Hierarchical Depth Buffer Back Annotation for GPU Power Management

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current graphics processing units (GPUs) face challenges in managing power efficiently due to increased heat generation and power consumption as they integrate more components, which can limit their usage and longevity, especially in battery-powered devices, and existing parallel graphics processing methods do not effectively address the need for efficient power management in hierarchical depth buffer back annotation.

Innovation Solution

Implementing a hierarchical depth buffer back annotation mechanism that allows for efficient communication between GPUs and host processors, utilizing a parallel processing architecture with SIMT architectures and dedicated circuitry to optimize graphics operations, and integrating GPUs with host processors to manage power consumption through advanced interconnects like PCIe or NVLink.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If additional components are integrated onto a single IC chip to improve functionality, then device capability is enhanced, but power consumption and heat generation increase

Engineering Contradiction:
Improvedevice capabilityVSAvoidpower consumption
Core Design Contradiction:
Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The patent divides the depth buffer processing into hierarchical levels (coarse depth buffer and fine depth buffer), allowing different processing stages to operate independently and efficiently, reducing overall power consumption while maintaining enhanced functionality

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements periodic depth buffer updates and rendering operations, enabling the GPU to enter low-power states between active processing periods, thus reducing average power consumption while maintaining device capability

Inventive Principle:
Principle #19Periodic action

2Adaptability or versatility

If additional components are integrated onto a single IC chip to improve functionality, then device capability is enhanced, but heat generation increases

Engineering Contradiction:
Improvedevice capabilityVSAvoidheat generation
Core Design Contradiction:
Adaptability or versatilityVSTemperature

Solution Approach 1:

By segmenting depth buffer processing into hierarchical levels, the patent distributes computational load across different processing units and time periods, preventing concentrated heat generation and improving thermal management

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Periodic rendering and depth buffer updates create intervals between high-power operations, allowing heat dissipation and reducing peak temperature while maintaining enhanced device functionality

Inventive Principle:
Principle #19Periodic action

3Productivity

If parallel processing techniques are implemented to increase processing efficiency, then productivity is improved, but device complexity increases

Engineering Contradiction:
Improveprocessing efficiencyVSAvoiddevice complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The hierarchical depth buffer approach segments rendering work into coarse and fine levels, enabling parallel processing at each level while keeping individual processing units relatively simple, thus improving productivity without excessive complexity increase

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent adds a hierarchical dimension to depth buffer processing, organizing parallel operations in multiple levels rather than a single flat structure, which improves processing efficiency while managing complexity through structured organization

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS11615584B2Hierarchical depth buffer back annotation
Publication Date: 2023.03.28 INTEL CORP
  • US11615584B2 patent drawing
  • US11615584B2 patent drawing
  • US11615584B2 patent drawing

AI summary

Briefly, in accordance with one or more embodiments, a processor performs a coarse depth test on pixel data, and performs a final depth test on the pixel data. Coarse depth data is stored in a coarse depth cache, and per pixel depth data is stored in a per pixel depth cache. If a result of the coarse depth test is ambiguous, the processor is to read the per pixel depth data from the per pixel depth cache, and to update the coarse depth data with the per pixel depth data if the per pixel depth data has a smaller depth range than the coarse depth data.