Optimizing OpenCL Code with Intel GPA

Optimizing OpenCL Code with Intel GPA

With Intel GPA, a key feature of the product is the Intel GPA Platform Analyzer. The tool collects
real-time trace data during the application run -- this trace data helps you visualize the execution profile of
the various tasks in your code over a period of time on both the CPU and the GPU.

Another key technology now available to developers is the Intel OpenCL SDK. OpenCL provides a uniform programming environment for software developers to write efficient, portable code. Version 1.1 now provides built-in capturing support, which
enables you to see the distribution of an application's OpenCL
commands (kernels and memory operations) across the CPU's software
threads.

Obviously having both of these technologies integrated together is a "must have" tool for game developers. This article shows you how to use these two technologies together to help you design, develop, and optimize your game to take advantage of the entire compute resources available on your customer's system. The end result is that your game is now playable on a wider range of devices, thereby increasing the potential marketplace for your games.

Please let us know your experiences using these two technologies together!

Regards,

Neal

3 posts / 0 nouveau(x)
Dernière contribution
Reportez-vous à notre Notice d'optimisation pour plus d'informations sur les choix et l'optimisation des performances dans les produits logiciels Intel.

Hello Maxim,

Thanks for the pointer to the updated article! Having OpenCL* traces is an important feature of the Platform Analyzer tool.

Regards,

Neal

Laisser un commentaire

Veuillez ouvrir une session pour ajouter un commentaire. Pas encore membre ? Rejoignez-nous dès aujourd’hui