Bounding Volume Hierarchy Leaf Node Compression via Integer Coordinate Differences

Solution Overview

Problem

Current node compression techniques in graphics processing are inefficient in compressing leaf nodes, leading to increased memory bandwidth consumption, higher power consumption, and complex determination of the number of primitives per leaf node.

Innovation Solution

A method involving the configuration of a bounding volume hierarchy (BVH) structure with compressed primitive IDs and geometry IDs, conversion of floating-point coordinates to integer values, calculation and modification of differences between these integer values, and storage of these compressed and modified values for efficient leaf node compression.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If current node compression techniques are used, then the compression process is simple, but the compression efficiency for leaf nodes is poor leading to increased memory bandwidth consumption

Engineering Contradiction:

Improvecompression efficiencyVSAvoidmemory bandwidth consumption

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The patent changes the representation parameters of floating-point coordinates by converting them to integer values through bit-shifting operations. This parameter transformation enables more efficient compression of leaf nodes in the BVH structure, reducing memory bandwidth consumption while maintaining compression effectiveness

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent applies different compression strategies to different parts of the BVH structure. Specifically, it uses specialized compression for leaf nodes (containing primitive geometry data) versus internal nodes, optimizing the compression process for each node type's specific characteristics and improving overall compression efficiency

Inventive Principle:

Principle #3Local quality

2Productivity

If current node compression techniques are used, then the compression process is simple, but the compression efficiency for leaf nodes is poor leading to higher power consumption

Engineering Contradiction:

Improvecompression efficiencyVSAvoidpower consumption

Core Design Contradiction:

ProductivityVSUse of energy by stationary object

Solution Approach 1:

The patent changes the representation parameters of floating-point coordinates by converting them to integer values through bit-shifting operations. This parameter transformation enables more efficient compression of leaf nodes in the BVH structure, reducing memory bandwidth consumption while maintaining compression effectiveness

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent applies different compression strategies to different parts of the BVH structure. Specifically, it uses specialized compression for leaf nodes (containing primitive geometry data) versus internal nodes, optimizing the compression process for each node type's specific characteristics and improving overall compression efficiency

Inventive Principle:

Principle #3Local quality

3Productivity

If current node compression techniques are used, then the compression process is simple, but the determination of number of primitives per leaf node becomes complex

Engineering Contradiction:

Improvecompression efficiencyVSAvoiddetermination complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent performs preliminary actions during the BVH construction phase by pre-determining the number of primitives per leaf node and pre-computing compression parameters. This advance preparation simplifies the later compression process and avoids complex runtime calculations, improving both efficiency and reducing operational complexity

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12205223B2Bounding volume hierarchy leaf node compression

Publication Date: 2025.01.21 QUALCOMM INC

AI summary

Aspects presented herein relate to methods and devices for graphics processing including an apparatus, e.g., a GPU. The apparatus may configure a BVH structure including a plurality of nodes each including one or more primitives, and each of the primitives being associated with a primitive ID, a geometry ID, and a set of floating-point coordinates. The apparatus may also compress the primitive ID or the geometry ID for each of the primitives. Further, the apparatus may convert a binary representation of each of the floating-point coordinates into an integer value of each of the floating-point coordinates. The apparatus may also calculate a difference between the integer value of each of the set of floating-point coordinates for each of the plurality of primitives. The apparatus may also store the compressed primitive ID or the compressed geometry ID and the calculated difference.