Frame Buffer Compression Using Wavelet Transform Residuals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing architectures face challenges in improving performance and power usage, particularly in shared memory hierarchies between CPU cores and graphics processors, with existing color buffer compression techniques being insufficient for further reductions in memory traffic, necessitating more complex algorithms to achieve significant advantages.
Innovation Solution
The implementation of aggressive and complex compression techniques, including lossy methods like discrete cosine transforms and wavelet-based transforms, is adapted for frame buffer compression, allowing for greater bandwidth savings and improved performance by encoding residuals between a smooth image representation and the original image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If simple compression algorithms are used, then device complexity is reduced, but memory bandwidth reduction is insufficient
Solution Approach 1:
The patent transforms the color buffer data by changing its representation parameters through wavelet transformation, converting spatial domain color data into frequency domain coefficients. This parameter transformation enables more efficient compression by exploiting the statistical properties of the transformed data, achieving greater bandwidth reduction without proportionally increasing algorithmic complexity.
Solution Approach 2:
The patent divides the color buffer into multiple tiles or blocks, applying compression independently to each segment. This segmentation allows the system to use more sophisticated compression techniques on smaller, manageable units while maintaining overall system performance, effectively reducing memory bandwidth without requiring the entire buffer to be processed by a single complex algorithm.
2Loss of information
If lossless compression is used, then image quality is maintained, but compression ratio is limited
Solution Approach 1:
The wavelet transformation changes the parameter representation of image data from spatial to frequency domain, where energy is concentrated in fewer coefficients. This parameter change enables lossless compression at higher ratios by efficiently encoding the transformed coefficients, and allows selective lossy compression of less important high-frequency coefficients when bandwidth reduction is prioritized over perfect reconstruction.
3Quantity of substance
If more complex compression algorithms are implemented, then memory bandwidth reduction improves, but power consumption increases
Solution Approach 1:
By dividing the compression task into smaller tile-based segments that can be processed independently and in parallel, the patent reduces the computational burden on any single processing unit. This segmentation enables more aggressive compression to be applied locally without proportionally increasing total power consumption, as the work can be distributed across multiple lower-power processing elements.
Data Source
AI summary
An apparatus and method for frame buffer compression. For example, one embodiment of a method comprises: identifying a wavelet function to compress an original frame buffer image; performing a wavelet transform on the original frame buffer image using the wavelet function; using a specified subset of wavelet coefficients of the wavelet function to construct an approximated image function; determining residuals comprising differences between the original frame buffer image and the approximated image function; and storing the residuals for subsequent use in recreating the original frame buffer image from the approximated image function.