CPU-GPU Collaborative Rendering for Virtual Scene Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In computer graphics, the graphics processing unit (GPU) often experiences higher resource occupancy rates than the central processing unit (CPU) during rendering, leading to increased pressure and reduced efficiency, especially when GPU resources are insufficient, causing the CPU to remain idle.
Innovation Solution
A collaborative processing method where the CPU takes on some of the GPU's computing tasks by determining the required image data and configuring rendering functions to perform computing, thereby reducing GPU pressure and improving efficiency without increasing hardware costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the graphics processing unit performs all rendering computing tasks independently, then the rendering quality and visual effects can be maintained, but the GPU resource occupancy rate becomes excessively high, leading to increased running pressure and reduced overall system efficiency
Solution Approach 1:
The patent segments the rendering computing tasks by separating compute shader tasks from the traditional rendering pipeline. The compute shader tasks are extracted and executed independently on the GPU, while the main rendering pipeline continues to operate. This segmentation allows the GPU to handle different types of computations simultaneously, reducing resource occupancy pressure while maintaining rendering quality.
Solution Approach 2:
The patent introduces a new execution dimension by utilizing the GPU's compute shader capability, which operates in a different computational dimension compared to the traditional rendering pipeline. This allows parallel execution of computing tasks alongside rendering tasks, effectively utilizing GPU resources in multiple dimensions simultaneously and improving overall system efficiency.
2Productivity
If the central processing unit is used to perform rendering computing tasks, then the GPU resource occupancy rate decreases and running pressure is reduced, but the CPU becomes overloaded and may bottleneck the rendering performance
Solution Approach 1:
Instead of offloading compute shader tasks to the CPU when GPU resources are constrained, the patent inverts the conventional approach by keeping compute shader tasks on the GPU and utilizing the GPU's general-purpose computing capability. This inversion allows the GPU to handle both traditional rendering and compute tasks, eliminating the need to transfer workload to the CPU and avoiding CPU bottlenecks.
3Reliability
If compute shader tasks are offloaded to the CPU for execution, then GPU resource occupancy is reduced, but data synchronization overhead increases and computing efficiency decreases
Solution Approach 1:
The patent utilizes the GPU's copy engine to transfer data between different memory spaces and processing units within the GPU architecture. This internal copying mechanism eliminates the need for CPU-GPU data synchronization, reducing synchronization overhead and time loss while maintaining balanced resource utilization.
Data Source
AI summary
A rendering processing method includes determining, in a process of rendering a virtual scene, image data required by a graphics processing unit to perform the rendering the virtual scene, configuring a rendering function in a central processing unit according to the required image data, and performing rendering computing through the configured rendering function in the central processing unit according to the required image data, to obtain a computing result. The method further includes outputting the computing result to the graphics processing unit, the computing result being configured for use by the graphics processing unit during the rendering the virtual scene.