In December 2010, we began our project by deciding upon what would be our main mechanic: a shifting environment. The vision was to have the setting serve as the main source of opposition through shifting walls that hindered the player's progress, and twisting halls that led them astray. It was a puzzle game, and the goal was simply to reach the end. Further discussion endowed the high concept with a more concrete visual component: the game would take place in tunnels, underground. A narrative was then applied: the game is set in the catacombs of "Neo Rome." At the center of it all, the actual mechanic was born: a decoy device that tricked the environment by simulating the bio-thermal signature of its user.
Design, Art, and Narrative aspects were rapidly developing, and the push-and-pull between them was inevitable. We were ambitious, and planned to have three tiers in the catacombs that become more advanced and more volatile as the player moved towards the surface. Concept art was generated for the three levels at the same time that puzzle mockups were designed and the pre-writing for the story was crafted.
The production method we used was a variation of the waterfall method. Each week, tasks were assigned to the team members. It was a fluid process, and we held meetings up to twice a week to share our progress and to ensure everyone was on the same page.
The three stages of development, for us, weren't so cleanly divisible. But it was certainly a great milestone when we created a functional prototype of the game's decoy mechanic in a reactive environment. This signaled the beginning of the next phase.
One aspect of our pre-production work that proved particularly helpful was the detailed documentation we generated. Between our design docs, art bibles, story bibles, and production docs, we had laid out a fairly clear path before us. Not all things could be anticipated, and many things changed along the way, but we prevented ourselves from wandering in complete darkness as we continued development.
Our one-man tech team proceeded with breakneck speed. And while design and art soldiered on, the script was one of the first things that needed to be completed, so as to allow for recording and audio remastering. Although the early puzzle blockouts and level map changed dramatically by the end, they proved instrumental in planning out narrative events. This was how the narrative was built: a story onto a game space.
The narrative team soon joined the art team in constructing the 2D and 3D assets that would serve as the building blocks for the level, each infused with futuristic Roman architecture. At the same time, the design team continued to refine the puzzles and gameplay, applying the feedback gathered from our bi-weekly playtest sessions. When enough assets were completed, the set-dressing and lighting process began. Here was when the Underground of New Rome was truly born. But time quickly grew short.
Mid-way through production, producers became aware of a grave issue: there was simply too much to do. We were (as is apparently typical) running behind schedule, but our deadlines could not change. So, we needed a plan to cut. In the end, three end puzzles did not make this version of the game, and the lower priority models remained unmade. The silver-lining: we had at least, to some degree, finished the game.
Yet at the same time, we had only really just begun.
With our first true development cycle ended, we ushered in a second one marked by little else except polish, polish, polish. Each spanned 10 weeks, and although building something more or less from the ground up is no small feat, the most noticeable changes to our project took place in post-production. Our goal was unchanged; we wanted to make an awesome game experience.
No discipline was spared. The design team implemented the puzzles which were previously cut, and were able to hold more playtesting sessions to improve upon the design of existing ones. The art team revised a few of the setpiece models, and continued to work through the lofty asset list. The narrative team wrote a whole new script that was more integrated with the gameplay, additionally introducing a clearer tutorial for the decoy device as well as non-linear elements. It was an intense and thorough make-over that gave our project a whole new life. Gone was “Romahedron” and in its place stood “Vestige.”
One of the biggest changes in our development process was the use of Scrum/Agile. We planned out two-week sprints, and used spreadsheets where our user stories were divided into tasks, with estimated hours at which we whittled away. This approach gave us more focus from day-to-day, as it concentrated our efforts and attention typically to just one task. Lessons were also learned about scope. With concrete numbers reflecting how much time we could devote to the project, we gained a much more realistic perspective on what could be done.
And there you have it. The game world became much more believable as an abandoned subway in a Rome-ruled future, our characters were palpable presences for players, and the story gained much needed clarity. Most of all, the experience was both challenging and fun. At the end of it all, we had in our hands a game of which we were, all things considered, quite proud.