Reverse Rendering Mechanism for 3D Image Overdraw Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing graphics rendering technologies face inefficiencies in computation and power consumption due to overdraw and unnecessary work in rendering synthetic 3D images, particularly in handling occluded pixels and intermediate images, which contribute minimally to the final image.
Innovation Solution
The implementation of Stream Sorting and Reverse Rendering techniques, which involve sorting streams of triangles into tile display lists, performing delayed visibility sorting, and rendering images in reverse order to eliminate unnecessary computations and memory transactions by focusing only on pixels and regions that contribute to the final image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If traditional forward rendering is used to render synthetic 3D images, then the rendering process is simple to implement, but significant computation and power are wasted on overdraw and intermediate images that do not contribute to the final image
Solution Approach 1:
The patent applies reverse rendering by inverting the traditional rendering sequence. Instead of rendering from first to last geometry (forward rendering), the system renders from last to first geometry, allowing early termination when occluded pixels are detected. This inversion enables the system to skip rendering of pixels that will be occluded by subsequent geometry, eliminating overdraw and reducing power consumption while maintaining implementation feasibility through modified rendering pipeline logic
2Loss of energy
If the entire intermediate image is rendered to ensure all possible contributions, then complete illumination effects are captured, but significant computation and power are spent on pixels that will not be sampled in subsequent passes
Solution Approach 1:
The patent implements partial rendering of intermediate images by rendering only the portions that will actually be sampled in subsequent rendering passes. The system identifies and renders only the relevant regions of intermediate images (such as shadow maps and reflection maps) that contribute to the final image, rather than rendering the entire intermediate image. This partial action approach reduces computation and power waste while maintaining the reliability of illumination effects by ensuring all sampled pixels are accurately rendered
3Loss of energy
If depth sorting structures are increased in size to reduce overdraw, then fewer pixels are overwritten, but the chip area and memory bandwidth requirements increase significantly
Solution Approach 1:
The patent eliminates the need for large depth sorting structures by applying reverse rendering. Instead of pre-sorting all geometry to minimize overdraw (which requires large chip area structures), the system renders in reverse order and terminates early when occlusion is detected. This approach achieves overdraw reduction without requiring increased chip area, as the rendering pipeline itself handles the optimization through backward traversal and early termination logic
Data Source
AI summary
Embodiments disclosed include a mechanism in a system and method for significantly reducing power consumption by reducing computation and bandwidth. This mechanism is particularly applicable for modern 3D synthetic images which contain high pixel overdraw and dynamically generated intermediates images. Only blocks of computation which contribute to the final image are performed. This is accomplished by rendering in reverse order and by performing multiple visibility sort in a streaming fashion through the pipeline. Rendering of dynamically generated intermediate images is performed sparsely by projecting texture coordinates from a current image back into one or more dependent images in a recursive manner. The newly computed pixel values are then filtered and control is returned to the sampling shader of the current image. When only visible pixels are projected optimal computation is performed. Several implementations are presented with increasing efficiency. An acceleration structure, termed a Draw Buffer, simplifies the process of projecting backward and utilizes a hardware managed dynamic memory object. This mechanism reduces computation by 50%, with significant bandwidth and power savings.