Hidden Culling in Tile-Based Graphics via Depth Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In tile-based computer graphics systems, non-visible objects are not efficiently culled before being stored in display lists, leading to unnecessary memory usage and bandwidth waste, as existing methods like partial rendering are slow and inaccurate.
Innovation Solution
A method that subdivides the rendering space into regions, derives depth values for objects, and updates depth thresholds to cull non-visible objects before they are included in display lists, allowing for efficient culling without requiring partial rendering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional z-buffering hidden surface removal is performed for all primitives, then visibility accuracy is maintained, but processing time and memory bandwidth increase significantly
Solution Approach 1:
The patent applies preliminary action by performing depth threshold comparisons before full z-buffering processing. Depth thresholds are pre-calculated for each tile based on extreme depth values, allowing primitives to be culled in advance without complete visibility analysis. This preliminary depth filtering removes obviously hidden primitives before they enter the expensive z-buffering pipeline, maintaining accuracy for visible primitives while reducing overall processing time.
Solution Approach 2:
The patent segments the rendering process into two distinct phases: a preliminary depth threshold comparison phase that quickly identifies and culls hidden primitives, and a subsequent z-buffering phase that processes only potentially visible primitives. This segmentation separates the filtering function from the detailed visibility determination, allowing each phase to be optimized independently and reducing the total number of primitives requiring full processing.
2Loss of energy
If depth threshold culling is applied before display list creation, then memory bandwidth is reduced, but culling accuracy decreases
Solution Approach 1:
The patent applies partial action by implementing depth threshold culling as a supplementary filter alongside traditional z-buffering, not as a complete replacement. The depth threshold test performs a partial visibility assessment that is less accurate than full z-buffering but sufficient to eliminate obviously hidden primitives. This partial filtering reduces memory bandwidth usage while maintaining acceptable culling accuracy for the overall rendering pipeline.
3Measurement precision
If tile-based rendering is used for high resolution images, then rendering precision is improved, but the number of tiles and memory requirements increase
Solution Approach 1:
The patent applies preliminary action by calculating and storing depth thresholds for each tile before the actual rendering process. These pre-calculated depth thresholds serve as compact filters that reduce the amount of geometric data and texture information that needs to be loaded into memory during rendering. By performing this depth analysis in advance, the system maintains high rendering precision through fine-grained tile processing while reducing peak memory requirements through early culling.
Data Source
AI summary
A method and system is provided for culling hidden objects in a tile-based graphics system before they are indicated in a display list for a tile. A rendering space is divided into a plurality of regions which may for example be a plurality of tiles or a plurality of areas into which one or more tiles are divided. Depth thresholds for the regions, which are used to identify hidden objects for culling, are updated when an object entirely covers a region and in dependence on a comparison between a depth value for the object and the depth for the region. For example, if the depth threshold is a maximum depth threshold, the depth threshold may be updated if an object entirely covers the tile and the maximum depth value of the object is less than the maximum depth threshold.