GPU-Attached Non-Volatile Memory for Multi-View E-Sports Rendering

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Cloud-based gaming systems struggle to efficiently utilize client devices with powerful graphics processing capabilities, and the rendering costs for multiple user views in shared virtual environments, such as e-sports events, are not optimally distributed.

Innovation Solution

Implementing a cloud gaming system that leverages cloud, edge, or client-based compute resources, allowing game servers to be downloaded and executed locally on clients with sufficient graphics processing capabilities, and utilizing GPU-attached non-volatile memory for improved rendering efficiency in e-sports spectator modes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If cloud-based gaming systems render each client view separately, then each user can access the game on their device, but the rendering costs are not optimally distributed and powerful client devices are not fully utilized

Engineering Contradiction:

Improveaccessibility across devicesVSAvoidrendering efficiency

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

The system segments rendering tasks by identifying which views can be rendered at the cloud server and which can be handled by client devices. Spectator views are rendered at the server, while player views can be offloaded to client devices with sufficient graphics capabilities, optimizing the distribution of rendering workload.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system creates a master rendering of the game scene at the cloud server, then generates derivative views from this master rendering for multiple spectators. This copying approach allows efficient distribution of visual information without requiring separate full renderings for each spectator view.

Inventive Principle:

Principle #26Copying

2Ease of operation

If each user views through a separate client device, then user perspectives are maintained, but the rendering costs for multiple views are not amortized efficiently

Engineering Contradiction:

Improveuser view perspectiveVSAvoidrendering cost

Core Design Contradiction:

Ease of operationVSUse of energy by stationary object

Solution Approach 1:

The system merges multiple spectator view renderings into a single master rendering at the cloud server. This combined approach maintains individual user perspectives while significantly reducing the total rendering computational cost by amortizing the rendering burden across all spectators through a single source rendering.

Inventive Principle:

Principle #5Merging (Combining)

3Quantity of substance

If cloud-based gaming systems support more spectators, then the platform capacity increases, but the rendering capacity required increases proportionally

Engineering Contradiction:

Improvenumber of spectatorsVSAvoidrendering capacity

Core Design Contradiction:

Quantity of substanceVSPower

Solution Approach 1:

The system applies partial rendering at the cloud server for spectator views, rendering only the necessary visual information rather than complete game states. This partial action approach allows supporting a large number of spectators without requiring proportional increases in rendering capacity, as the server only performs essential rendering tasks.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS12555298B2Systems and methods for improved efficient E-sports spectator mode through use of GPU attached non-volatile memory

Publication Date: 2026.02.17 INTEL CORP

AI summary

Described herein is a cloud-based gaming system in which multiple views of a spectated E-sports event can be rendered and combined into an immersive video having at least three degrees of freedom. Low-latency generation of the immersive video is facilitated via the use of GPU-controlled non-volatile memory on which rendered data for multiple viewpoints are stored.