Marbleground Construct
Marbleground construct is a 3D puzzle platformer. Set in a children's classroom you complete these pre-build puzzles by placing in different objects that help you control the marble.
Main Role: Level Design
Secondary Roles: Game & UI Design
Timeframe: 8 Weeks
Platform: PC
Team size: 8
Software used:
SketchUp
Trello
Unreal Engine 4
Perforce
G-Suite
Main Responsibilities
--------------------------------------------------------------------------------​
-
Initiating the level design process and initial visualization of the concept,
-
Creating modular sets to block out level concepts with the idea's generated by team-wide brainstorming,
-
Implementing level geometry, missions system, and Interactable level elements.
-
Game & level design documentation and upkeep of professional communicative documents.
-
Researching, concepting, prototyping, and implementation of UI elements.
-
Making UX-focused prototypes for concept features that helped us cater to our target audience.
Pre-Production Phase​
As a game designer
Being partially responsible for the game design as well, I assigned myself to most UI/UX related topics. I worked on concepting the in-game UI, I worked on our game design document and made the UI flowchart.
As a level designer
Initially I was put together with this team by my teachers since they were in need of a level designer, so I wanted to show the team early on what I could mean for them. I did this by making a modular set and level sketch in SketchUp that fit what they showed with the initial concept art, which gave us a great view of what the game would actually look like in 3D. Besides that I also set up the level design documentation and made an asset list with what models I need plus their measurements.
.gif)
After receiving the modular set from the artists, and the programming team being done with the physics system, I wanted to get a first version of the level into engine. This helped the team get a grip on what direction the game was heading and steering it the right way.
We came to the conclusion that the first level wasn't exiting enough, there wasn't enough unique elements and fun things to discover. Together with an artists i concepted a theme and it's respective visual elements that we could use as unique level ingredients. with those elementes I sketched out a new level which ended up being the final level in the game.
.png)
First iteration level
Since we had no artists during Pre-production (other than 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.
Visual scripting
Placeholder functionality
Because our programmers were very busy figuring out how to make animations work with the marble, I created some placeholder functionality that had the same end result. By doing this we didn't have to wait to start playtesting and gather feedback.
UX prototyping
Part of design philosophy was to have a very interactive playfield, thinking of something like the playing field of Hearthstone. I concepted and prototyped multiple ways to achieve this goal. I made a prototype for the playfield falling into place when the game starts and that it explodes when the game is finished. Additionally, I made the key elements of the level (musical instruments in our case) fully interactable with nice sound effects.
Mission System
Being fully responsible for the mission system, I also had to script it into existence. I made the functionality for the UI in-game, and the backend functionality that checks for the individual objectives. After linking them together I finished the whole mission loop and implemented it in the level.
​
​
Mission System
I made a big part of the level design focused on having multiple paths in the level ranging from easy to extreme in terms of difficulty. The mission system supported this level design approach and made each level more valuable and really uplifted the gameplay experience. Having all the paths in one level made players look at the level from many different angles which worked beautifully with the type of game we were making.
Playtesting
Preparing
Our target audience was a quite specific age group and we had a nice playtesting day planned, but that had to be prepared for. I made a playtesting forum for us to fill in during the testing session, as well as a tiny guide on how to playtest (since there were artists and programmers that joined us to do the playtesting)
Besides that I also made a nice results document in which we could analyze the testing data and make conclusion from the results.​
Target Audience
As mentioned before, we had a big playtesting day with our target audience. One of our team members managed to get us a playtesting session at a local scouting group where they volunteered. We went there as a team and set up everything for preparation and did playtests with about 20 kids, which offered us very valuable information.
Release!
The crew that worked on Marbleground Construct had a strong bond and we had a great opportunity to show off what we made during the playtesting day at our University. We received the main booth for the first years and got very positive comments from our peers, the teaching staff and industry visitors from various companies.
Personal Post-Mortem
Other Projects
My stance in the project:
Even though I joined the game later into the project, I feel like I had a ton of impact on the project by staying very flexible regarding my tasks and skillset. The early concepts I made helped the team pull together a strong vision that lasted throughout the whole project that ended up very successful (at least in my eyes).
​
My biggest mistakes:
Through regular peer reviews I realized that I wasn't giving the team enough space to think and concept with wild ideas. I was often too quick in disposing of idea's since in my head they would never work instead of first talking enough to make me truly understand what they ment.
My approach to the leveldesign changed often and resulted in a very messy pipeline. I should've set out a more specific plan for level creation. If I had done that we might've been able to actually create more then one level for our vertical slice.
​
My best learnings:
Marbleground Construct was my first project participating in the four cycles of game development. I've learned that being open to tasks of all kinds outside your specialization can be very valuable to a project, especially in a smaller project this flexibility is very valuable. The project has also acted as the initial lure for me to specialize in level design, which has helped me greatly in picking a career path.
