Low Resolution Buffer Pixel Culling in GPU Binning Pass
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current graphics processing units (GPUs) face inefficiencies in rendering pixels on displays due to the lack of effective methods for determining which image surfaces contribute to the final image, leading to unnecessary processing of occluded pixels.
Innovation Solution
A low resolution z-buffer (LRZ) is updated during a binning pass, where the GPU divides image surfaces into blocks and stores surface identifiers in a buffer corresponding to blocks of pixels, allowing for efficient rendering by determining which image surfaces contribute to the final image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the GPU processes all pixels on the display to generate pixel values, then complete image rendering is achieved, but computational resources are wasted on occluded pixels that do not contribute to the final image
Solution Approach 1:
The patent applies preliminary action by performing a binning pass before the main rendering process. During this preliminary pass, the GPU determines which image surfaces are visible and which are occluded by storing surface identifiers in a low-resolution buffer. This allows the system to identify and cull occluded pixels before investing computational resources in rendering them, thereby improving rendering speed while reducing wasted energy on non-contributory pixels
2Productivity
If the GPU divides image surfaces into blocks and stores surface identifiers in a low resolution buffer during binning pass, then occluded pixels are identified and culled efficiently, but the device complexity increases due to additional buffer management
Solution Approach 1:
The patent applies segmentation by dividing the display into multiple blocks and using a low-resolution buffer where each storage location corresponds to a block of pixels. This segmentation allows efficient parallel processing of different regions of the image surface while reducing the complexity of tracking individual pixels. The block-based approach enables the GPU to process and manage surface identifiers in manageable units, improving processing efficiency without requiring complex per-pixel management
3Measurement precision
If the GPU updates the low resolution buffer with surface identifiers for each image surface during binning pass, then visibility determination is improved, but the time required for the binning pass increases
Solution Approach 1:
The patent applies dimensionality change by using a low-resolution buffer that operates at a coarser granularity than full-resolution pixel data. Instead of tracking visibility at the pixel level, the system uses blocks of pixels represented by surface identifiers in the low-resolution buffer. This dimensional reduction allows the GPU to determine visibility with high accuracy while significantly reducing the time required for the binning pass, as fewer storage locations need to be updated compared to full-resolution processing
Data Source
AI summary
Aspects of this disclosure are directed to updating a low resolution buffer during a binning pass of an image surface by a graphics processing unit (GPU). For example, during the binning pass of the image surface, the GPU may divide the image surface into a plurality of blocks of surface pixels of the image surface. The GPU may then store a surface identifier of the image surface within storage locations of the low resolution buffer to correspond to each of the blocks of the surface pixels of the image surface.