Adaptive Frame Buffer Compression for Power Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data processing systems that store image data in a frame buffer face challenges in reducing power consumption, particularly in mobile devices where battery life is a constraint, as existing methods do not effectively balance energy expenditure in compressing and decompressing image data based on static and dynamic image portions.
Innovation Solution
The system employs compression and decompression circuitry that adjusts compression control parameters to increase the compression ratio for static image portions, while maintaining losslessness and increasing energy consumption for compression, thereby reducing the amount of data written and read from the frame buffer, and uses larger block sizes for static areas to optimize burst data accesses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If compression ratio is increased for static image portions, then data transfer volume is reduced, but energy consumed by compression circuitry increases
Solution Approach 1:
The system dynamically adjusts compression control parameters based on the temporal stability of image portions. Compression ratio is increased for static portions that remain unchanged across multiple frames, while dynamic portions use lower compression. This dynamic adaptation resolves the contradiction by applying high compression only when energy expenditure is justified by sustained data transfer reduction.
Solution Approach 2:
The system changes compression control parameters (such as compression ratio and block size) based on detected image stability. For static portions, parameters are adjusted to achieve higher compression ratios with larger block sizes, reducing data transfer volume while the energy cost is amortized over multiple frames where the compressed data remains valid.
2Quantity of substance
If compression ratio is increased, then data transfer volume is reduced, but power consumption is increased
Solution Approach 1:
The system dynamically adapts compression strategy based on image stability detection. High compression ratios are applied only to portions of the image that remain static across multiple frames, allowing the energy investment in compression to be amortized over multiple read operations, thereby reducing overall power consumption relative to the data transfer volume reduction achieved.
Solution Approach 2:
Different compression ratios are applied to different portions of the image based on their temporal characteristics. Static portions use high compression ratios while dynamic portions use lower compression, optimizing the balance between data transfer volume reduction and power consumption on a local basis rather than uniformly across the entire image.
3Productivity
If larger block sizes are used for static areas, then burst data access efficiency is improved, but compression complexity increases
Solution Approach 1:
The image is segmented into static and dynamic portions, with different compression strategies applied to each. Static portions use larger block sizes optimized for burst data access, while dynamic portions use smaller blocks. This segmentation allows the system to exploit burst access efficiency for static content without unnecessarily increasing compression complexity across the entire image.
Solution Approach 2:
Different block sizes are used for different regions of the image based on their temporal stability. Static areas use larger blocks to optimize burst data access efficiency, while dynamic areas use smaller blocks. The compression circuitry adapts its complexity locally, applying more sophisticated large-block compression only where it provides benefit.
Data Source
AI summary
Image data is subject to compression and decompression when it is respectively written to and read from a frame buffer. If a portion of the image data is identified as static (subject to less than a threshold amount of change for greater than a threshold time), then compression control parameters used for compression of that portion of the image are adjusted so as to increase the compression ratio achieved, hold the degree of lossiness substantially constant and increase the energy consumed while compressing that portion. The increased energy consumption during this high compression ratio compression is likely compensated for by a reduction in energy subsequently consumed when writing that frame-buffer image data to the frame buffer and reading that frame-buffer image data multiple times from the frame buffer. The compression characteristics varied may be to increase the block size used in the compression. Other variations in compression applied may be to change from single-pass compression to multi-pass compression, switch compression on and off altogether, or reorder the data when it has been compressed so as to match the order it will be read and so achieve support for longer read burst.