Compressed Triangle Blocks With Implicit Indexing for Ray Tracing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing ray tracing technologies face challenges in efficiently handling highly complex and irregular geometries, leading to increased memory footprint and processing time, particularly in constructing acceleration structures for ray tracing, which limits performance and adoption.

Innovation Solution

A new highly compressed triangle block format (HCTB) that includes implicit triangle identifiers, arithmetic vertex compression, and strip topology encoding, allowing for increased compression and efficient rendering of complex geometries, with features like implicit indexing and reduced memory footprint.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the number of triangles is increased to represent complex geometries, then the geometric complexity and realism are improved, but the memory footprint and processing time increase

Engineering Contradiction:

Improvegeometric complexityVSAvoidmemory footprint

Core Design Contradiction:

Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

The patent divides the triangle mesh into clusters of up to 128 triangles, where each cluster is processed and stored independently. This segmentation allows the system to handle large complex geometries by breaking them into manageable units, reducing the memory footprint required to store the entire mesh while maintaining the ability to represent complex shapes through multiple clusters.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent combines multiple triangles into single-block compressed triangle structures, merging adjacent triangles that share vertices into unified data representations. This merging reduces the total number of separate data entries needed, thereby reducing memory footprint while preserving the geometric complexity through shared vertex references within the compressed block.

Inventive Principle:

Principle #5Merging (Combining)

2Adaptability or versatility

If the number of triangles is increased to represent complex geometries, then the geometric complexity and realism are improved, but the processing time for constructing acceleration structures increases

Engineering Contradiction:

Improvegeometric complexityVSAvoidprocessing time

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

By segmenting the triangle mesh into clusters of up to 128 triangles, the patent enables parallel processing of acceleration structure construction. Each cluster can be processed independently, allowing concurrent operations that significantly reduce the total processing time while maintaining the ability to handle complex geometries through hierarchical organization.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary compression and organization of triangle data into standardized blocks before acceleration structure construction. This preliminary action pre-processes the data into an optimized format, reducing the computational overhead during subsequent ray tracing and rendering operations.

Inventive Principle:

Principle #10Preliminary action

3Reliability

If explicit vertex indices are stored for all triangles, then the vertex reuse is improved, but the storage requirements increase

Engineering Contradiction:

Improvevertex reuseVSAvoidstorage requirements

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The patent extracts and stores only the unique vertex indices within each triangle cluster, eliminating redundant vertex index storage across multiple triangles. By taking out only the necessary unique vertex references and using them to implicitly define all triangles in the cluster, the system maintains vertex reuse reliability while significantly reducing storage requirements.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent uses implicit vertex copying through shared references within compressed triangle blocks. Instead of storing explicit indices for every triangle vertex, the system copies vertex data through shared memory references within the block, maintaining vertex reuse while reducing storage overhead through efficient memory access patterns.

Inventive Principle:

Principle #26Copying

4Adaptability or versatility

If legacy data structures are used for ray tracing, then the hardware compatibility is improved, but the performance with complex geometries deteriorates

Engineering Contradiction:

Improvehardware compatibilityVSAvoidrendering performance

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

The patent changes the data representation parameters from traditional per-triangle formats to compressed block formats with implicit topology. This parameter change maintains compatibility with legacy ray tracing hardware while improving performance by reducing data throughput requirements and enabling more efficient processing of complex geometries through the compressed representation.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20260080605A1Highly Compressed Triangles

Publication Date: 2026.03.19 NVIDIA CORP

AI summary

Raytracing and pathtracing systems use vertex compression including per-component shifts, unsigned arithmetic deltas, and base vertex shortening. Topology encodings include enhanced explicit vertex indexing and implicit triangle strip based indexing including left, right, start tokens, turn back, and degenerate tokens and including rotation. Substantial lossless compression ratio increases are realized, especially for highly quantized vertex data.