Game development studios often face the challenge of creating a large number of diverse characters within tight deadlines. Manual character creation can be time-consuming and resource-intensive, leading to delays in game development cycles. Additionally, ensuring character diversity and uniqueness is crucial for creating engaging gameplay experiences.
Factory provides a solution by automating the generation of unique 3D characters while offering customization options to meet specific design requirements. Its robust randomization algorithms allow developers to quickly generate a wide variety of characters by adjusting parameters such as body proportions, facial features, clothing styles, and animations.
01
Objectives and Research
The primary objectives for Factory were to enhance productivity by automating character generation, provide extensive customization options to meet diverse design needs, and ensure ease of use to minimize the learning curve for developers. To achieve these goals, we conducted user research focusing on game developers and character artists.
We started with interviews to understand the specific pain points and requirements of our target users. Game developers emphasized the need for a tool that could produce a large variety of characters quickly, without sacrificing quality. Character artists highlighted the importance of customization to ensure that generated characters could fit the specific aesthetic and narrative requirements of their games.
Although this app is primarily built for RTFK, our interviews revealed that it will also benefit creatives from other industries. We developed personas representing our primary users based on these insights.
→ Game Development
Quickly generate diverse NPCs and player avatars to populate game worlds.
→ Animation and Film
Create varied background characters to add depth and realism to scenes.
→ Virtual Reality
Enhance VR experiences with a variety of unique characters.
→ Prototyping and Design
Rapidly prototype character concepts for projects.
We mapped out the user flow and identified these key stages:
(1) User imports or creates Parts in Factory. (2) User specifies population rules in the rules editor. (3) User generates the population. (4) User visualizes the population and curates. (5) User exports the population, e.g. into Hydrogen.
This process involved working closely with the Chief Product Officer, engaging with 3D designers to understand their pipeline, and breaking down the journey into the most relevant stages to pinpoint critical touchpoints where users needed the most support and interaction.
Factory creates and curates *populations* of creatures. Each such population will represent variants of a “character”. To wit, the specification for such a population will consist of several *groups* such as “head”, “torso”, “legs”, “hat”. Each group will in turn contain a selection of options for *parts* which can be used to fill a segment for a particular creature. The definition for an instance of a character will contain a single part from each and every group.
04
Usability Testing and Iteration
Usability testing played a crucial role in refining the application. We conducted task-based testing sessions where artists performed specific tasks using the prototypes. This helped us identify areas for improvement and gather qualitative feedback on the overall user experience. In addition to these sessions, we had weekly meetings with RTFKT where we iterated the application based on their feedback. This collaborative approach led to multiple changes in the app, significantly enhancing its market fit.
