BVH Geometry Compression With Quantized and Residual Triangle Blocks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing graphics processing technologies face inefficiencies in ray tracing due to the large size of bounding volume hierarchies (BVHs) caused by storing both compressed and uncompressed triangle data, which increases memory footprint and traversal time, and the need for repeated compression/decompression, while traditional triangle filters increase power consumption and performance penalties.

Innovation Solution

A progressive geometry compression scheme is employed, where quantized triangle data is stored in quantized blocks, with separate blocks storing residual precision for higher-precision operations, reducing BVH footprint and traversal-time bandwidth without performance penalties, and deduplication techniques are used to efficiently assign primitives to quantized triangle blocks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If both compressed and uncompressed triangle data are stored in the BVH, then higher precision operations can be performed, but the memory footprint and traversal time increase substantially

Engineering Contradiction:
ImproveprecisionVSAvoidmemory footprint
Core Design Contradiction:
Measurement precisionVSVolume of stationary object

Solution Approach 1:

The triangle data is segmented into two separate blocks: a quantized triangle block storing compressed data with lower precision, and a residual block storing the difference data. This segmentation allows the system to use only the necessary precision for each operation, reducing the memory footprint while maintaining the capability for high-precision operations when needed.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The residual data block is nested within the BVH structure alongside the quantized triangle block. The residual block contains the difference between the original and quantized data, allowing reconstruction of full-precision data when required. This nested structure enables progressive precision without duplicating the entire triangle data set.

Inventive Principle:
Principle #7Nested doll (Nesting)

2Measurement precision

If both compressed and uncompressed triangle data are stored in the BVH, then higher precision operations can be performed, but traversal time increases

Engineering Contradiction:
ImproveprecisionVSAvoidtraversal time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

By segmenting triangle data into quantized and residual blocks, the system can perform traversal operations using only the compact quantized data for bounding volume tests, avoiding the need to traverse or process the larger residual data during normal traversal, thus reducing traversal time.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs partial precision operations during traversal by using only the quantized data for most operations. Full precision is applied only when necessary for final intersection testing, reducing the overall computational time while maintaining accuracy where needed.

Inventive Principle:
Principle #16Partial or excessive action

3Productivity

If traditional triangle filters are used for ray tracing, then intersection testing can be performed, but power consumption increases and performance penalties occur

Engineering Contradiction:
Improveray tracing performanceVSAvoidpower consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The invention changes the precision parameter of triangle data storage by using quantized data instead of full-precision data in the quantized triangle block. This parameter change reduces memory bandwidth requirements and power consumption during traversal while maintaining the ability to achieve accurate intersection results when needed.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS12573128B2Progressive compression of geometry for graphics processing
Publication Date: 2026.03.10 APPLE INC
  • US12573128B2 patent drawing
  • US12573128B2 patent drawing
  • US12573128B2 patent drawing

AI summary

A graphics processor includes circuitry configured to generate a bounding volume hierarchy (BVH) data structure for a graphics scene to be rendered. The BVH includes a leaf node that includes: node data that specifies quantization frame data for quantized primitive children of the leaf node, a quantized primitive block that includes quantized vertex components of the primitive children, and a vertex residual block that includes additional vertex data for the primitive children. The vertex residual block may not include redundant data with the quantized primitive block and the quantized primitive block and vertex residual block may include, for the primitive children, sufficient data to reconstruct an original-precision representation of their vertices. The circuitry may traverse the BVH to perform ray tracing operations during render of the graphics scene, including to perform one or more ray primitive intersect tests based on the quantized primitive block and the vertex residual block.