User Guide

  • 2020.3
  • 07/10/2020
  • Public Content
Contents

While analyzing your graphics application you might need to measure GPU time to discover how much time your application spends within each region of the captured scene.
To measure GPU time and to normalize all available metrics:
  1. In the
    Scene Overview
    pane
    , select
    Regions Only
    or
    Full Tree
    view from the drop-down list.
  2. Select the GPU Duration metric.
  3. Arrange regions by time in the
    Old
    column
    (if nothing has been changed) or in the
    New
    column
    (if you modified something)
    and see which of the selected regions takes more time.
    NOTE
    You can use the
    D3DPerf_BeginEvent/EndEvent
    API calls to provide meaningful names for each region, such as "terrain pass" or "lighting effects", instead of using the default numbered names.
  4. Select the check box of the most time-consuming region to highlight it in the
    Visualization
    pane
    .
The following screenshot shows that the graphics application spends more time on the "39 to 67" region within the captured scene.

Product and Performance Information

1

Intel's compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice.

Notice revision #20110804