Sub-Block Alpha Compression for Guaranteed Memory Bandwidth Savings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data compression methods in GPUs and CPUs face challenges in balancing memory bandwidth and power consumption while maintaining high-quality rendering, particularly in mobile devices, where memory space and bandwidth are limited resources.
Innovation Solution
A lossy data compression method that divides pixel blocks into sub-blocks and analyzes alpha channel values to select from a set of candidate compression modes, ensuring a guaranteed compression threshold is met by combining lossless and lossy techniques, reducing memory requirements and bandwidth.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If data compression is applied to reduce memory bandwidth and power consumption, then power consumption and memory bandwidth usage are reduced, but data quality may deteriorate
Solution Approach 1:
The pixel block is divided into multiple sub-blocks, and different compression modes are applied to different sub-blocks based on their local characteristics. This segmentation allows the system to apply aggressive compression only where acceptable, preserving quality in critical regions while achieving overall compression goals.
Solution Approach 2:
The patent applies compression selectively rather than uniformly across all pixels. By analyzing alpha channel values and applying compression only to sub-blocks where it is acceptable, the system achieves partial compression action that balances quality preservation with bandwidth reduction.
2Productivity
If lossy compression modes are used to achieve higher compression ratios, then memory bandwidth is reduced more effectively, but rendering quality deteriorates
Solution Approach 1:
Different compression qualities are applied to different regions of the image based on local characteristics. Sub-blocks with uniform or less important visual characteristics receive higher compression, while other regions maintain higher quality, achieving local optimization of the compression-quality tradeoff.
Solution Approach 2:
The patent changes compression parameters dynamically based on local image characteristics, specifically analyzing alpha channel values to determine appropriate compression modes for each sub-block, thereby adapting the compression ratio to local quality requirements.
3Loss of information
If compression analysis is performed on every pixel block to maintain quality, then data quality is preserved, but processing time and complexity increase
Solution Approach 1:
By segmenting the pixel block into sub-blocks, the patent reduces the processing complexity compared to analyzing every pixel individually. The segmentation allows for efficient parallel processing and reduces the computational burden while maintaining quality through localized analysis.
Solution Approach 2:
The patent performs preliminary analysis of alpha channel values before applying compression, allowing the system to pre-determine appropriate compression modes for each sub-block. This preliminary action prevents unnecessary processing and optimizes the compression pipeline efficiency.
Data Source
AI summary
Lossy methods and hardware for compressing data and the corresponding decompression methods and hardware are described. The lossy compression method comprises dividing a block of pixels into a number of sub-blocks and then analysing, for each sub-block, and selecting one of a candidate set of lossy compression modes. The analysis may, for example, be based on the alpha values for the pixels in the sub-block. In various examples, the candidate set of lossy compression modes comprises at least one mode that uses a fixed alpha channel value for all pixels in the sub-block and one or more modes that encode a variable alpha channel value.