The Perfect Computer
Narrative & Level Designer | Solo Project | Fallout 4
Mirelurks and Malfunctions
"The Perfect Computer" is a single-player Fallout 4 level. The player meets Leo, an intrepid explorer, who tells them about Vault 18, a hidden vault that has a mirelurk problem. As the player navigates through the environment, they fight mirelurks and robots, and find out that the vault is ruled strictly over by a computer. They are then given the choice to repair the central computer, eliminating the mirelurk threat, but forcing the residents of Vault 18 back inside, or destroying the central computer, freeing the residents, but perhaps subjecting them to a worse fate. This level incorporated stakeholder feedback heavily, and the design focused on robustness and approachability.
Environmental Storytelling
The base story of high-concept science-fiction freedom vs safety is something not all players want, so the secondary supporting story, of mirelurks and robots conflicting, is emphasized throughout the level, in combat encounters, environment design, and scripted sequences like the above, where a patrolling robot is ambushed by a mirelurk.
Clear Landmarks
Whether Megaton or Diamond City, Fallout 4 has always had significant landmarks to guide the player. I wanted to emphasize that here, in both the exterior and interior of Vault 18. The exterior has the vault clearly highlighted, and secondary objectives illuminated with lights or in motion with a flag. Inside, the first thing the player sees is the final terminal, with a patrolling robot.
Integration with Base Game
On massive teams, sometimes it is hard to integrate everyone's vision into a central vision. This is doubly so when done as a modder, but I focused on and succeeded in creating unique environments that nonetheless fit into the established Fallout 4 lore and setting, whether expanding Red Rocket, or building an entirely new place of interest on the coast.
Environment Design
Vaults in Fallout 4 are frequently treated as one off jokes or raider nests. For my narrative, a core point was the people living inside Vault 18, and I took that into the environmental design, building onto the core inhuman vault aesthetic with touches of homeliness, like individuals were trying to make an abode out of what they had available.
Unique Interactions
Taking basic Fallout 4 systems, like hacking and terminals, I retained the structure but developed more intriguing methods of interacting. Above is a password entry terminal, which the player finds the passcode on a pip-boy readable chip on the corpse of a robot.
Satisfying Responsiveness
As an RPG over an action game, Fallout 4 sometimes has sequences that feel unsatisfying to a player, with the results relayed off-screen or over text, such as in the game's conclusion. When designing this level, I wanted to ensure that player's actually felt like they were engaged in a physical world and not just code, and the above example of a destructible wall is an illustration of this.
The Final Decision
The core of the story of The Perfect Computer is the player interacting with the individual characters of the vault. This culminates in a decision with direct and instantaneous feedback.
Destroy Computer
This decision results in immediate feedback, with the vault alarm blaring, and a bright red light covering the environment. Outside, the situation is even more severe, with the former inhabitants under attack from a mirelurk queen. A player that chooses this option is more interested in action, and that is what I delivered.
Repair Computer
This decision results in more dialogue, because a player that chooses this is one likely to be interested in the characters as they are and want to preserve their safety. Friendly robots appear, and current and new characters gain new dialogue lines providing the player an explanation of the future trajectory of the vault.
Postmortem
What Went Well?
​Planning.
Going into every week, I had a very distinct plan of what I wanted to get done. I had a list of fixes, a list of things to get done, and a list of future plans if I had more time. My time estimates were very accurate, and anything I wanted to finish on time got finished on time.
​​
Time management.
I made a decision that I was not going to rush around during the week and only work on the weekends. It was good, because it gave me a definite time slot to work in, and I had no wasted time of trying to get back into the flow of things. I spent approximately 10 hours every Saturday and Sunday working on Creation Kit.
​
Quest pacing.
Following from my original plan, the quest I built has a very natural flow, with the player progressing through the different areas at a steady pace and in a logical order. It feels very natural, and that result is due to efficient planning and playtesting.
Environmental design.
I built the environment with the player’s experience in mind, so all of the elements are prioritized towards the player’s view and the player’s experience. If I did not start from this baseline, I would not have been able to mold it later. It starts from the base geometry and is therefore a very important early consideration.
Asset use.
I was limited to only the base Fallout 4 assets. Within that region, I was very creative in the use of these assets. I found the big, disassembled plane, and used the pieces as landmarks in my exterior. I took the existing vault assets and twisted them into my own. When I needed spaces a little bigger, I used pillars as flooring and ceiling, as well as recreating my own catwalks with individual pieces. In all, I used the base kit cleverly and expanded it in clean manners.
Tool efficiency.
I used a lot of the default papyrus scripts, as well as efficient Creation Kit workflows. By using the tools already developed to their maximum efficiency, I was able to save a lot of time and get more done than I would have otherwise.
What Went Wrong?
The initial plan was not properly balanced in complexity.
My original plan had an overly complex interior, and a very bare exterior. I ended up expanding my exterior a lot and cutting a whole section of my interior. When designing my initial plan, it would have been better to strike a better balance.
The initial plan did not have exterior verticality.
When designing an interior space, the required inclusion of ceilings provides an easy reference for verticality. When outside, there are no such forced ceilings, and therefore in a design it is harder to visualize how exactly a space will function. Better drafting tools could assist in this, as well as an awareness of the intended purpose of a space.
Stakeholder expectations were not clearly understood.
At several points I received feedback, which I thought I fixed, but were fixed in an improper way. For instance, I was told the atrium was too small, so I widened the walls. The actual intended fix was to make the catwalks bigger, which I finally did once I understood. I would have saved development time if I understood this sooner.
Project requirements were not clearly understood.
Some areas of vagueness in the requirements I accepted without looking for clarification. For instance, I was told to integrate my quest start into the already existing Fallout 4 geometry. I did, but the actual instruction was to build my own area and then integrate that into existing Fallout 4 geometry, which is a big difference, and understanding it sooner would have saved development time.
Inconsistency with Fallout 4.
Some aspects of the developed quest were contrary to the actual design of Fallout 4. These changes were done based off of stakeholder feedback, and it would have been useful to communicate as soon as possible to be aware of the expected differences.
Tool updating.
In the final week of development, there was a Creation Kit update. I carefully did not update, but I found some critical bugs when giving my files to others to playtest. I updated, and fixed those bugs, but many more arose that were literally impossible to fix as they were inherent to Creation Kit. I think it would have been better to fix the bugs I knew I could fix, rather than hoping an update would fix them.
What Did I Learn?
Consistency in design is incredibly important.
When building something, both for the player’s experience, and the speed of development, having a consistent design is essential. This applies to the initial design document, as well. It is very important to consider this throughout the entire design process.
Exterior spaces, due to the lack of a ceiling, are important to visualize differently.
When designing exterior spaces, be sure to have some form of vertical reference, otherwise the tendency is to build a flat plane. Consider how to implement verticality from the very start. Use tools that allow one to properly describe vertical space.
You are building a level for the stakeholder and not yourself.
Stakeholder feedback is very important. The players are the ones who ultimately decide if a level is good, but a stakeholder is the one who gives a go or no-go response. Their feedback is consequently more important in terms of presence than quality.
Always be aware of the end design goal.
Know where you want the player to end at the end of what you are designing, and everything that is being built will flow smoother and more logically with that awareness.
Communicate with stakeholders on differing aspects of design.
When the design changes from the base template, be sure to communicate these changes, and ask for clarification. This saves time, ensures the final project fits the specifications, and provides immediate feedback.
Be very cautious when updating tools.
It is almost always better to deal with outdated tools than untested broken new tools. And it is very hard to revert. If things are not proven to be stable, don’t gamble.