Hierarchical Ray Tracing Structure With Implicit Nodes

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing ray tracing systems face challenges in reducing data transfer and computational overheads due to large acceleration structures, which hinder real-time rendering performance and increase power consumption, especially in mobile devices.

Innovation Solution

Implementing a hierarchical acceleration structure with implicitly represented nodes, where data defining the regions is inferred from descendants, reducing the amount of data stored and transferred, and dynamically adjusting the intersection testing process based on system conditions to optimize performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If acceleration structures are used to reduce intersection tests, then rendering efficiency is improved, but memory bandwidth and power consumption increase due to large data transfer

Engineering Contradiction:
Improverendering efficiencyVSAvoidpower consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent extracts only the essential data needed for intersection testing by implementing implicit node representation. Instead of storing complete node data including region definitions for all nodes, the system extracts and stores only the minimum necessary information (such as child node references and selective region data), eliminating redundant data storage and transfer while maintaining the acceleration structure's functionality for efficient ray tracing

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent inverts the traditional approach by not explicitly storing node region data but instead inferring it dynamically during ray traversal. Rather than having the acceleration structure store all region boundary information upfront, the system inverts the paradigm by computing region information on-demand from the hierarchical structure itself, reducing memory bandwidth requirements while preserving rendering efficiency

Inventive Principle:
Principle #13The other way round (Inversion)

2Measurement precision

If complete node data is stored explicitly, then intersection testing accuracy is improved, but data transfer and memory usage increase

Engineering Contradiction:
Improveintersection testing accuracyVSAvoiddata transfer
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The acceleration structure serves itself by enabling dynamic inference of node region data during ray traversal. Instead of relying on pre-stored complete node data, the system allows the hierarchical structure to generate the necessary region information on-demand from its own internal relationships (parent-child node connections), eliminating the need for redundant data storage while maintaining precise intersection testing capability

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent changes the representation parameters of node data from explicit complete storage to selective implicit storage. By transforming how node information is encoded (storing only essential identifiers and relationships rather than full region definitions), the system reduces data quantity while maintaining the ability to perform accurate intersection tests through dynamic parameter reconstruction during traversal

Inventive Principle:
Principle #35Parameter changes

3Device complexity

If hierarchical acceleration structures are implemented, then computational overhead is reduced, but silicon area and storage requirements increase

Engineering Contradiction:
Improvecomputational overheadVSAvoidsilicon area
Core Design Contradiction:
Device complexityVSArea of stationary object

Solution Approach 1:

The patent implements a universal data representation where the same compact node structure serves multiple functions: it stores hierarchical relationships, enables region inference, supports ray traversal, and provides intersection testing capability. This multi-functional design eliminates the need for separate data structures for different purposes, reducing overall silicon area while maintaining low computational overhead through the unified efficient structure

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP3929877B1Hierarchical acceleration structures for use in ray tracing systems
Publication Date: 2026.04.01 IMAGINATION TECH LTD
  • EP3929877B1 patent drawingFigure 1
  • EP3929877B1 patent drawingFigure 2a~2b
  • EP3929877B1 patent drawingFigure 3

AI summary

Ray tracing systems and computer-implemented methods are described for generating a hierarchical acceleration structure to be used for intersection testing in a ray tracing system. Nodes of the hierarchical acceleration structure are determined, wherein each of the nodes represents a region in a scene, and wherein the nodes are linked to form the hierarchical acceleration structure. Data is stored representing the hierarchical acceleration structure. The stored data comprises data defining the regions represented by a plurality of the nodes of the hierarchical acceleration structure. At least one node of the hierarchical acceleration structure is an implicitly represented node, wherein data defining a region represented by an implicitly represented node is not explicitly included as part of said stored data but can be inferred from said stored data. Also described are ray tracing systems and computer-implemented methods for performing intersection testing in the ray tracing system in which, based on conditions in the ray tracing system, a determination is made as to whether testing of one or more rays for intersection with a region represented by a particular node of a sub-tree is to be skipped. The one or more rays are tested for intersection in accordance with said determination.