver in but the lack of time meant we couldn’t set dress and polish something that was bigger. The last week of the competition looked like this: 26–30 hours of work followed up by 6–8 hours of sleep and then… work again. We slept at different times so there’s always at least 1 person awake who can look at Discord and see if someone helped us with an issue. Even with this crunch we were able to submit only an hour before the deadline which got us 5th place in the competition.

To fix the gameplay issues we will need more… You guessed it — TIME. Currently we have gathered a lot of player feedback so we know what needs fixing and we have a plan on how to approach the problems.