VRAM Overlay Blending for Cloud Gaming Resource Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional overlay/underlay processes for multimedia content in cloud gaming platforms require significant system resources, leading to low performance and high power consumption, especially in high CCU systems, due to the need for capturing, copying, and blending images across different video sources.

Innovation Solution

The proposed solution involves rendering the primary video source in VRAM and overlaying/underlaying secondary multimedia content directly within the same VRAM location, eliminating the need for an extra buffer and reducing system bus, system memory, and CPU usage by performing blending operations within VRAM.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If conventional overlay/underlay process is used (capturing, copying, blending in frame buffer), then multimedia content can be overlaid, but system bus usage, system memory usage, and CPU usage increase significantly

Engineering Contradiction:
Improveoverlay capabilityVSAvoidsystem resource usage
Core Design Contradiction:
Ease of manufactureVSDevice complexity

Solution Approach 1:

The patent merges the overlay/underlay blending operation directly into the VRAM rendering process. Instead of capturing the primary video source to system memory, copying it to a frame buffer, and blending the overlay there, the system blends the overlay content directly into the VRAM frame buffer at the same location where the primary source is rendered. This consolidation eliminates multiple memory copy operations and reduces CPU intervention, thereby reducing system bus usage, system memory usage, and CPU usage while maintaining the overlay capability.

Inventive Principle:
Principle #5Merging (Combining)

2Ease of manufacture

If conventional overlay/underlay process is used, then multimedia content can be blended, but power consumption increases in high CCU systems

Engineering Contradiction:
Improveblending capabilityVSAvoidserver power consumption
Core Design Contradiction:
Ease of manufactureVSUse of energy by stationary object

Solution Approach 1:

The patent combines the blending operation with the rendering process in VRAM, eliminating the need for separate capture and copy operations in system memory. By performing the blend directly in the VRAM frame buffer where the primary video source is already being rendered, the system reduces the total number of memory access operations and CPU cycles required, thereby reducing power consumption in high CCU server environments.

Inventive Principle:
Principle #5Merging (Combining)

3Quantity of substance

If overlay is performed in VRAM at the same location, then no extra buffer is needed, but the process must be more efficient

Engineering Contradiction:
Improvememory buffer usageVSAvoidprocessing efficiency
Core Design Contradiction:
Quantity of substanceVSProductivity

Solution Approach 1:

The patent merges the overlay rendering with the primary video rendering in the same VRAM location. The GPU performs both the primary source rendering and the overlay blending in a unified rendering pass, using the same frame buffer for both operations. This approach eliminates the need for additional buffer memory while maintaining high processing efficiency by leveraging the GPU's parallel processing capabilities and avoiding costly memory copy operations.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS10521879B2Overlaying multi-source media in VRAM
Publication Date: 2019.12.31 KK UBITUS
  • US10521879B2 patent drawing
  • US10521879B2 patent drawing
  • US10521879B2 patent drawing

AI summary

Methods, apparatuses, and computer program products for overlaying multisource media in VRAM are described. The primary media source is rendered in VRAM by an application program, and then the secondary media source(s) are rendered and blended to the primary source in VRAM at the same location of the primary source in VRAM, so no extra buffer is needed. This improves system performance and reduces power consumption, through reduced system bus, system memory, and CPU usage.