Adaptive Hierarchical Primitive Listing for Tile-Based Graphics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In tile-based graphics processing systems, the varying sizes of primitive lists for different tiles and frames lead to inconsistent memory requirements, making it challenging to optimize memory usage and reduce memory bandwidth and power consumption.
Innovation Solution
A method and apparatus that determine sub-regions for listing graphics primitives based on a cost function, assessing the cost of listing primitives in different sets of sub-regions, and selecting the most cost-effective set for inclusion in primitive lists, allowing for efficient memory management and reduced bandwidth and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If primitives are listed for all possible tiles to ensure complete coverage, then rendering completeness is improved, but memory bandwidth and power consumption increase
Solution Approach 1:
The patent changes the parameter of primitive list organization by introducing hierarchical levels (tile level, block level, frame level) instead of a single flat structure. This allows the system to select the appropriate level of detail based on rendering requirements, reducing unnecessary memory operations while maintaining completeness.
Solution Approach 2:
The patent segments the primitive listing system into multiple hierarchical levels: individual tile primitive lists, block primitive lists (grouping multiple tiles), and frame primitive lists (grouping multiple blocks). This segmentation allows selective processing and reduces the total number of memory operations required.
2Reliability
If primitives are listed for all possible tiles to ensure complete coverage, then rendering completeness is improved, but memory bandwidth consumption increases
Solution Approach 1:
The patent merges primitive lists at higher hierarchical levels (block and frame levels) to avoid redundant storage and transmission of primitive data across multiple tile lists. This combining reduces memory bandwidth consumption by eliminating duplicate primitive listings while maintaining rendering completeness through the hierarchical structure.
3Measurement precision
If primitive lists are prepared for each individual tile, then rendering precision is improved, but device complexity increases
Solution Approach 1:
The patent introduces dynamic selection of hierarchical levels based on rendering needs. The system can adaptively choose to process at tile level for high precision requirements or at block/frame levels for lower precision needs, making the data structure flexible rather than statically complex.
Solution Approach 2:
The patent adds a hierarchical dimension to the primitive listing system, organizing data in multiple levels (tile → block → frame) rather than a single flat dimension. This additional organizational dimension reduces complexity by grouping related data while maintaining access precision when needed.
4Manufacturing precision
If varying sized primitive lists are used for different tiles, then rendering accuracy is improved, but memory management difficulty increases
Solution Approach 1:
The patent creates a universal hierarchical data structure that can accommodate varying primitive list sizes at different levels. The same hierarchical framework handles both small tile-level lists and large frame-level lists uniformly, simplifying memory management while preserving rendering accuracy through selective level processing.
Data Source
AI summary
A primitive listing and sorting arrangement for a tile-based graphics rendering system in which primitive lists can be prepared for at least two different sets of sub-regions of the render target area. Two or more alternative solutions for listing the primitive for rendering using the sub-regions of at least two of the at least two different sets of sub-regions of the render target area that primitive lists can be prepared for are determined (S62). A cost function is then used to assess the cost of listing the primitive for each determined alternative primitive listing solution (S63) and the primitive listing arrangement to use is then selected based on the cost assessment (S64). Embodiments use an adaptive cost function that can be modified in use based on analysis of the costs for the primitive listing process actually being experienced in use.