Adaptive Sub-patches System for Graphics Processing Efficiency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current graphics processing systems face limitations in parallel processing efficiency due to reliance on host processor synchronization, which restricts the amount of parallel processing possible, especially in SIMT architectures.

Innovation Solution

The implementation of adaptive sub-patches during data processing in graphics processors, which allows for dynamic tessellation factors and adaptive tessellation granularity, enabling more efficient processing by subdividing patches into sub-patches with varying tessellation factors based on performance and scene perspective.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If host processor synchronization is used to coordinate parallel threads in SIMT architecture, then system reliability is improved, but parallel processing productivity is limited

Engineering Contradiction:
Improvesystem reliabilityVSAvoidparallel processing productivity
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent introduces a synchronization mechanism that operates independently of the host processor, using a dedicated synchronization unit within the graphics processing system. This intermediary layer coordinates parallel threads without requiring host processor involvement, thereby maintaining system reliability while enabling higher parallel processing productivity through autonomous thread synchronization.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Device complexity

If fixed function computational units are used for graphics processing, then device complexity is reduced, but processing adaptability is limited

Engineering Contradiction:
Improvedevice complexityVSAvoidprocessing adaptability
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent implements a hybrid architecture where fixed function computational units are dynamically configured to perform different operations. The system includes programmable units that can be selectively activated to provide adaptability, while the fixed function units handle routine operations. This dynamic configuration allows the system to balance between simplicity and versatility based on processing requirements.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The graphics processing system is segmented into distinct fixed function computational units, each specialized for specific operations such as vertex processing, fragment processing, or geometry processing. This segmentation allows the system to maintain low complexity within each unit while achieving high overall adaptability through the coordinated operation of multiple specialized units.

Inventive Principle:
Principle #1Segmentation

3Device complexity

If uniform tessellation factors are applied to entire patches, then device complexity is reduced, but manufacturing precision of rendering quality deteriorates

Engineering Contradiction:
Improvetessellation control complexityVSAvoidrendering quality precision
Core Design Contradiction:
Device complexityVSManufacturing precision

Solution Approach 1:

The patent implements adaptive tessellation that applies different tessellation factors to different regions within a patch based on local geometric characteristics and importance. The system evaluates metrics such as surface curvature, distance from camera, and geometric complexity to determine local tessellation requirements, thereby achieving high rendering quality precision without requiring complex global tessellation control.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The tessellation system dynamically adjusts tessellation factors during rendering based on real-time evaluation of scene geometry and view conditions. This dynamic adaptation allows the system to maintain low control complexity by using standardized algorithms while achieving high rendering precision through context-aware parameter adjustment.

Inventive Principle:
Principle #15Dynamics

4Manufacturing precision

If excessive tessellation is applied to ensure rendering quality, then manufacturing precision is improved, but computational resource efficiency deteriorates

Engineering Contradiction:
Improverendering qualityVSAvoidcomputational resource efficiency
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The patent implements cull tessellation that applies tessellation selectively to only those regions of a patch that require high rendering quality. The system identifies and processes only the necessary portions of geometry based on visibility, importance, and geometric complexity metrics, thereby achieving high rendering quality while minimizing computational resource consumption by avoiding unnecessary tessellation of irrelevant regions.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3926587A1Adaptive sub-patches system, apparatus and method
Publication Date: 2021.12.22 INTEL CORP
  • EP3926587A1 patent drawingFigure 1
  • EP3926587A1 patent drawingFigure 2A
  • EP3926587A1 patent drawingFigure 2B

AI summary

Systems, apparatuses and methods may provide a way to subdivide a patch generated in graphics processing pipeline into sub-patches, and generate sub-patch tessellations for the sub-patches. More particularly, systems, apparatuses and methods may provide a way to diverge tessellation sizes to a configurable size within an interior region of a patch or sub-patches based on a position of each of the tessellations. The systems, apparatuses and methods may determine a number of tessellation factors to use based on one or more of a level of granularity of one or more domains of a scene to be digitally rendered, available computing capacity, or power consumption to compute the number of tessellation factors.