Identifying Graphics Elements with Complex Geometry

Geometry analysis can help you both debug rendering issues with the complex objects and optimize your application performance. To analyze geometry for the selected API call:

  1. In the Main bar chart or from the API Log, select a single API call that results in a rendering issue, or contributes the most to the overall frame rendering time.

    NOTE

    Graphics Frame Analyzer displays geometry objects only if you select a single API call.

  2. Select the geometry thumbnail in the Resource List. The Geometry Viewer opens, displaying the object in the 3D coordinate system. You can assess the object from different angles by pressing the mouse wheel to switch the axes, or rotate the object by clicking and dragging the object with the mouse pointer.
  3. Analyze various geometry aspects by switching between the available visualization modes:

    Solid (default) - displays the object as a solid model.

    Wireframe - displays a wireframe model to inspect the backside of the object and to understand how it is built from triangles.

    Solid Wireframe - displays a wireframe model on top of the solid model.

    Normals - displays a normal map for your object to analyze how the normal vectors are plotted.

  4. Optionally, export geometry data in the Wavefront OBJ format by clicking the button.

Using geometry analysis, you can identify and optimize inessential elements that take up a lot of resources. For example, if you find an element in the scene that includes a lot of polygons or complicated artwork but contributes very little to the visual effect of the scene, you can gain significant performance improvements by reducing the level of detail for the geometry of this element.

 

See Also

Resources Pane

For more complete information about compiler optimizations, see our Optimization Notice.