Bounding Active Pixel Regions to Reduce GPU Power
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current techniques for pixel processing in graphics processing units (GPUs) do not effectively track and report inactive regions of rendered content, leading to inefficient processing and increased power consumption.
Innovation Solution
A method that involves obtaining an indication of active pixel regions for a frame of graphics content, computing bounding areas for active pixels, configuring a workload for composition or reprojection on these areas, and outputting a configured workload indication to skip processing on inactive pixels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If all pixels are processed in the graphics processing pipeline, then complete visual content is rendered, but power consumption and computational resources are wasted on inactive regions
Solution Approach 1:
The patent divides the rendered content into active regions and inactive regions based on bounding box information. By segmenting the pixel processing into these two categories, the system can selectively process only active pixels while skipping inactive ones, thus improving processing efficiency and reducing power consumption without compromising the rendering of important content
Solution Approach 2:
The patent applies different processing quality to different regions of the display. Active regions within bounding boxes receive full processing and rendering, while inactive regions are skipped. This local differentiation ensures that computational resources are allocated based on content importance, resolving the contradiction between complete rendering and energy efficiency
2Reliability
If pixel processing is performed on inactive regions, then complete frame rendering is achieved, but computational resources are wasted
Solution Approach 1:
The patent performs preliminary identification of active and inactive regions using bounding box information before the main pixel processing pipeline. This preliminary action allows the system to pre-determine which pixels need full processing and which can be skipped, ensuring rendering completeness for active regions while saving computational resources on inactive regions
Solution Approach 2:
The patent introduces an intermediary mechanism that uses bounding box data to mediate between the rendering pipeline and the display output. This intermediary layer filters and directs pixel processing based on active/inactive region classification, maintaining rendering reliability for important content while reducing overall computational resource consumption
Data Source
AI summary
This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for reducing AR power consumption via bounding sparse content. A processor obtains, for a frame of graphics content, a first indication of at least one of: the frame, a graphics processor tracked active pixel region associated with the frame, tile tracking compression metadata of the frame, or render associated tiles of the frame. The processor computes, based on the first indication, a set of bounding areas for a set of active pixels of the frame. The processor configures a workload for at least one of a composition or a reprojection on the computed set of bounding areas. The processor outputs a second indication of the configured workload for at least one of the composition or the reprojection on the computed set of bounding areas for the set of active pixels of the frame.