银河飞将
In certain example embodiments, the “game device” may be a device that is hosted within a cloud-based environment (e.g., on Amazon's AWS or Microsoft's Azure system). In such a scenario, the game (or other application program) may be hosted on a virtual machine in the cloud computer system and the input devices and display devices may be local the user. The user may also have a “thin” client application or computer that is communicating with the cloud-based service (e.g., communicate data from the device and receive and display images that are received from the cloud to the television). In this type of implementation, user input is passed form the user's computer/input device to the cloud-based computer system that is executing the video game application 108. Images are generated by the game engine, transformed by the neural network (e.g., upconverted) and then transmitted to the user's display (or a computer that then outputs the images to the display).
查看详细资料
TOP
In certain example embodiments, the techniques herein may advantageously take advantage of NVIDIA's tensor cores (or other similar hardware). A tensor core may be a hardware unit that multiplies two 16×16 FP16 matrices (or other sized matrices depending on the nature of the hardware), and then adds a third FP16 matrix to the result by using fused multiply—add operations, and obtains an FP16 result. In certain example embodiments, a tensor core (or other processing hardware) can be used to multiply two 16×16 INT8 matrices (or other sized matrices depending on the nature of the hardware), and then add a third INT32 matrix to the result by using fused multiply-add operations and obtain an INT32 result which can then be converted to INT8 by dividing by the appropriate normalization amount (e.g., which may be calculated during a training process, such as described in connection with FIG. 9). Such conversions may be accomplished using, for example, a low processing cost integer right shift. Such hardware acceleration for the processing discussed herein (e.g., in the context the separable block transforms) may be advantageous.
