• 03/25/2021
  • Public Content
Contents

Analysis Tips

To start, begin by measuring your app’s baseline power usage in multiple cases and at different loads:
  • At near idle, as in the user interface (UI) during videos
  • Under an average load during a typical scene with average effects
The worst-case load may not occur where you expect it to. We have seen very high frame rates (1000 frames per second (fps)) during cut-scene video playback in certain apps, a situation that can cause the GPU and CPU to use unnecessary power. As you study your application, try a few of these tips:
  • Measure how long (on average) your application can run on battery power, and compare its performance with other, similar apps. Measuring power consumption regularly informs you if any recent changes caused your app to use more power.
  • The Battery Life Analyzer (BLA) is a good (Windows-only) tool for this work. For more information, see this article showcasing BLA, and how it can collect high-level data and analyze an app's power use. If the data BLA provides shows that you have issues residing in the wrong C-states for too long, it is time to look deeper.
  • If your app is reported as deficient or there are unexpected wakeups, start optimizing for power. To do so, you will want to look at the Windows Performance Analyzer (WPA) tool, which showcases workflow using WPA for CPU analysis.
  • VTune™ Profiler XE is also useful to get power call stacks, since it can identify the cause of the wakeup.
Use the data gained through these methods to reduce or consolidate wakeups, thus remaining in a lower power state longer.

Product and Performance Information

1

Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.