Conservative Rasterization Error Term for Pixel Coverage Precision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional graphics processing pipelines face accuracy issues during rasterization, particularly when rendering three-dimensional objects at different resolutions, leading to inconsistent coverage results and inefficient memory allocation due to 'snapping' of primitives to a grid, resulting in over- or under-allocation of frame buffer memory.
Innovation Solution
The implementation of conservative rasterization using an error term, where edge equations are modified by an error term to enlarge primitives, ensuring accurate identification of pixels intersecting the primitives across varying pixel resolutions, thereby compensating for precision discrepancies and snapping inconsistencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If fixed point rasterization calculations are performed with fixed fractional-pixel precision, then computation speed is improved, but manufacturing precision deteriorates due to snapping to a grid
Solution Approach 1:
The patent applies preliminary action by pre-calculating an error term based on the primitive geometry and resolution parameters before the actual rasterization process. This error term is then added to the fixed-point rasterization results to compensate for the snapping inaccuracies, thereby maintaining high precision without sacrificing computation speed during the main rendering pipeline.
2Productivity
If low resolution rasterization is performed first to determine coverage, then productivity is improved, but manufacturing precision deteriorates due to coarser grid snapping
Solution Approach 1:
The patent implements feedback by using the error term calculated from high-resolution reference data to correct the low-resolution rasterization results. The error term captures the discrepancy between coarse and fine grid alignments, and this information is fed back into the low-resolution process to adjust pixel coverage determination, ensuring accurate memory allocation decisions.
3Device complexity
If primitives are snapped to a grid for efficient processing, then device complexity is reduced, but measurement precision deteriorates due to alignment errors at different resolutions
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the error term based on the specific primitive geometry, resolution level, and grid alignment characteristics. Rather than using a fixed correction value, the error term is computed as a function of these parameters, allowing the system to maintain simple fixed-point processing while achieving high precision through parameter-dependent correction.
Data Source
AI summary
A system, method, and computer program product are provided for conservative rasterization of primitives using an error term. In use, an edge equation is determined for each edge of a primitive, the edge equation having coefficients defining the edge of the primitive. Each edge of the primitive is shifted to enlarge the primitive by modifying coefficients of the edge equation defining the edge by an error term that is a predetermined amount. Pixels that intersect the primitive are then determined using the enlarged primitive.