Quote Originally Posted by hurleybird View Post
No, completely wrong. A PhysX processor needs to communicate with the CPU, not graphics card. There is no intricate connection between the two like Nvidia claims.
Nice, so you know more about how a company's software works than the company itself? Read, learn, then speak. Preferably in that order

The Fermi pipeline is optimized to switch to below 25 microseconds, a significant improvement over last generation GPUs. Besides improved performance, this allows developers to create applications that take greater advantage of frequent kernel-to-kernel communication, such as fine-grained interoperation between graphics and PhysX applications.