# US 2026/0065407 A1 · Graphics processing

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

- Number: US 2026/0065407 A1
- Document ID: US20260065407A1
- Status: Published application
- Kind: A1
- Application number: 18/819,003
- Filed: 2024-08-29
- Published: 2026-03-05
- Earliest claimed: 2024-08-29
- Assignees: Arm Limited
- Applicants: Arm Limited
- Inventors: Jakob Axel Fries; Olof Henrik Uhrenholt; Wing-Tsi Henry Wong
- Web page: https://patentgrep.com/patent/US20260065407A1

## CPC

- G06T 1/20: Processor architectures; Processor configuration, e.g. pipelining
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A tile-based graphics processing system is disclosed. The system selects a rendering configuration for generating one or more rendering tiles of a render output based on a compression unit size and/or shape that is to be used to encode the one or more rendering tiles of the render output.

## Claims

### Claim 1

A method of operating a tile-based graphics processing system; the method comprising:
selecting a rendering configuration for generating one or more rendering tiles of a render output based on a compression unit size and/or shape that is to be used to encode the one or more rendering tiles of the render output;
generating the one or more rendering tiles of the render output using the selected rendering configuration; and
encoding the one or more rendering tiles of the render output using the compression unit size and/or shape.

### Claim 2 (depends on 1)

The method of claim 1, wherein the graphics processing system comprises an accumulation buffer and an encoder; and wherein encoding the one or more rendering tiles comprises:
the accumulation buffer accumulating render output data for one or more compression units, and providing a complete compression unit to the encoder for encoding; and
the encoder encoding the complete compression unit;
wherein the accumulation buffer providing a complete compression unit to the encoder for encoding comprises:
determining whether the accumulation buffer has not accumulated all of the render output data for the compression unit; and
when it is determined that the accumulation buffer has not accumulated all of the render output data for the compression unit:
requesting the render output data for the compression unit that the accumulation buffer has not accumulated from a memory system; and
combining render output data returned from the memory system with render output data for the compression unit that the accumulation buffer has accumulated so as to provide a complete version of the compression unit for encoding.

### Claim 3 (depends on 1)

The method of claim 1, wherein the graphics processing system comprises an encoder and an accumulation buffer operable to provide render output data to the encoder for encoding; and wherein selecting a rendering configuration for generating the one or more rendering tiles comprises: selecting a rendering configuration that will cause the accumulation buffer to provide render output data for a complete compression unit to the encoder without requiring a request for render output data from a memory system.

### Claim 4 (depends on 1)

The method of claim 1, wherein selecting a rendering configuration for generating the one or more rendering tiles comprises selecting a size and/or shape of the one or more rendering tiles based on the compression unit size and/or shape.

### Claim 5 (depends on 4)

The method of claim 4, comprising selecting a size and/or shape of the one or more rendering tiles that will reduce a number of compression units overlapping a rendering tile.

### Claim 6 (depends on 1)

The method of claim 1, wherein selecting a rendering configuration for generating the one or more rendering tiles comprises selecting a processing order for processing the one or more rendering tiles based on the compression unit size and/or shape.

### Claim 7 (depends on 6)

The method of claim 6, comprising selecting a processing order for processing the one or more rendering tiles that will result in rendering tiles that overlap a compression unit being processed successively.

### Claim 8 (depends on 1)

The method of claim 1, wherein generating the one or more rendering tiles comprises:
issuing the one or more rendering tiles for processing in a processing order; and
obliging processing of the one or more rendering tiles to finish in the processing order.

### Claim 9 (depends on 1)

The method of claim 1, wherein the graphics processing system comprises one or more processing cores, and wherein generating the one or more rendering tiles of the render output comprises assigning one or more rendering regions of the render output to a respective processing core for processing; and
wherein selecting a rendering configuration for generating the one or more rendering tiles comprises selecting a size and/or shape of the one or more rendering regions based on the compression unit size and/or shape.

### Claim 10 (depends on 9)

The method of claim 9, comprising selecting a size and/or shape of the one or more rendering regions such that a rendering region encompasses one or more complete compression units.

### Claim 11 (depends on 1)

A non-transitory computer readable storage medium storing software code which when executing on a processor performs the method of claim 1.

### Claim 12

A tile-based graphics processing system comprising:
a configuration circuit configured to select a rendering configuration for generating one or more rendering tiles of a render output based on a compression unit size and/or shape that is to be used to encode the one or more rendering tiles of the render output;
a rendering circuit configured to generate one or more rendering tiles of a render output using a rendering configuration selected by the configuration circuit; and
an encoder configured to encode one or more rendering tiles of a render output generated by the rendering circuit using a compression unit size and/or shape that a rendering configuration used by the rendering circuit to generate the one or more rendering tiles of the render output was selected based on.

### Claim 13 (depends on 12)

The system of claim 12, comprising an accumulation buffer configured to accumulate render output data generated by the rendering circuit for one or more compression units, and provide a complete compression unit to the encoder for encoding;
wherein the accumulation buffer is configured to provide a complete compression unit to the encoder for encoding by:
determining whether the accumulation buffer has not accumulated all of the render output data for the compression unit; and
when it is determined that the accumulation buffer has not accumulated all of the render output data for the compression unit:
requesting the render output data for the compression unit that the accumulation buffer has not accumulated from a memory system; and
combining render output data returned from the memory system with render output data for the compression unit that the accumulation buffer has accumulated so as to provide a complete version of the compression unit for encoding.

### Claim 14 (depends on 12)

The system of claim 12, comprising an encoder and an accumulation buffer operable to provide render output data to the encoder for encoding;
wherein the configuration circuit is configured to select a rendering configuration for generating one or more rendering tiles by selecting a rendering configuration that will cause the accumulation buffer to provide render output data for a complete compression unit to the encoder without requiring a request for render output data from a memory system.

### Claim 15 (depends on 12)

The system of claim 12, wherein the configuration circuit is configured to select a rendering configuration for generating one or more rendering tiles by selecting a size and/or shape of the one or more rendering tiles based on the compression unit size and/or shape.

### Claim 16 (depends on 15)

The system of claim 15, wherein the configuration circuit is configured to select a size and/or shape of the one or more rendering tiles that will reduce a number of compression units overlapping a rendering tile.

### Claim 17 (depends on 12)

The system of claim 12, wherein the configuration circuit is configured to select a rendering configuration for generating one or more rendering tiles by selecting a processing order for processing the one or more rendering tiles based on the compression unit size and/or shape.

### Claim 18 (depends on 17)

The system of claim 17, wherein the configuration circuit is configured to select a processing order for processing the one or more rendering tiles that will result in rendering tiles that overlap a compression unit being processed successively.

### Claim 19 (depends on 12)

The system of claim 12, wherein the rendering circuit comprises one or more processing cores, and an assigning circuit configured to assign one or more rendering regions of a render output to a respective processing core for processing; and
wherein the configuration circuit is configured to select a rendering configuration for generating one or more rendering tiles by selecting a size and/or shape of the one or more rendering regions based on the compression unit size and/or shape.

### Claim 20 (depends on 19)

The system of claim 19, wherein the configuration circuit is configured to select a size and/or shape of the one or more rendering regions such that a rendering region encompasses one or more complete compression units.

### Claim 21

A tile-based graphics processor comprising:
a rendering circuit configured to generate one or more rendering tiles of a render output using a rendering configuration selected based on a compression unit size and/or shape that is to be used to encode the one or more rendering tiles of the render output; and
an encoder configured to encode one or more rendering tiles of a render output generated by the rendering circuit using a compression unit size and/or shape that a rendering configuration used by the rendering circuit to generate the one or more rendering tiles of the render output was selected based on.
