Extrapolating Nonresident Mipmap Data Using Resident Levels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The appearance of texture map data is compromised when high resolution mipmaps are retrieved from high latency storage resources, leading to blurry images that only sharpen once the mipmaps are available in physical memory, and there is a need to smoothly transition from extrapolated filtering to interpolated filtering during this process.
Innovation Solution
A multi-threaded graphics processor extrapolates low resolution mipmaps stored in physical memory to produce sharper texture values while high resolution mipmaps are retrieved from high latency storage, using an extrapolation threshold LOD to determine when to switch from extrapolated to interpolated filtering, and computes a deltaLOD to determine an extrapolation weight for producing filtered texel values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If high resolution mipmaps are stored in high latency storage resources, then memory capacity is improved, but image quality deteriorates during retrieval
Solution Approach 1:
The patent segments mipmap data into resident and non-resident portions, storing high resolution mipmaps in high latency storage while maintaining low resolution mipmaps in fast memory. This segmentation allows the system to access frequently used low resolution data quickly while keeping high resolution data available but not constantly loaded, resolving the contradiction between memory capacity and image quality during retrieval.
Solution Approach 2:
The patent performs preliminary extrapolation of high resolution texture values from available low resolution mipmaps before the high resolution mipmaps are fully retrieved from storage. This preliminary action provides usable high quality texture data during the retrieval process, eliminating the image quality deterioration that would otherwise occur while waiting for high resolution data to load.
2Speed
If low resolution mipmaps are used during retrieval, then access speed is improved, but image quality deteriorates
Solution Approach 1:
The system performs preliminary extrapolation to generate high resolution texture values from low resolution mipmaps that are already in fast memory. This allows the system to maintain fast access speeds using low resolution data while simultaneously producing high quality output through extrapolation, resolving the contradiction between access speed and image quality.
Solution Approach 2:
The patent changes the resolution parameter of texture data through extrapolation algorithms. By taking low resolution mipmap data and extrapolating it to produce high resolution texture values, the system effectively transforms the resolution parameter to achieve high image quality while maintaining fast access speeds using the lower resolution source data.
3Manufacturing precision
If extrapolated filtering is used during transition, then image quality is improved, but filtering complexity increases
Solution Approach 1:
The patent dynamically adjusts the filtering approach based on the availability of mipmap data. When high resolution mipmaps are not yet retrieved, the system uses extrapolated filtering from low resolution data. Once high resolution mipmaps become available, the system transitions to standard interpolation filtering. This dynamic adaptation resolves the contradiction by applying complex extrapolation only when necessary, rather than always.
Data Source
AI summary
A multi-threaded graphics processor is configured to use to extrapolate low resolution mipmaps stored in physical memory to produce extrapolated texture values while high resolution nonresident mipmaps are retrieved from a high latency storage resource and converted into resident mipmaps. The extrapolated texture values provide an improved image that appears sharper compared with using the low resolution mipmap level texture data in place of the temporarily unavailable high resolution mipmap level texture data. An extrapolation threshold LOD is used to determine when extrapolated magnification or minification texture filtering is used. The extrapolation threshold LOD may be used to smoothly transition from using extrapolated filtering to using interpolated filtering when a nonresident mipmap is converted to a resident mipmap.