Get the latest Education e-news
  • Postmortem: Dead West

    - Patrick Dietz

  • What Went Wrong

    1. Scope and Scope and Scope

    The fiction presented above is most of what's presented to the player in the final product, aside from voice acting that added flavor and instruction to the game. There were plans for a more involved story. Perhaps predictably, fiction was among the first elements to hit the cutting room floor, so to speak, but it wasn't alone. Originally, we designed for three campaign maps, a day camp to handle the character building and inventory elements, and at least two more bosses. Over the seven-month development cycle, this was slowly whittled down through a series of agonizing meetings and milestones as we became more realistic about the output we could pursue.

    As I mentioned earlier, we made smart decisions and good compromises, for the betterment of the game, but it's never encouraging to lose something that could add that much more to the to the fiction, gameplay and production value. We maintained the vision for the game, but that vision got smaller as the weeks went by. At the end, we had a really solid prototype for a much larger game, consisting of a very unique single campaign level, a single boss battle, and a survival mode.

    As the design lead, I did my best to keep the scope realistic, but I definitely overreached. In a sense, this was also the case with our gameplay. Nearly every system that we designed made it into the game: wagons and upgrades, inventory, special ammo types, and skill trees to name a few. We did have a pretty great design for special abilities such as "Fan the Hammer," a move that allowed Jack Dixon to unload his clip in a rapid, 45 degree arc ahead of him. Truth was, even without the time pinch, our game was complicated enough. These abilities could have been really deep and fun, particularly when combined with our ammo system, but at that point in the project all signs were pointed toward simplifying the design.

    Indeed, if we had designed the game's fiction and functionality from the start with the appropriate scope in mind, the entire game would probably be easier to communicate. As you might imagine, it's a bit tough to quickly communicate what an old west, survival horror, third person, tower-defense, action-rpg, shooter is really like. It's not the team didn't know what we were building; in context all of those elements gel pretty well, but there was an identity crisis from the start, particularly when trying to effectively distill the Dead West experience such that others would really understand what we were working toward.

    2. Animation and Capability Gaps

    Even in a large company, finding the right people at the right time to meet capability gaps is an enormous challenge. For all the aforementioned dedication and talent on this team, it's impractical to expect that we'll be able to meet every need with a limited pool of student talent split amongst several teams. Five games made it through the initial pitches into pre-production, and three went all the way through development. Next to previous classes, our only basis for comparison, this left less manpower overall to begin with.

    Animation was perhaps the most pervasive of these gaps throughout the development cycle. Our game necessitated more animations than the other two projects, and with only one adept, but overworked artist dedicated to the task, this became a major bottleneck to content.

    It wasn't only a manpower issue. In addition to learning the intricate process of animation, our artists and programmers were learning to translate that data from Motionbuilder into our lesser-known graphics engine. The fledgling, German-built, Trinigy engine had a number benefits out of the box in addition to the source code gave our engineers awesome flexibility, but it lacked the robustness in features and documentation often found in more established commercial engines. Combined with our collective lack of experience and the fickle nature of animations, it was very nearly disastrous.

    Animation is as much a specialized technical process as it is an art. Even our most proficient artists were still learning the skill which was complicated by motion capture data that needed to be cleaned up, and a nubile engine. It became much more than we could ask of a single animator, and it required more than a few long nights and headaches from our artists and programmers to get them in the game and feeling natural.

    Could we have done it differently? Hand-keying the animations may have worked, but would likely have felt significantly less natural, particularly given our time constraints. We're proud of getting so many animations in the game and looking great, but this is a capability gap that haunted us throughout the project, and lead to a number of our hardest cuts for scope and time. This is one pipeline I wish we had nailed down earlier, planned around more realistically, and had more manpower and experience to draw on.

    3. Accountability and Motivation

    The Dead West team was highly motivated most of the time and we had nothing beyond the success of our project to motivate us. There were, of course, tough periods. It's impossible to sustain that level of effort forever; there were periods of burnout for every one of us. At times folks lost heart and at others we simply had to work around serious personnel issues.

    Part of the industry that FIEA cannot replicate is the motivation and accountability inherent in a competitive workforce: the promise of upward mobility and the threat of censure or termination. The worst consequence for letting work or quality slide, procrastination or other productivity drains, was letting the team down and the end product suffer. The types of individuals not motivated by this will inevitably lose the respect of their team and be unable to function effectively on a project of this nature.


comments powered by Disqus