CPU Performance Control via Frame Buffer Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional power management algorithms in mobile computing devices are ineffective in accurately determining when to modify the operating mode of CPUs and GPUs, leading to inefficiencies in energy usage and performance, particularly for interactive and visual workloads.
Innovation Solution
Implementing performance control algorithms that monitor cycle-to-cycle jitter, GPU utilization rates, and specific instruction types to dynamically adjust the operating mode of CPUs, including voltage and frequency, based on feedback from components like frame buffers and memory controllers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If conventional power management algorithms monitor only CPU workload duration to determine operating mode changes, then energy is conserved for short bursty workloads, but interactive performance and animation smoothness deteriorate because the algorithms cannot accurately detect when mode changes are needed for visual workloads
Solution Approach 1:
The patent implements feedback loops that continuously monitor frame rate metrics (such as frames per second) and UI animation smoothness, using this feedback to dynamically adjust CPU operating modes. This closed-loop control enables the system to respond to actual visual performance needs rather than relying on predetermined workload duration thresholds, thereby maintaining both energy efficiency and interactive performance.
Solution Approach 2:
The system dynamically adjusts CPU operating modes based on real-time monitoring of visual workload characteristics rather than static thresholds. The operating mode transitions are made adaptive and flexible, allowing the system to optimize performance for interactive applications while conserving energy during less demanding periods, thus resolving the contradiction between energy conservation and interactive performance.
2Productivity
If CPU voltage and frequency are increased to improve animation smoothness and UI responsiveness, then interactive performance improves, but energy consumption increases
Solution Approach 1:
The patent applies partial action by increasing CPU voltage and frequency only to the extent necessary to achieve target frame rates and animation smoothness thresholds, rather than maintaining constantly high performance. The system provides just enough performance headroom to ensure smooth visuals while avoiding excessive energy consumption during periods when maximum performance is not required.
Solution Approach 2:
The system dynamically changes CPU operating parameters (voltage and frequency) based on monitored visual workload demands. By adjusting these parameters in response to actual performance needs rather than maintaining fixed high settings, the system achieves smooth animations only when necessary, thereby balancing visual performance with energy conservation.
3Productivity
If CPU operating mode is frequently adjusted to respond to varying workload demands, then performance is optimized for different tasks, but system stability and power delivery complexity increase
Solution Approach 1:
The patent implements periodic sampling and monitoring of performance metrics at structured intervals, rather than continuous adjustment. This periodic approach allows the system to respond to workload changes systematically while avoiding the instability that would result from overly frequent mode transitions, thus managing complexity while maintaining performance optimization.
Solution Approach 2:
The patent employs a unified power management framework that handles multiple workload types (interactive UI, graphical rendering, compute tasks) through a single coherent control mechanism. This universal approach simplifies the power management system compared to having separate specialized controllers for each workload type, reducing overall system complexity while maintaining optimized performance across diverse tasks.
Data Source
AI summary
The invention provides a technique for targeted scaling of the voltage and/or frequency of a processor included in a computing device. One embodiment involves scaling the voltage/frequency of the processor based on the number of frames per second being input to a frame buffer in order to reduce or eliminate choppiness in animations shown on a display of the computing device. Another embodiment of the invention involves scaling the voltage/frequency of the processor based on a utilization rate of the GPU in order to reduce or eliminate any bottleneck caused by slow issuance of instructions from the CPU to the GPU. Yet another embodiment of the invention involves scaling the voltage/frequency of the CPU based on specific types of instructions being executed by the CPU. Further embodiments include scaling the voltage and/or frequency of a CPU when the CPU executes workloads that have characteristics of traditional desktop/laptop computer applications.