3D Graphics Rendering Area Prediction for Buffer Efficiency

The patent divides the screen into multiple regions (first region and second region) based on object movement predictions. Instead of rendering the entire screen when a scene changes, only the second region containing moving objects is rendered, while the first region reuses rendered data from the previous frame. This segmentation approach maintains display completeness for moving objects while significantly reducing unnecessary rendering time and memory usage.

The patent performs preliminary prediction of object movement between frames to determine which screen regions will be updated. By analyzing object trajectories and predicting future positions, the system pre-determines the update region before actual rendering occurs. This preliminary action enables efficient resource allocation by preparing only the necessary rendering data for predicted movement areas, avoiding wasted computation on static regions.

The patent segments the rendering process into two parts: rendering the first region (containing stationary objects) from previous frame data, and rendering only the second region (containing moving objects) from current frame data. This segmentation reduces memory space consumption by loading and processing only the necessary second region data, while maintaining display completeness through the combination of both regions.

The patent extracts and processes only the essential rendering data for moving objects in the second region, while excluding static objects from current frame rendering. By taking out only the necessary moving object data and using previous frame data for stationary objects, the system reduces memory space requirements while preserving display completeness for all objects.

The patent segments the rendering operation into selective updates: the first region uses rendered data from the previous frame without reprocessing, while the second region performs new rendering operations for moving objects. This segmentation maintains update accuracy for all objects through the combination of reused and newly rendered data, while significantly improving processing rate by eliminating repeated operations on static regions.

The patent applies partial action by rendering only the necessary second region containing moving objects rather than performing complete rendering of all regions. This partial rendering approach maintains sufficient update accuracy for displayed objects while dramatically improving processing rate by avoiding excessive rendering operations on static areas.