Video Processor Tile Compression for Memory Bandwidth Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern video processing devices face increased memory bandwidth requirements due to higher pixel resolutions, leading to inefficiencies in memory usage and bandwidth consumption, particularly when handling large video or graphical frames that exceed the memory capacity of the device.
Innovation Solution
The implementation of frame buffer compression techniques, utilizing span units and tile units, which compress data objects like video and graphical frames into compressed frame buffers, reducing memory bandwidth needs by leveraging spatial correlation and optimizing processing efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pixel resolution of video frames is increased, then picture quality is improved, but memory bandwidth required for processing increases
Solution Approach 1:
The patent segments video frames into smaller blocks (e.g., 16x16 pixel blocks) that can be independently compressed and stored. This segmentation allows the system to process and store only the essential information from each block, significantly reducing the total memory bandwidth required while maintaining overall picture quality.
Solution Approach 2:
The patent extracts only the most important information from video frame blocks by analyzing spatial correlation and removing redundant data. This extraction process eliminates unnecessary memory bandwidth consumption while preserving the critical visual information needed for high picture quality.
2Quantity of substance
If frame buffer compression is implemented, then memory bandwidth is reduced, but processing complexity increases
Solution Approach 1:
The patent applies different compression strategies to different regions of video frames based on local characteristics. By analyzing spatial correlation locally and applying appropriate compression techniques to each region, the system achieves effective memory bandwidth reduction without uniformly increasing processing complexity across the entire frame.
Solution Approach 2:
The patent performs preliminary analysis of video frame blocks to identify spatial correlation patterns before compression. This preliminary action allows the system to pre-determine the most efficient compression approach for each block, reducing overall processing complexity while achieving effective memory bandwidth reduction.
3Quantity of substance
If data is compressed into smaller units, then memory bandwidth efficiency is improved, but granularity loss increases
Solution Approach 1:
The patent changes the parameter of data representation by transforming video pixel data into compressed forms that preserve essential visual information while reducing data volume. By carefully selecting compression parameters and thresholds, the system achieves memory bandwidth efficiency improvements while minimizing granularity loss through intelligent parameter selection.
Data Source
AI summary
A video processing device includes a video processing unit that decodes a video input signal into a decoded video signal in accordance with a video compression protocol, based on uncompressed video frame data. A tile engine includes a tile accumulation module that accumulates the uncompressed video frame data into a plurality of tile units, wherein each of the plurality of tile units includes a plurality of video span units. A tile compression/decompression module generates compressed video frame data for storage in a compressed video frame buffer by compressing the plurality of video span units into a plurality of compressed video span units and further that retrieves the compressed video frame data from the compressed video frame buffer by retrieving the plurality of compressed video span units and generating the uncompressed video frame data by decompressing the plurality of compressed video span units.