Image Drawing Apparatus Memory Efficiency via Windowed Address Conversion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional frame buffers with non-rectangular display areas suffer from memory inefficiency due to unused areas, which decrease memory usage in graphic display systems.

Innovation Solution

An image drawing apparatus that divides the frame buffer into multiple windows and blocks within a rectangular storage area, assigning physical addresses only to displayed areas, and using a transaction conversion unit to convert virtual addresses to physical addresses based on pixel positions, thereby eliminating unassigned areas from memory storage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a frame buffer is used to store image data for non-rectangular display areas, then graphic display can be performed on non-rectangular displays, but many unused areas are generated in the frame buffer decreasing memory efficiency

Engineering Contradiction:
Improvedisplay area shape adaptabilityVSAvoidmemory efficiency
Core Design Contradiction:
Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

The frame buffer is divided into multiple windows, where each window corresponds to a specific display area. This segmentation allows the system to allocate memory only for the portions of the frame buffer that correspond to actual display areas, eliminating unused memory regions while maintaining support for non-rectangular display shapes.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different regions of the frame buffer are assigned different properties based on whether they correspond to display areas or unused areas. The transaction conversion unit applies different address mapping rules to different windows, converting virtual addresses to physical addresses only for windows that correspond to actual display areas, thereby optimizing memory usage locally in each region.

Inventive Principle:
Principle #3Local quality

2Quantity of substance

If the frame buffer is divided into multiple windows with assigned areas, then memory efficiency is improved, but the device complexity increases due to address conversion requirements

Engineering Contradiction:
Improvememory efficiencyVSAvoidaddress conversion complexity
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

A transaction conversion unit is introduced as an intermediary component between the graphic processing unit and the frame buffer. This unit handles the complex address conversion tasks by maintaining a mapping between virtual addresses and physical addresses for each window, thereby isolating the complexity from the main processing paths and allowing the GPU and frame buffer to operate with simplified interfaces.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Loss of substance

If physical addresses are assigned only to displayed areas in each window, then wasteful data storage is reduced, but the difficulty of detecting and measuring pixel positions increases

Engineering Contradiction:
Improvedata storage wasteVSAvoidpixel position mapping
Core Design Contradiction:
Loss of substanceVSDifficulty of detecting and measuring

Solution Approach 1:

The system performs preliminary address conversion by the transaction conversion unit before data access occurs. Virtual addresses are pre-mapped to physical addresses based on the window configuration and display area definitions. This preliminary action ensures that when the GPU or display controller accesses frame buffer data, the address translation has already been completed, making pixel position detection straightforward without requiring complex runtime calculations.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11100904B2Image drawing apparatus and display apparatus with increased memory efficiency
Publication Date: 2021.08.24 KK TOSHIBA
  • US11100904B2 patent drawing
  • US11100904B2 patent drawing
  • US11100904B2 patent drawing

AI summary

According to embodiments, an image drawing apparatus includes: an SRAM; and a transaction conversion unit configured to convert a transaction based on a virtual address indicating a pixel position in a storage area of the SRAM into a transaction based on a physical address in the SRAM. When the storage area is divided into a plurality of windows in a row direction and a column direction so that each window includes one or more lines, and an assigned area which is assigned the physical address in the SRAM is set in each of the windows, the transaction conversion unit converts the transaction based on the virtual address into the transaction based on the physical address based on whether the pixel position indicated by the virtual address is in the assigned area.