Kagemusha - Collaboration

A collaboration with Game Design & Programming

 

This project involved creating the complete sound and music experience for a stealth VR game with a nostalgic twist—visually and tonally inspired by the PlayStation 1 era. The game blends stealth with hack-and-slash combat mechanics, targeting male players aged 18–35. I was solely responsible for audio elements: environmental ambience, interactive SFX, adaptive music, and the overarching score. Using Logic Pro, Pro Tools, Wwise, and Unity, I worked independently on sound while collaborating remotely with the development team. The primary goal was to evoke classic 90s ninja nostalgia while leveraging VR’s immersive capabilities, consistent with findings that responsive spatial audio enhances presence (Begault, 2000).

​

To build the game's sonic identity, I studied soundtracks from Ninja Gaiden, Assassin’s Creed VR, and historically influenced scores like House of Flying Daggers. I researched traditional Japanese instruments (e.g., koto, shakuhachi, and taiko) to ground the aesthetic in cultural context (Malm, 1996). To modernise the palette, I blended these with contemporary production styles such as phonk and trap, aligning with recent analyses on genre-crossing dynamic game music (Collins, 2008).

​

The low-poly visual design prompted an audio aesthetic to match—I intentionally degraded sounds using bitcrushing to simulate 90s console fidelity and reduce runtime load (Murphy, 2013). Environmental and interactive sounds were shaped with subtractive synthesis in Vital: layered fabric recordings for weapon effects and pitch-shifted nature recordings for ambience. I particularly refined enemy footsteps on stone by combining multiple texture layers and adding randomised pitch and stereo variations to enhance realism and avoid repetition.

​

Musically, I opted for an adaptive scaffolding structure. Instead of one continuous track, I built four ~90s loops, each broken down into stems (melody, rhythm, bass, and pad). These were distributed across Unity emitters, enabling dynamic layering based on player location. This approach—similar to vertical layering—allows smooth transitions and prevents abrupt cuts (Heather, 2012).

​

Organisation in Wwise was essential. I adopted a clear naming convention to prevent naming conflicts and simplify event management in Unity (Audiokinetic, 2021). This workflow reduced confusion and errors when linking triggers.

​

Following traditional Japanese musical structure (ensemble–solo–ensemble), I featured instruments individually before returning to full texture. Daytime themes used sparse, airy instrumentation of koto, while night themes introduced drones and low taiko for suspense. Upon detection, a percussion-forward remix of the night theme triggered—using stingers like sharp woodblocks as musical cues. I implemented adaptive transitions in Wwise via states and containers, allowing smooth layering and a responsive musical experience.

​

Despite these preparations, not all audio features made it into the final game. Personal setbacks and communication breakdowns with the primarily Unreal-based development team delayed full middleware integration. I pivoted by writing a Unity script to manage random emitter playback when Wwise container logic failed, ensuring variation and avoiding repetitive loops. My prior coursework in Wwise served as a reference framework to prevent implementation issues and stabilise integration efforts (Smith, 2019).

​

Reflecting on the development process, the greatest difficulty lay in insular team communication. As the audio specialist joining mid-project, I had limited influence on design decisions and was often left out of milestone planning. This isolation made it difficult to align sound with evolving game mechanics. It emphasised how crucial early, ongoing collaboration is in game development (Ruh, 2008).

​

Still, I’m proud of the adaptive audio system I built — early testers noted how well the sound adapted to their movement and transitions between stealth and combat. One peer commented, “The audio responded naturally to what I did, making me feel part of the VR world.” This positive feedback underscored the impact of immersive sound design.

​

Going forward, I will structure clearer communication pathways and ensure audio is integrated from the start. I’ll prototype middleware integration early and build iteration time into the schedule. I also aim to deepen my understanding of Wwise’s container types and Unity scripting and to maintain strong documentation for easier teamwork. Overall, the project enhanced my technical skills in synthesis, adaptive music, and engine integration—and reinforced that sound must drive immersion, not just complement it.