Position-Only Shading Pipeline Reduces Memory Traffic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current graphics processing systems face high memory traffic, which increases power usage and reduces performance, and existing techniques do not effectively manage memory traffic in graphics processing.
Innovation Solution
Implementing a position-only shading pipeline that sorts triangles based on their positions and depths, using a cull pipe for initial culling and a replay pipe for efficient vertex shading, reducing memory traffic by processing triangles in bins and using representative coordinates to optimize binning and rasterization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If traditional full vertex shading is executed for all triangles, then complete shading coverage is achieved, but memory traffic to color and depth buffers increases significantly
Solution Approach 1:
The patent applies preliminary action by performing position-only shading and culling tests before full vertex shading to identify and discard triangles that will not be rendered. The cull pipe executes position-only shading and performs backface culling, view frustum culling, and zero coverage culling in advance, recording surviving triangles in a bit-stream. This preliminary filtering reduces the number of triangles that proceed to full vertex shading, thereby reducing memory traffic to color and depth buffers and lowering power consumption.
2Quantity of substance
If position-only shading pipeline with culling is implemented, then memory traffic is reduced, but processing complexity increases due to multiple pipes and sorting operations
Solution Approach 1:
The patent applies segmentation by dividing the graphics processing pipeline into two separate pipes: the cull pipe and the replay pipe. The cull pipe handles position-only shading and culling tests, while the replay pipe executes full vertex shading for surviving triangles. This segmentation allows each pipe to be optimized for its specific function, reducing overall memory traffic despite the increased structural complexity of having multiple specialized processing paths.
Data Source
AI summary
In the cull pipe, positions of the vertices of a triangle have already been computed and these coordinates may be exploited by taking and sorting triangle groups based on these coordinates. As one example, all the triangles in a tile may constitute a group. The triangle groups are sorted into bins. Within each bin the triangles are sorted based on their depths.