Developer Central
China  |  India
  • Home
  • Tools & SDKs
  • Resources
  • Community
  • Partners
  • Support
  • Home
  • Tools & SDKs
  • Resources
  • Community
  • Partners
  • Support
  • Home
  • Tools & SDKs
  • Resources
  • Community
  • Partners
  • Support
  • Home
  • Tools & SDKs
  • Resources
  • Community
  • Partners
  • Support

Tools & SDKs

  • Heterogeneous Computing
    • Aparapi
    • APP Kernel Analyzer
    • APP Profiler
      • Getting Started
        • System Requirements
        • Installation
        • Successful Installation
        • Using The Visual Studio Plugin
        • Using the Command Line Interface
        • Known Issues
        • Support
      • User Guide
        • APP Profiler Session Explorer
        • APP Profiler Session
        • APP Profiler Timeline
        • APP Profiler Summary Pages
        • APP Profiler Code Viewer
        • APP Profiler Kernel Occupancy Viewer
        • APP Profiler Settings
        • APP Profiler Session Parameters
        • Description of Configuration Files
        • APP Profiler Kernel Occupancy
        • CLPerfMarkerAMD Library
    • Accelerated Parallel Processing (APP) SDK
      • Downloads
        • Download Archive
      • System Requirements & Driver Compatibility
      • Documentation
      • Bolt C++ Template Library
        • AMD Bolt Sample Code Contest Official Rules
      • Samples & Demos
      • Compute System Recommendations
      • Intro OpenCL Tutorial
      • Open Source Libraries
      • Support
    • Accelerated Parallel Processing Math Libraries (APPML)
    • CodeAnalyst Performance Analyzer
      • CodeAnalyst for Linux®
      • CodeAnalyst for Windows®
      • PMU Extension Driver
      • CodeAnalyst Archive
    • CodeXL
    • gDEBugger
      • gDEBugger User Guide
        • Getting started with gDEBugger
        • Toolbars
        • Views
        • Dialogs
        • How Do I…
        • FAQ
        • Supported Extensions
        • System Requirements & Installation
        • Legal Information
        • Contact Information
  • CPU Development
    • Advanced Platform Management Link (APML)
    • Core Math Library (ACML)
      • Downloads & Resources
        • Building with ACML
        • How to use ACML with different versions of GCC/GFORTRAN
      • Archive Downloads
      • Product Features
      • Redistribution Agreements
    • LibM
      • ACML_MV
      • LibM Functions
    • Open64 SDK
      • Building with ACML
    • SimNow™ Simulator
    • Tools for DMTF DASH
    • x86 Open64 Compiler Suite
      • Developer Guide
      • Apps & Libraries Built by x86 Open64
      • Inactive Files
  • Graphics Development
    • Display Library (ADL) SDK
    • ATI GPU Services (AGS) Library
    • GPU PerfStudio 2
      • API Trace
      • Frame Debugger
      • Frame Profiler
      • Shader Debugger
      • Shader Editor
      • Previous versions
    • GPU ShaderAnalyzer
    • GPUPerfAPI
    • OpenGL ES SDK
    • Quad-Buffer SDK
    • FirePro SDK
      • FirePro News & Events
    • Radeon™ SDK
  • Open Source
    • GCC & GNU Toolchains
    • OpenCL™ Emulator-Debugger
    • Tapper

Home > Tools & SDKs > Graphics Development > GPU PerfStudio 2 > Frame Debugger

Frame Debugger

The Frame Debugger gives you access to the drawcalls within your application and allows you to view their states and resources. It is possible to pause your application on any given frame and analyze the elements that make up the current frame buffer image. The user may scrub through the draw calls to locate the draw call of interest. The Frame Debugger specializes in viewing the active state for any draw call in a frame and has specialized data viewers for image and description based data. Each data viewer is a dockable window that can be placed and resized by the user to make custom layouts that can be saved and reloaded.

What’s new

  • Pipeline Stage View
    • Along the left side of the window is a control that opens a layout for each stage of the graphics pipeline.
    • Each layout gives the user access to the resources used in the currently selected draw call.
    • The icons for each stage indicate if it is in use at the current draw call – blue is active.
    • Clicking a stage will open the associated layout in the Content Pane.
    • Layouts can be edited and saved by the user to create custom layouts.
    • The image below shows a DX11 application with the pixel shader stage selected.



  • Image Resource Thumbnail Viewer
    • The above image shows the new image resource viewer.
    • Thumbnails of images relevant to the selected stage are included by default.
    • The user can edit which images are included in the viewer.
    • Clicking on a thumbnail will open it in the larger image window at the bottom.
    • Double-clicking will open the image in a large modal image viewer.
  • Image based data includes:
    • RenderTargets
    • Textures
    • Swapchain
    • Depth buffer
    • Stencil buffer
    • Input geometry


Screen shot of the texture viewer showing the resource and resource view information. Users can select the mipmap level and index of the cube array to be displayed, zoom and pan around the image.

  • Index and Vertex Buffer layout
    • The Index and Vertex Buffer Window allows you to see information about the geometry being passed into the graphics pipeline.
    • Input Topology – Indicates the way the index buffer is interpreted by the pipeline and how the primitives are assembled.
    • Index Buffer – Shows the values contained within the index buffer. Offsets and counts that are specified by the draw calls will be reflected in the starting index of the table and the number of elements it contains.
    • Vertex Buffer – Shows the values of each component of the input elements.
    • Input Elements – Contains a list of input elements that are specified by the application. Selecting an item in this list will cause its description to be shown in the textbox to the right.



Above is a screen shot of the DX11 Input Assembly layout showing the Index and Vertex buffer contents, topology and input layout information.

Draw Call Slider



Located along the bottom of the window is a bar graph / slider that allows you to select which draw call you are investigating and immediately identify how costly it is compared to other draw calls. The height of the bar represents the relative time that the draw call took to execute relative to the most expensive draw call in the current frame. The vertical orange bar represents the currently selected draw call and the blue region around it shows a magnified view of the draw calls on either side of the selected call. The heights of the bars do not change, but they do become wider. Clicking anywhere on the bar graph area will change the selected draw call. Clicking in the blue zoomed area allows easy selection of the next or previous draw call. You can also click and drag the orange bar to continually change the selected draw call and get a quick understanding of how the application is rendering.

The small blue area to the top right of the bar graph can also be used to smoothly scroll through the frame. Move the orange bar a little to increase / decrease the selected draw call by one; the further the orange bar is moved from the center, the faster the slider moves through the frame.

GPU PerfStudio Main Page

Feedback or questions?

  • Get the hcNewsFlash.

    Your email address:

    No SPAM.
    Easy unsubscribe.

    HSA is going to rock your world.

    Learn more about Heterogeneous System Architecture.

    Got Questions?

    Ask the Developer Forums Community. They’ve got answers.

Tools & SDKs

  • Heterogeneous Computing
    • Aparapi
    • APP Kernel Analyzer
    • APP Profiler
      • Getting Started
        • System Requirements
        • Installation
        • Successful Installation
        • Using The Visual Studio Plugin
        • Using the Command Line Interface
        • Known Issues
        • Support
      • User Guide
        • APP Profiler Session Explorer
        • APP Profiler Session
        • APP Profiler Timeline
        • APP Profiler Summary Pages
        • APP Profiler Code Viewer
        • APP Profiler Kernel Occupancy Viewer
        • APP Profiler Settings
        • APP Profiler Session Parameters
        • Description of Configuration Files
        • APP Profiler Kernel Occupancy
        • CLPerfMarkerAMD Library
    • Accelerated Parallel Processing (APP) SDK
      • Downloads
        • Download Archive
      • System Requirements & Driver Compatibility
      • Documentation
      • Bolt C++ Template Library
        • AMD Bolt Sample Code Contest Official Rules
      • Samples & Demos
      • Compute System Recommendations
      • Intro OpenCL Tutorial
      • Open Source Libraries
      • Support
    • Accelerated Parallel Processing Math Libraries (APPML)
    • CodeAnalyst Performance Analyzer
      • CodeAnalyst for Linux®
      • CodeAnalyst for Windows®
      • PMU Extension Driver
      • CodeAnalyst Archive
    • CodeXL
    • gDEBugger
      • gDEBugger User Guide
        • Getting started with gDEBugger
        • Toolbars
        • Views
        • Dialogs
        • How Do I…
        • FAQ
        • Supported Extensions
        • System Requirements & Installation
        • Legal Information
        • Contact Information
  • CPU Development
    • Advanced Platform Management Link (APML)
    • Core Math Library (ACML)
      • Downloads & Resources
        • Building with ACML
        • How to use ACML with different versions of GCC/GFORTRAN
      • Archive Downloads
      • Product Features
      • Redistribution Agreements
    • LibM
      • ACML_MV
      • LibM Functions
    • Open64 SDK
      • Building with ACML
    • SimNow™ Simulator
    • Tools for DMTF DASH
    • x86 Open64 Compiler Suite
      • Developer Guide
      • Apps & Libraries Built by x86 Open64
      • Inactive Files
  • Graphics Development
    • Display Library (ADL) SDK
    • ATI GPU Services (AGS) Library
    • GPU PerfStudio 2
      • API Trace
      • Frame Debugger
      • Frame Profiler
      • Shader Debugger
      • Shader Editor
      • Previous versions
    • GPU ShaderAnalyzer
    • GPUPerfAPI
    • OpenGL ES SDK
    • Quad-Buffer SDK
    • FirePro SDK
      • FirePro News & Events
    • Radeon™ SDK
  • Open Source
    • GCC & GNU Toolchains
    • OpenCL™ Emulator-Debugger
    • Tapper

©2013 Advanced Micro Devices, Inc. OpenCL and the OpenCL logo are trademarks of Apple, Inc., used with permission by Khronos.

  • Contact Us
  • |
  • Careers
  • |
  • Site Map
  • |
  • Terms and Conditions
  • |
  • Privacy
  • |
  • Trademarks