Nvidia's acquisition of Ageia earlier this year was a smart move aimed at bring particle acceleration in its mid-range / high-end graphics cards. Nvidia now plans to make the PhysX API available independently of another acceleration hardware card. As previously reported, the graphics specialist announced that it is about to complete the conversion of Ageia's API to the CUDA development language.

When it finally breaks free, the CUDA drivers would allow GeForce 8 and GeForce 9 cards owners to enjoy PhysX-enabled games without having to add and external physics processing unit to their system. Of course, Nvidia will have its benefits from including physics processing capabilities into its graphics cards.

According to Nvidia's CEO Jen-Hsun Huang, the company expects to sell more units of its mid-range and high-end graphics solutions than before the integration.

"Our expectation is that this is gonna encourage people to buy even better GPUs. It might - and probably will - encourage people to buy a second GPU for their SLI slot," Huang claimed earlier this year. "And for the highest-end gamer, it will encourage them to buy three GPUs. Potentially two for graphics and one for physics, or one for graphics and two for physics," he continued.

More than that, this would be a quick and sharp stab at Intel's upcoming Larrabee multi-core architecture. During this spring's Intel Developer Forum, Intel said that it would release the Larrabee chip through its channel partners, a strategy widely adopted by Nvidia and its rival ATI.

The same manufacturers that thrive on building Nvidia and AMD graphics cards could be convinced to adopt Intel's Larrabee, and the chip giant has all the financial means to push its products forward.

However, Nvidia's sharp response to porting the PhysX API to its CUDA development engine could reflect its worries regarding the upcoming Intel GPUs.