An aspect of the present invention proposes a solution to allow a dynamic adjustment of a performance level of a GPU based on the user observed screen area. According to one embodiment, a user's focus in one or more display panels is determined. The GPU that performs rendering for that region and/or display panel will dynamically adjust (i.e., increase) the level of performance in response to the user's focus, whereas all other GPUs (e.g., the GPUs that perform rendering for other regions/display panels) will experience a reduced level of performance. According to such an embodiment, dynamically reducing the performance of GPUs outside of the area of focus can result in any one or more of a significant number of benefits, including lower power consumption rates, less processing, less (frequent) memory accesses, and reduced heat and noise levels