# US 2025/0095277 A1 · Early release of resources in ray tracing hardware

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0095277 A1
- Document ID: US20250095277A1
- Status: Published application
- Kind: A1
- Application number: 18/964,127
- Filed: 2024-11-29
- Published: 2025-03-20
- Earliest claimed: 2020-06-18
- Assignees: NVIDIA Corporation
- Applicants: NVIDIA Corporation
- Inventors: Gregory MUTHLER; John BURGESS; Ronald Charles BABICH; William Parsons Newhall
- Web page: https://patentgrep.com/patent/US20250095277A1

## CPC

- G06T 15/06: Ray-tracing
- G06F 9/48: Program initiating; Program switching, e.g. by interrupt
- G06F 9/5027: {the resource being a machine, e.g. CPUs, Servers, Terminals}
- G06T 17/10: Constructive solid geometry [CSG] using solid primitives, e.g. cylinders, cubes
- G06T 2210/21: Collision detection, intersection
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Techniques are disclosed for improving the throughput of ray intersection or visibility queries performed by a ray tracing hardware accelerator. Throughput is improved, for example, by releasing allocated resources before ray visibility query results are reported by the hardware accelerator. The allocated resources are released when the ray visibility query results can be stored in a compressed format outside of the allocated resources. When reporting the ray visibility query results, the results are reconstructed based on the results stored in the compressed format. The compressed format storage can be used for ray visibility queries that return no intersections or terminate on any hit ray visibility query. One or more individual components of allocated resources can also be independently deallocated based on the type of data to be returned and/or results of the ray visibility query.

## Claims

### Claim 1

A ray tracing hardware method comprising:
receiving a plurality of thread requests including a first and second thread requests, each thread request including information defining a ray and a primitive to test for intersection with the ray;
allocating a resource for storing intersection test results;
processing the first and second thread requests to provide an intersection test result for each thread request;
reporting intersection test results stored in the allocated resource to a cooperating processor for a first type of primitive intersection test result; and
reporting ray intersection test results to the cooperating processor using data that was predetermined before the intersection test was performed for at least one second type of primitive intersection test result that is different from the first type of primitive intersection test result.

### Claim 2 (depends on 1)

The ray tracing hardware method of claim 1, wherein the first type of primitive intersection test results indicates that the ray intersects the primitive.

### Claim 3 (depends on 1)

The ray tracing hardware method of claim 1, wherein the second type of primitive intersection test results indicates that the ray does not intersect the primitive.

### Claim 4 (depends on 1)

The ray tracing hardware method of claim 1, wherein the ray intersection test results are concurrently reported for the plurality of thread requests.

### Claim 5 (depends on 1)

The ray tracing hardware method of claim 1, wherein the ray intersection test results are concurrently reported in groups for two or more thread requests of the plurality of thread requests.

### Claim 6 (depends on 1)

The ray tracing hardware method of claim 1, wherein the ray intersection test results are concurrently reported for a portion of the plurality of thread requests when the intersection test is completed for a predetermined number of thread requests.

### Claim 7

A ray tracing hardware coupled to a processor and comprising circuitry configured to perform operations comprising:
receive a plurality of thread requests including a first and second thread requests, each thread request including information defining a ray and a primitive to test for intersection with the ray;
allocate a resource for storing intersection test results;
process the first and second thread requests to provide an intersection test result for each thread request;
report intersection test results stored in the allocated resource to a cooperating processor for a first type of primitive intersection test result; and
report ray intersection test results to the cooperating processor using data that was predetermined before the intersection test was performed for at least one second type of primitive intersection test result that is different from the first type of primitive intersection test result.

### Claim 8 (depends on 7)

The ray tracing hardware of claim 7, wherein the first type of primitive intersection test results indicates that the ray intersects the primitive.

### Claim 9 (depends on 7)

The ray tracing hardware of claim 7, wherein the second type of primitive intersection test results indicates that the ray does not intersect the primitive.

### Claim 10 (depends on 7)

The ray tracing hardware of claim 7, wherein the ray intersection test results are concurrently reported for the plurality of thread requests.

### Claim 11 (depends on 7)

The ray tracing hardware of claim 7, wherein the ray intersection test results are concurrently reported in groups for two or more thread requests of the plurality of thread requests.

### Claim 12 (depends on 7)

The ray tracing hardware of claim 7, wherein the ray intersection test results are concurrently reported for a portion of the plurality of thread requests when the intersection test is completed for a predetermined number of thread requests.

### Claim 13

A ray tracing hardware method comprising:
receiving a plurality of thread requests, each thread request including information defining a ray and a primitive to test for intersection with the ray;
allocating a resource for each of the plurality of thread requests for storing intersection test results;
processing the plurality of thread requests to provide an intersection test result for each thread request;
deallocating the resources allocated for one or more of the plurality of thread requests for which a predetermined ray intersection test result is determined before the intersection test results are reported; and
reporting intersection test results stored in the allocated resource to a cooperating processor.
