Display Pipe Active Region Pixel Fetching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current graphics processing systems face challenges in efficiently managing memory access for high-resolution image and video frames, leading to increased bandwidth demands and potential battery depletion in portable devices, due to the need to fetch and process large volumes of pixel data, especially when using scaling algorithms like bilinear and bicubic interpolation.
Innovation Solution
The implementation of a display pipe system that supports user interface units with programmable registers to define active regions within frames, allowing only pixels within these regions to be fetched and processed, treating pixels outside the active regions as transparent with an Alpha value of zero, thereby reducing memory access and processing demands. This system includes built-in scalers that use context pixels to generate output pixels based on available active pixels, using bilinear scaling algorithms and handling inactive pixels by determining their color values based on adjacent active pixels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the entire frame is fetched from memory for display, then complete image data is available for processing, but memory bandwidth demand increases and processing power consumption increases
Solution Approach 1:
The patent divides the frame into active regions (containing valid pixel data) and inactive regions (containing no valid data). The fetch unit is configured to selectively fetch only the active regions from memory, excluding inactive regions. This segmentation approach reduces the volume of data transferred from memory, thereby reducing memory bandwidth demand and power consumption while maintaining complete image data for the displayed portion.
2Manufacturing precision
If scaling algorithms are applied to upscale pixels, then output resolution is improved, but processing complexity and memory access requirements increase
Solution Approach 1:
The patent applies scaling algorithms locally within active regions rather than processing the entire frame. The blend unit performs scaling operations only on fetched active pixels and their neighboring pixels within the active region boundaries. This localized processing approach maintains output pixel accuracy for displayed areas while reducing overall processing complexity and memory access requirements by excluding inactive regions from scaling operations.
3Loss of information
If pixels outside active regions are fetched and processed, then no data is lost, but memory access time and processing time increase
Solution Approach 1:
The patent extracts and processes only the relevant pixel data from active regions, excluding pixels outside active regions that contain no valid information. The fetch unit is configured to retrieve only active pixels and their necessary neighboring pixels for scaling operations, while the blend unit processes only these extracted pixels. This extraction approach prevents information loss within active regions while eliminating wasted processing time on inactive pixels.
Data Source
AI summary
A display pipe may include fetch circuitry and a scaler unit, and registers programmable with information that defines active regions of an image frame. Pixels within the active regions are active pixels to be displayed, pixels outside of the active regions are inactive pixels not to be displayed. The fetch circuitry may retrieve frames from memory, retrieving the active pixels and not retrieving the inactive pixels as defined by the programmed contents of the registers. A scaler unit may produce scaled pixels from the fetched pixels, basing each scaled pixel on a respective corresponding set of pixels. When a given pixel of the respective corresponding set of pixels is an inactive pixel, the scaler unit may assign an estimated value to the given pixel based on one or more active pixels in the respective corresponding set of pixels. The scaler unit may provide the scaled pixels to a blend unit for blending with other pixels.


