Abstract: One embodiment of the invention sets forth a method for toggling between video scanouts generated by a plurality of graphics processing units. The method includes the steps of transmitting a set of programming instructions to a first graphics processing unit and to a second graphics processing unit, configuring a first state machine within the first graphics processing unit to cause a trigger signal to be included with each video frame transmitted by the first graphics processing unit for display, and configuring a second state machine within the second graphics processing unit to cause a trigger signal to be included with each video frame transmitted by the second graphics processing unit for display. The method advantageously creates a direct relationship between the transmission frequencies of the individual graphics processing units and the switching frequency of a video bridge, not relying on a driver to control the video bridge switching.

Abstract: A method for synchronizing two of more graphics processing units. The method includes the steps of determining whether the phase of a first timing signal of a first graphics processing unit and the phase of a second timing signal of a second graphics processing unit are synchronized, and adjusting the frequency of the first timing signal to the frequency of the second timing signal if the first timing signal and the second timing signal are not synchronized.

Abstract: A first graphics application produces surface data using a first graphics applications programming interface. A second graphics application uses a second graphics applications programming interface that is incompatible with the first graphics applications programming interface to process the surface data for display. A device driver requests driver level information from the second graphics application and stores the surface data in a portion of memory specified by the driver level information. Although the first graphics application is incompatible with the second graphics application, surface data produced by the first graphics application may be processed for display by the second graphics application.

Abstract: A method for synchronizing two of more graphics processing units. The method includes the steps of determining whether the phase of a first timing signal of a first graphics processing unit and the phase of a second timing signal of a second graphics processing unit are synchronized, and adjusting the frequency of the first timing signal to the frequency of the second timing signal if the first timing signal and the second timing signal are not synchronized.

Abstract: A method for synchronizing two of more graphics processing units. The method includes the steps of determining whether the phase of a first timing signal of a first graphics processing unit and the phase of a second timing signal of a second graphics processing unit are synchronized, and adjusting the frequency of the first timing signal to the frequency of the second timing signal if the first timing signal and the second timing signal are not synchronized.