Connected vs Offline Analysis in PVRTune

PVRTune supports two main approaches when analysing application performance on a device

There are two main approaches to analysing an application in PVRTune:

  • connected analysis - which displays real-time data

  • offline analysis - which displays saved data from a PVRTUNE file.

How to perform a connected analysis

Connected analysis is the assessment of real-time data from a target device.

The main steps to performing a connected analysis are as follows:

  1. Ensure that the target device is using PowerVR device drivers.
  2. Boot the target device.
  3. Install PVRPerfServer on the target device.
  4. If necessary, initialise the PowerVR device drivers.
  5. Run PVRPerfServer. If successful, PVRPerfServer outputs the server name, IP address, and port number on the console.
  6. Run PVRTune GUI on the host machine.
    Important: Both PVRTune and PVRPerfServer must have matching versions.
  7. If PVRPerfServer is on the same subnet as PVRTune, the target device will be broadcasting and therefore visible in the PVRTune GUI (see the Connecting to a Target Device in PVRTune - Remote Deployment section). If this is not the case, then manually enter the IP address of the server.
  8. Once successfully connected, identify the bottleneck in the application by analysing the output of PVRTune (for more information, see the Identifying Bottlenecks section).
  9. Attempt to resolve the bottleneck. For more information on optimisation techniques, see our document “PowerVR Performance Recommendations”.
  10. Repeat the necessary steps with the newly optimised application until performance is at the required level, or no further bottlenecks can be identified.

Offline Analysis

It is possible to analyse an application without being directly connected to PVRPerfServer. This requires a PVRTUNE file of the target application to have been created prior to analysis. This approach is particularly useful if data is being lost due to network/Wi-Fi/USB performance or high CPU usage on the client machine.

A PVRTUNE file can be created either through the use of the –sendto= command-line parameter for PVRPerfServer, or by saving received data to a file using PVRTune GUI.

Note: PVRPerfServer only saves counter data for a single counter group when using the –sendto= parameter. By default, the group is set to zero, but this can be changed using the –group= command-line parameter.

With a PVRTUNE file created and copied across to an accessible location, perform the following steps:

  1. Run PVRTune GUI on the host machine. It is important that both PVRTune and PVRPerfServer have matching versions.
  2. Open the file in the PVRTune GUI. The PVRTune GUI will then display the tuning data.
  3. Identify the bottleneck.
  4. Attempt to resolve the bottleneck. For more information on optimisation techniques, see the document entitled “PowerVR Performance Recommendations”.
  5. Repeat the necessary steps with the newly optimised application until performance is at the required level or no further bottlenecks can be identified.