Unified rasterization and ray tracing rendering environments

• US 9,424,685 B2
• Filed: 07/30/2013
• Issued: 08/23/2016
• Est. Priority Date: 07/31/2012
• Status: Active Grant

• Current Assignee: Imagination Technologies Limited (Imagination Technologies Group Ltd.)
• Original Assignee: Imagination Technologies Limited (Imagination Technologies Group Ltd.)
• Inventors: Howson, John W., Peterson, Luke Tilman
• Primary Examiner(s): Ma, Tize
• Application Number:
  US13/953,754
• Publication Number:
  US 20140063016A1
• Time in Patent Office: 1,120 Days
• US Class Current: 1/1
• CPC Class Codes: G06T 1/20 Processor architectures; Processor configuration, e.g. pipelining
• CPC Class Codes: G06T 15/005 General purpose rendering architectures
• CPC Class Codes: G06T 15/06 Ray-tracing
• CPC Class Codes: G06T 15/80 Shading
• CPC Class Codes: G06T 2200/04 involving 3D image data
• CPC Class Codes: G06T 2200/28 involving image processing hardware
• CPC Class Codes: G06T 2207/20021 Dividing image into blocks, subimages or windows

Abstract

A graphics processor architecture provides for scan conversion and ray tracing approaches to visible surface determination as concurrent and separate processes. Surfaces can be identified for shading by scan conversion and ray tracing. Data produced by each can be normalized, so that instances of shaders, being executed on a unified shading computation resource, can shade surfaces originating from both ray tracing and rasterization. Such resource also may execute geometry shaders. The shaders can emit rays to be tested for intersection by the ray tracing process. Such shaders can complete, without waiting for those emitted rays to complete. Where scan conversion operates on tiles of 2-D screen pixels, the ray tracing can be tile aware, and controlled to prioritize testing of rays based on scan conversion status. Ray population can be controlled by feedback to any of scan conversion, and shading.

Claims

1. A machine-implemented method of rendering,images in a computer graphics system, comprising:

identifying one or more visible surfaces, from among surfaces in a 3-D scene, the identified on or more visible surfaces comprising visible surfaces for a plurality of pixels located in 2-D screen space;

preparing, concurrently with the identifying, to execute shaders associated with respective visible surfaces of pixels of the one or more visible surfaces that have been identified, the preparing comprising completing a respective normalized set of inputs to be provided to each shader for use during execution, the normalized set of inputs comprising a specified set of attributes regardless of whether the one or more identified surfaces were identified by ray intersection testing or scan conversion;

executing each of the shaders in a computation cluster, wherein each of the executing shaders comprises one or more operations, wherein at least on of the executing shaders comprises defining one or more rays to be tested for intersection with surfaces in the 3-D scene;

testing at least some of the rays for intersection concurrently with the identifying and the executing of the shaders; and

shading identified intersections for rays completing intersection testing within the computation cluster.

2. The machine-implemented method of rendering of claim 1, further comprising producing a 3-D viewer position from a 2-D screen space view position, and projection matrix data.

3. The machine-implemented method of rendering of claim 1, wherein the intersection testing of rays is comprised by the identifying, and the preparing comprises performing an attribute interpolation for a visible surface identified through intersection testing of a ray.

4. The machine-implemented method of rendering of claim 2, wherein the producing of the 3-D viewer position is performed during the completion of the normalized set of inputs.

5. The machine-implemented method of rendering of claim 2, wherein the completion of the normalized set of inputs comprises producing interpolated data for attributes for a particular point on a surface, for which image attribute data is unavailable.

6. The machine-implemented method of rendering of claim 1, wherein the performing visible surface identification for the plurality of pixels is performed sequentially for 2-D regions of pixels in the plurality of pixels, and for collections of rays associated with respective 3-D volumes in 3-D scene space.

7. The machine-implemented method of rendering of claim 1, wherein the one or more visible surfaces comprise surfaces that have been determined visible from the view position, but which may be obscured by another surface closer to the view position.

8. The machine-implemented method of rendering of claim 1, further comprising performing an entirety of the identifying of the one or more visible surfaces for each particular pixel before executing a shader for the identified one or more visible surfaces for that pixel.

9. The machine-implemented method of rendering of claim 1, further comprising completing the execution of an executing shader before an intersection for a ray defined by that shader has been shaded.

10. The machine-implemented method of rendering of claim 1, further comprising queuing rays defined by executing shaders and beginning to test groups of queued rays for intersection within a hierarchical acceleration structure.

11. The machine-implemented method of rendering of claim 1, further comprising, during executing of the shader, accessing an API semantic that returns an interpolated attribute for the visible surface being shaded by that executing shader.

12. The machine-implemented method of rendering of claim 1, further comprising during executing of the shader, accessing an API semantic that returns 3-D locational data for parameters, outputted from the identifying, and specified in the 2-D screen space.

13. The machine-implemented method of rendering of claim 1, further comprising, during executing of the shader, interpolating vertex parameter data in 3-D space to produce parameter data for a point on a visible surface identified by a rasterization unit.

14. The machine-implemented method of rendering of claim 1, further comprising, during executing of the shader, accessing 3-D coordinates for vertexes defining the visible surface, and determining a location on the visible surface to use as an origin for the ray.

15. The machine-implemented method of rendering of claim 1, wherein the completing of the respective normalized set of inputs comprises producing said specified set of attributes, controlled by a template associated with the visible surface, for shader invocations originating from both scan conversion and from ray tracing.

16. An apparatus for graphics rendering, comprising:

a visible surface determination module capable of scan converting 3-D scene geometry from a viewpoint and identifying visible surfaces for a set of pixels in 2-D screen space, and comprising a ray tracing intersection unit for testing rays for intersection with 3-D scene geometry, wherein the visible surface determination module performs scan conversion in a regular pattern in 2-D screen space and performs ray intersection testing by collecting rays according to regions of 3-D space and deferring commencement of testing for individual rays until dispatch within a packet of rays;

a normalizer configured to provide normalized shader data for use by a shader during execution, the normalized shader data comprising a specified set of attributes regardless of whether the visible surfaces were identified by ray intersection testing or scan conversion;

a pixel fragment shader configured to shade a surface visible from a pixel fragment in said 2-D screen space, identified by said visible surface determination unit, wherein the pixel fragment shader is operable to emit a ray originating from the surface being shaded, that is to be tested for intersection in the 3-D scene by the ray tracing intersection unit using ray tracing, and

a ray intersection shader operable to contribute to shading o fthe visible surface associated with the pixel fragment shader, and b said ray intersection shader being independently schedulable from the pixel fragment shader that emitted the ray.

17. The apparatus for graphics rendering of claim 16, wherein the surfaces in the 3-D scene are specified in a 3-D coordinate system, and the ray intersection shader is operable to express shading operations in the 3-D coordinate system.

18. The apparatus for graphics rendering of claim 16, further comprising a collector to group ray intersections according to a common shader that is scheduling to be executed for each ray intersection of a collection.

19. A system for graphics rendering, comprising:

a visible surface determination module operable to concurrently execute both ray tracing and scan conversion operations, to determine a respective visible surface within a 3-D scene, from each of one or more view positions specified to the visible surface determination module;

a normalizer configured to provide normalized shader data for use by a shader during execution, the normalized shader data comprising a specified set of attributes regardless of whether the visible surfaces were identified by ray intersection testing or scan conversion; and

a processor coupled to a non-transitory memory, the non-transitory memory storing instructions for configuring the processor to execute one or more surface shaders, each associated with a surface in the 3-D scene, each of the surface shaders operable to

activate in response to a scan conversion input, from the visible surface determination module, comprising data for a surface in the 3-D scene visible from a pixel fragment in a 2-D screen space, and to interface with the visible surface determination module to emit a ray originating from the surface visible from the pixel fragment, that will be tested for intersection in the 3-D scene by the visible surface determination module, and

activate in response to a ray tracing input, from the visible surface determination module, comprising an indication of a detected ray intersection in the 3-D scene, and responsively determine a shading effect for one or more pixel fragments the 2-D screen space.

20. The system for graphics rendering of claim 19, further comprising a coordinate transform unit operable to receive origin information for the ray to be emitted, specified in a 2-D coordinate system, to convert the ray origin information into an origin specified in a 3-D coordinate system and to provide the origin specified in the 3-D coordinate system to the visible surface determination module.