Spatial Computing & RealityEngine

The web is evolving beyond flat screens. Koda Zenith includes RealityEngine, a native spatial computing stack that allows developers to build immersive AR/VR experiences using familiar web paradigms.

The .zen Language

We introduced the .zen extension specifically for defining spatial realities. It combines reactive state management with low-level 3D rendering instructions.

• Reality DSL: Declarative syntax for defining 3D scenes.
• Zero-Latency Interactions: Sub-1ms input-to-photon latency on supported spatial hardware.
• Physical Integration: Direct access to spatial tracking, hand gestures, and environmental mapping.

Building for the Immersive Era

RealityEngine doesn’t just render 3D models; it manages the entire spatial context, including physics, spatial audio, and haptic feedback.

reality App() {
  state {
    rotation: Degree(0)
  }

  render {
    Scene([
      AmbientLight(intensity: 0.5),
      PointLight(position: [2, 5, 2]),
      
      // Load and animate a native Zenith model
      Model('./assets/industrial_core.glb') {
        transform: rotateY(state.rotation),
        onHover: () => state.rotation.animate(to: 360, duration: 2s)
      }
    ]);
  }
}

Reality-First UX

Zenith’s spatial components are designed to be “World Anchored”. They exist in 3D space rather than being fixed to a viewport. This opens up new possibilities for:

• Industrial Dashboards: Digital twins of real machinery.
• Collaborative Workspaces: Multi-user shared realities.
• Data Visualization: Exploring complex datasets in 3 dimensions.

Native Performance

By bypassing the traditional browser DOM and rendering directly to the Metal or Vulkan buffers, RealityEngine achieves performance that was previously only possible in native game engines like Unreal or Unity.