Volumesight

Volumesight is a volumetric rendering engine designed for non-polygonal datasets, such as point clouds and block models, that exceed the capacity of traditional polygonal rendering techniques.

Scaling with Resolution

Traditional renderers often struggle as models grow in complexity. Volumesight instead scales its memory and compute usage with screen resolution rather than the total size of the model.

This allows for the visualization of large datasets with consistent performance. As the camera moves, Volumesight dynamically adjusts the level of detail to maintain fidelity.

GPU Features

Volumesight leverages modern GPU capabilities:

• Occlusion Culling: Skips rendering for data hidden behind other objects to improve performance.
• Efficient Streaming: Integrates with Dagger to fetch only the data required for the current camera position and resolution.
• VRAM Management: Uses optimized GPU memory structures to cache and reuse data, minimizing CPU-GPU transfers.

Interaction

Volumesight provides the foundation for interacting with volumetric data, including raycasting for selection and spatial queries against the rendered representation. This enables pixel-perfect interaction with massive datasets without the overhead of maintaining a separate CPU-side collision mesh for every point or block.