Piwi and the Floating Mountains’s development began during my time at GriN but was cut short when I decided to leave the company and transition into freelance consulting. It had a brief revival (during the ‘Sonic Death Monkey’ days) while I made a few minor tweaks to the engine and improved performance. However, due to limited interest—and perhaps my own lack of marketing skills—the project came to an end around early 2008.

Piwi and the Floating Mountains follows the adventures of a little chicken named Piwi, who lives in the floating mountains high above the clouds. His world is under threat from various ‘evil’ animals, and to save the kingdom, Piwi must search for three hidden keys in an open-world setting.
In hindsight, Piwi and the Floating Mountains was loosely inspired by another arcade game that left a big impression on me—The New Zealand Story (also known as Kiwi Kraze)—developed and published by Taito in 1988.

When developing a game of this scale—especially given the technical constraints of the early 2000s, during the Director 8.5 era—achieving smooth gameplay on contemporary hardware was a top priority. Beyond design, story, audio, and music, there were substantial technical considerations and challenges, especially around optimizing performance for a fluid gaming experience.

A key factor in maintaining smoothness was managing the polygon count in 3D models. Higher polygon counts allowed for more detailed, realistic models, but they demanded more processing power, which could lead to lag or choppy graphics on less powerful machines. By carefully controlling the polygon count, we could ensure the game ran well while still providing visually rich content.

Keeping track of the poly-count is one of the many ways to optimize 3D content as illustrated with the Piwi-model from the Piwi and the floating mountains game. The polygonal evolution of 3D characters has come a long way. 

Another crucial tool was the Level of Detail (LOD) modifier in Director. LOD allowed us to dynamically reduce the polygon count for models viewed from a distance, preserving the appearance of detail up close while optimizing performance as objects moved further away. This feature, which players could adjust in the game’s settings, enabled smoother rendering and gameplay by reducing processing demands, especially on lower-spec systems. Using LOD strategically helped balance visual quality with performance, making it possible to deliver a visually engaging game experience without sacrificing smoothness or playability.

In addition to using the LOD (Level of Detail) modifier, we implemented fog effects to enhance the sense of depth and immersion while subtly masking the work of the LOD adjustments. Fog in Director was a powerful tool to create atmospheric layers, allowing distant objects to gradually blend into the background. By adjusting the fog’s density and color, we could make distant areas appear hazy, naturally fading out details that the LOD modifier was reducing. This not only made the environment feel more expansive but also smoothed transitions between different detail levels, preventing abrupt shifts that might otherwise distract the player.

Another important optimization technique was adjusting the game’s viewport or resolution. By reducing the rendering size of the viewport, we effectively lessened the processing load, as fewer pixels needed to be drawn on the screen, significantly improving frame rates. These combined techniques—LOD, fog, and optimized viewport—allowed us to deliver a visually engaging experience that ran smoothly on contemporary machines, striking a balance between graphical fidelity and performance.

I also developed a custom particle behavior using Director’s particle capabilities to add extra visual effects, enhancing the game’s immersive quality. For instance, when Piwi dashes through water, dynamic ripples appear, and on sandy surfaces, dust clouds trail behind him, adding a sense of realism to his movements. To streamline the creation and testing of these effects, I built a standalone application where I could upload one or more images to use as particle textures. The application featured a UI that allowed me to adjust particle settings in real-time and export the configurations directly into the behavior script. This approach not only saved time but also allowed me to immediately see how each effect performed visually, refining the overall look and feel of the game without needing to constantly recompile or reload scenes.

Particle systems undoubtedly bring a new level of realism and vibrancy to games—like the water ripples that appear when Piwi dashes through puddles. Although particle effects could be quite processor-intensive, the results back then were impressive, especially given the capabilities of contemporary hardware. They allowed for dynamic visuals that ran smoothly, enhancing immersion in a way that felt cutting-edge at the time.

Van Ziepe

Een reactie achterlaten

Je e-mailadres zal niet getoond worden. Vereiste velden zijn gemarkeerd met *