Interconnected ALUs for 3D Graphics Rendering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Portable or handheld devices face challenges in quickly performing realistic 3-D graphics rendering due to limited capabilities and high power consumption, which restricts their ability to efficiently execute complex graphics operations.

Innovation Solution

The implementation of a software-programmable arithmetic logic stage in a graphics pipeline with interconnected arithmetic logic units (ALUs) that include multipliers and adders, allowing for flexible configuration and reuse of hardware components to perform complex operations like multiply-add and multidimensional dot products in a single pass, thereby optimizing graphics operations without increasing hardware requirements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If specialized graphics processing units with multiple hardware functional units are used to perform complex graphics operations, then rendering speed and realism are improved, but device complexity and power consumption increase

Engineering Contradiction:
Improverendering speedVSAvoidhardware complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent implements a universal arithmetic logic unit that can perform multiple graphics operations including vertex transformation, lighting calculations, and texture coordinate generation through software-programmable interconnect circuitry. This single ALU replaces what would traditionally require multiple dedicated hardware functional units, reducing device complexity while maintaining rendering capability across different graphics operations

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Productivity

If multiple dedicated hardware functional units are implemented to perform complex operations in parallel, then rendering throughput is improved, but power consumption increases

Engineering Contradiction:
Improverendering throughputVSAvoidpower consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent employs dynamic interconnect circuitry that can be reconfigured via software to create different data flow patterns for different graphics operations. This dynamic reconfiguration allows the same hardware to adapt to different computational requirements, maintaining high throughput for complex operations without requiring permanently activated multiple functional units, thereby reducing power consumption

Inventive Principle:
Principle #15Dynamics

3Adaptability or versatility

If more hardware components are added to perform complex multidimensional operations, then operational capability is improved, but device size and cost increase

Engineering Contradiction:
Improveoperational capabilityVSAvoidhardware quantity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent introduces software-programmable interconnect circuitry as an intermediary between the arithmetic logic unit and the rest of the graphics pipeline. This interconnect layer enables complex multidimensional operations by dynamically routing data between different units through software control, rather than requiring dedicated hardwired paths for each operation type, thus improving adaptability without proportionally increasing hardware quantity

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS9448766B2Interconnected arithmetic logic units
Publication Date: 2016.09.20 NVIDIA CORP
  • US9448766B2 patent drawing
  • US9448766B2 patent drawing
  • US9448766B2 patent drawing

AI summary

An arithmetic logic stage in a graphics pipeline includes a number of arithmetic logic units (ALUs). The ALUs each include, for example, a multiplier and an adder. The ALUs are interconnected by circuitry that, for example, routes the output from the multiplier in one ALU to both the adder in that ALU and an adder in another ALU.