Compressed Z-Plane Storage for Faster Graphics Z-Fill

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional Z (depth) pipelines in graphics processing are constrained by slower Z-fill operations due to limitations in hardware and bandwidth, leading to reduced performance.

Innovation Solution

Implementing a compressed depth plane computation and storage system using HiZ hardware, allowing for direct writing to cache in a compressed format and lossless decompression using unsigned mathematics, along with a common compressed data header for efficient compression/decompression across various data types.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If compressed depth plane computation and storage system is implemented using HiZ hardware, then graphics processing performance is enhanced and Z-fill operations run at HiZ rate, but device complexity increases due to compression/decompression hardware requirements

Engineering Contradiction:
Improvegraphics processing performanceVSAvoidcompression/decompression hardware
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent transforms depth data from uncompressed format to compressed format by changing its parameter representation. The compression algorithm encodes depth values using fewer bits by exploiting spatial coherence and redundancy in depth buffers, while the decompression hardware reconstructs the original depth values. This parameter transformation enables faster Z-fill operations at HiZ rate while managing hardware complexity through efficient encoding/decoding mechanisms.

Inventive Principle:
Principle #35Parameter changes

2Quantity of substance

If direct writing to cache in compressed format is implemented, then bandwidth utilization is improved and memory efficiency increases, but lossless decompression complexity increases

Engineering Contradiction:
Improvememory bandwidth utilizationVSAvoidlossless decompression
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent creates a compressed copy of the depth buffer data that can be stored in cache memory with reduced bandwidth requirements. The compression process generates a compact representation that preserves all necessary depth information. The decompression hardware then reconstructs the full-depth data from this compressed copy when needed, enabling efficient memory utilization while maintaining data integrity through lossless reconstruction.

Inventive Principle:
Principle #26Copying

3Measurement precision

If per-pixel depth test hardware is used for Z-fill operations, then depth precision is maintained, but operation speed is limited by hardware bandwidth

Engineering Contradiction:
Improvedepth precisionVSAvoidZ-fill operation speed
Core Design Contradiction:
Measurement precisionVSSpeed

Solution Approach 1:

The patent segments the depth buffer into hierarchical regions (HiZ blocks) that can be processed independently at different levels of detail. The coarse depth test operates on compressed HiZ blocks to quickly determine visibility, while per-pixel depth testing is performed only on visible fragments. This segmentation allows the system to maintain depth precision for visible pixels while achieving faster overall Z-fill performance through parallel processing of hidden regions.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4095789B1Generation and storage of compressed z-planes in graphics processing
Publication Date: 2025.10.29 INTEL CORP
  • EP4095789B1 patent drawingFigure 1
  • EP4095789B1 patent drawingFigure 2A
  • EP4095789B1 patent drawingFigure 2B

AI summary

Generation and storage of compressed z-planes in graphics processing is described. An example of a processor includes a rasterizer to generate a fragment of pixel data including blocks of pixel data; a depth pipeline to receive the fragment, the pipeline including a first and second depth test hardware, the first depth test hardware to perform a coarse depth test including determining minimum and maximum depths for each block; and a depth buffer, wherein the processor is to determine whether the fragment meets requirements that the fragment fully covers a tile of pixel data and passes a first depth test, and that each of the minimum and maximum depths of the fragment has a same sign and exponent, and, upon determining that the fragment meets the requirements, to generate a compressed depth plane utilizing the first depth test and update the depth buffer with the compressed depth plane.