IP Library Granted Patent US 10,846,908
Granted Patent B2
US 10,846,908 · App. 16/427,658 · Granted Nov 24, 2020

Graphics processing apparatus based on hybrid GPU architecture

Inventor: Hyung Min Yoon (Seoul, KR)
Assignee: SILICONARTS, INC.
G06T15/005G06T1/60G06T15/04G06T15/06G06T15/80G06T2210/21
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,846,908
App. No.
16/427,658
Granted
Nov 24, 2020
Kind
B2
Abstract

Disclosed is a graphics processing apparatus based on hybrid GPU architecture. The graphics processing apparatus includes: an acceleration structure generation unit generating an acceleration structure based on geometry data related to a 3D scene; a ray tracing unit performing ray tracing based on the acceleration structure; and a rasterization unit including a sharing interface for shader-sharing with the ray tracing unit and performing rasterization rendering by sharing shading information generated from a result of the ray tracing.

Claims (23)

1. A graphics processing apparatus based on hybrid GPU architecture, the apparatus comprising:

an acceleration structure generation unit generating an acceleration structure based on geometry data related to a 3D scene;

a ray tracing unit performing ray tracing based on the acceleration structure; and

a rasterization unit including a sharing interface for shader-sharing with the ray tracing unit and performing rasterization rendering by sharing shading information generated from a result of the ray tracing,

wherein the ray tracing unit performs the ray tracing by using a shader shared directly from the rasterization unit through the sharing interface,

wherein the rasterization unit performs the rasterization rendering by using shading information shared directly from the ray tracing unit through the sharing interface, and

wherein the acceleration structure generation unit, the ray tracing unit, and the rasterization unit are each implemented via at least one processor.

2. The apparatus of claim 1 , wherein the acceleration structure generation unit includes:

a primitive cache providing a primitive scene corresponding to the 3D scene;

a working memory storing primitives during a process for generating an acceleration structure; and

an acceleration structure buffer storing the acceleration structure; and

generates a dynamic acceleration structure for each frame in conjunction with the working memory and stores the generated dynamic acceleration structure in a memory through the acceleration structure buffer.

3. The apparatus of claim 1 , wherein the ray tracing unit performs the ray tracing by performing ray generation, traversal testing, intersection testing, shading, and texture mapping; and

when the intersection testing is completed, the ray tracing unit performs the shading after performing synchronization with the rasterization unit by determining whether shader-sharing is possible through the sharing interface.

4. The apparatus of claim 3 , wherein the ray tracing unit performs the ray tracing by using a shader shared directly from the rasterization unit through the sharing interface at the shading or the texture mapping.

5. The apparatus of claim 1 , wherein the rasterization unit performs the rasterization rendering by sequentially performing vertex shading, triangle setup, fragment shading, and fragment operation.

6. The apparatus of claim 5 , wherein the rasterization unit performs the rasterization rendering by using the shading information at the fragment shading or the fragment operation.

7. The apparatus of claim 1 , further comprising a hybrid graphics processing unit generating a hybrid image related to the 3D scene through selective graphics processing between the ray tracing unit and the rasterization unit,

wherein the hybrid graphics processing unit is implemented via at least one processor.

8. The apparatus of claim 7 , wherein the hybrid graphics processing unit generates the hybrid image by sequentially performing:

determining a 2D frame related to the 3D scene;

partitioning the 2D frame into a first area including dynamic objects, a second area including static objects, and a third area related to a background of the 3D scene; and

generating an image about the 2D frame by selectively applying ray tracing rendering or rasterization rendering to the first to the third areas.

Assignments (3)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 11, 2025
From: SILICONARTS TECHNOLOGY INC.
To: SILICONARTS TECHNOLOGY US INC.
Reel/Frame 070476/0964 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 2, 2024
From: SILICONARTS INC.
To: SILICONARTS TECHNOLOGY INC.
Reel/Frame 066179/0677 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 31, 2019
From: YOON, HYUNG MIN
To: SILICONARTS, INC.
Reel/Frame 049337/0688 →
Priority Claims (1)
KR 10-2019-0042397 · Apr 11, 2019 · national
Continuity (1)
Related Publication 20200327712A1 · Oct 15, 2020