Serial Ray Intersection Testing for Memory-Limited Ray Tracing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Ray tracing systems face challenges in performing intersection testing at a rate suitable for real-time rendering on devices with tight constraints on silicon area, cost, and power consumption, such as mobile devices, due to the high processing requirements and memory needs, even with current acceleration structures.

Innovation Solution

A hierarchical acceleration structure with a serial-mode ray intersection process that generates a limited number of new ray requests and stores intersection information for further nodes, reducing the number of intersection tests and memory overflow risks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a hierarchical acceleration structure is used to reduce intersection testing work, then processing efficiency is improved, but memory requirements increase leading to potential memory overflow

Engineering Contradiction:
Improveintersection testing efficiencyVSAvoidmemory usage
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

The patent segments the hierarchical acceleration structure into multiple levels (e.g., BVH levels) and processes rays through each level sequentially. Instead of loading all geometry data into memory simultaneously, the system divides the scene into coarse-to-fine bounding volume hierarchies, processing only the necessary portions at each level. This segmentation reduces peak memory usage while maintaining intersection testing efficiency.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary organization of geometry data into hierarchical acceleration structures before ray tracing begins. Bounding volumes are pre-computed and stored in a hierarchical format, allowing rays to be tested against simplified representations first. This preliminary structuring enables efficient culling of non-intersecting geometry without requiring all detailed geometry to be loaded into memory simultaneously.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If every ray is tested against every primitive to ensure accurate intersection detection, then measurement precision is improved, but processing time increases significantly

Engineering Contradiction:
Improveintersection detection accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the set of primitives into hierarchical groups using bounding volumes. Rays are first tested against coarse bounding volumes at upper hierarchy levels, and only if they intersect do they proceed to finer levels. This segmentation creates a culling effect where most primitives are eliminated early without detailed intersection testing, ensuring accuracy for relevant primitives while dramatically reducing total processing time.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies partial action by testing rays against only the necessary subset of primitives at each hierarchy level rather than all primitives. The hierarchical structure allows the system to perform intersection testing on a partial set of primitives (those within intersecting bounding volumes) at each level, accumulating results as it descends the hierarchy. This approach achieves complete and accurate intersection detection while avoiding excessive testing of obviously non-intersecting primitives.

Inventive Principle:
Principle #16Partial or excessive action

3Measurement precision

If more ray requests are generated to capture all intersections, then measurement precision is improved, but device complexity and processing burden increase

Engineering Contradiction:
Improveintersection completenessVSAvoidray request management complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments ray request generation into discrete hierarchy levels. At each level, the system generates ray requests only for bounding volumes that are intersected by incoming rays. This segmented approach ensures that ray requests are generated systematically and completely for all relevant geometry while avoiding redundant requests. The hierarchical segmentation provides a natural framework for tracking and managing ray requests without excessive complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements dynamic ray request management where the number and scope of ray requests adapt based on the hierarchical traversal state. As rays descend through the hierarchy and intersect bounding volumes, new ray requests are dynamically generated for child nodes. This dynamic generation ensures complete intersection detection while automatically adjusting the workload based on scene geometry and ray paths, avoiding static over-provisioning of ray requests.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS12620162B2Intersection testing for ray tracing
Publication Date: 2026.05.05 IMAGINATION TECH LTD
  • US12620162B2 patent drawing
  • US12620162B2 patent drawing
  • US12620162B2 patent drawing

AI summary

A system and method for performing intersection testing of rays in a ray tracing system. The ray tracing system uses a hierarchical acceleration structure comprising a plurality of nodes, each identifying one or more elements able to be intersected by a ray. The system makes use of a serial-mode ray intersection process, in which, when a ray intersects a bounding volume, a limited number of new ray requests are generated.