Abstract: A computer-implemented method for parallel detection of disparities in video from multiple customer devices under test in a system unit that comprises a central processing unit and is connected to a graphics card with a graphics processing unit, the method comprising: receiving a video signal from each of the devices under test; processing each of the received video signals in a dedicated CPU thread controlled by the CPU and a dedicated GPU thread controlled by the graphics processing unit; wherein within the CPU thread successive threads are uploaded to a memory for the graphics processing unit and the GPU thread is executed to process the frame, and the processed frames are collected from the memory for the graphics processing unit to check within the CPU thread whether disparities have occurred in the processed frame; wherein all the CPU threads are controlled by a main thread controlled by the CPU.

Abstract: A computer-implemented method for parallel detection of disparities in video from multiple customer devices under test in a system unit that comprises a central processing unit and is connected to a graphics card with a graphics processing unit, the method comprising: receiving a video signal from each of the devices under test; processing each of the received video signals in a dedicated CPU thread controlled by the CPU and a dedicated GPU thread controlled by the graphics processing unit; wherein within the CPU thread successive threads are uploaded to a memory for the graphics processing unit and the GPU thread is executed to process the frame, and the processed frames are collected from the memory for the graphics processing unit to check within the CPU thread whether disparities have occurred in the processed frame; wherein all the CPU threads are controlled by a main thread controlled by the CPU.