Get Started

  • 2021.3
  • 09/23/2021
  • Public Content

Get Started with Intel® Graphics Performance Analyzers (Intel® GPA)

Use this documentation to get started with Intel® Graphics Performance Analyzers (Intel® GPA), which is a toolset for graphics performance analysis and optimization of games and other graphics-intensive Microsoft* DirectX*, Apple* Metal*, Vulkan*, and OpenGL* applications. Intel® GPA is available on Windows*, macOS*, and Ubuntu* hosts.

Supported Graphics APIs

API
Windows Host
Ubuntu Host
macOS Host
DirectX
yes
no
no
Vulkan
yes
yes
no
Metal
no
no
yes
OpenGL
yes*
yes
yes
*on Windows OS, OpenGL support is limited to Trace mode.
For details on software and hardware requirements for Intel® GPA, see the product Release Notes.

System Analysis: Understand High-Level Performance Profile of Your Game

  • View CPU, GPU, and Graphics API metrics in real time to determine whether your application is CPU or GPU bound
  • Experiment with graphics pipeline state overrides to perform a high-level iterative analysis of your game without changing a single line of code
  • Capture frames and traces for detailed analysis with Graphics Frame Analyzer and Graphics Trace Analyzer, respectively

Stream Analysis: Spot Frames with Potential Performance Bottlenecks

  • Quickly pinpoint problem areas with fast, efficient iterations over stream data
  • Spot intermittent anomalies and multiframe algorithms
  • No need to recapture data to look at different frames
Intel GPA: Stream Analysis

Frame Analysis: Pinpoint Performance Bottlenecks within a Frame

  • Explore a variety of metrics at the API call level to find performance bottlenecks
  • Analyze all graphics resources and textures used in the frame
  • Experiment with rendering states, graphics primitive parameters, as well as shaders without recompiling your game code
  • Analyze pixel history for your application
  • Correlate rendering issues with the exact stage of the graphics pipeline

Platform Analysis: Visualize the Interaction of Your App and Its Threads Across All CPUs and the GPU

  • Correlate CPU and GPU activity to understand whether your app is effectively using all compute resources, or it is CPU/GPU bound
  • Explore GPU usage and analyze a software queue for GPU engines at each moment of time
  • Identify GPU and CPU application frame rate and how it depends on vertical synchronization
  • Explore the performance of your application per selected GPU metrics over time
Intel GPA: Platform Analysis

Select Your Host System to Get Started

Click to learn more about system-specific workflows for Windows*, Linux* (Ubuntu*), or macOS*:
Click for Getting Started steps on Windows* host 
				Click for Getting Started steps on Linux* host 
				Click for Getting Started steps on macOS* host

Product and Performance Information

1

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