GPU Stroke Rendering via Point Primitives for Display Performance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Display devices face performance deterioration and increased power consumption when rendering a large number of strokes due to the need for multiple rendering pipelines, especially when handling intersecting strokes with different styles and pressures, which requires a significant amount of memory and processing power.
Innovation Solution
A display device and method utilizing a graphic processing unit (GPU) that renders strokes using point primitives within a redraw region, with conversion and reverse conversion information calculated based on the maximum radius of the strokes, allowing for efficient rendering by minimizing the number of rendering pipelines and optimizing memory usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a large number of strokes are rendered using traditional rendering methods, then the strokes can be displayed, but the rendering time increases significantly and use performance deteriorates
Solution Approach 1:
The patent divides the display area into multiple regions and processes strokes in parallel across different GPU cores. Each GPU core handles a specific region, allowing simultaneous rendering of multiple strokes. This segmentation enables the system to process large numbers of strokes efficiently by distributing the workload across multiple processing units rather than processing them sequentially.
2Manufacturing precision
If multiple rendering pipelines are used to handle intersecting strokes with different styles, then stroke accuracy is improved, but the number of processing pipelines increases and power consumption rises
Solution Approach 1:
The patent performs preliminary classification of strokes by style and depth information before rendering. Strokes are grouped into categories based on their stylistic properties, and a single rendering pipeline is configured to handle multiple styles within each category. This preliminary organization allows the system to process intersecting strokes of different styles efficiently without requiring separate pipelines for each style, thereby reducing overall pipeline complexity while maintaining rendering accuracy.
3Productivity
If traditional rendering methods are used for strokes, then strokes can be rendered, but memory usage and processing power requirements increase significantly
Solution Approach 1:
The patent changes the rendering parameters by utilizing depth information and style classification to optimize the rendering process. By adjusting rendering parameters based on stroke characteristics (such as rendering only visible portions of strokes considering depth information), the system reduces unnecessary processing and memory operations. This parameter optimization decreases both power consumption and processing requirements while maintaining rendering throughput.
Data Source
AI summary
A display device is provided. The display device includes: a graphic processing unit rendering strokes; a display displaying an image including the rendered strokes; and a processor identifying a redraw region when a redraw event occurs on the image, and providing rendering information for rendering strokes included in the redraw region to the graphic processing unit, wherein the graphic processing unit renders the strokes included in the redraw region using point primitives on the basis of the rendering information.