Active Silicon supports real-time GPU processing

  • Published November 2017

We are pleased to announce that all our frame grabbers are compatible with both AMD‘s DirectGMA and NVIDIA’s GPUDirect for Video.

These APIs enable many filter, convolution and matrix-vector-operations to be performed directly on data from a frame grabber using a GPU without the need to be processed by system buffers or by the CPU. This makes data acquisition very fast with very low latency as the GPU memory is made directly accessible to the frame grabber. Modern GPUs are extremely efficient at processing images and graphics, and their parallel structure makes them particularly well suited to uses where large blocks of data need to be processed in parallel.

We’ve published details on our website to help you understand the processes involved when running both Windows and Linux OS, and the setup requirements for each one to help you know what’s needed and how best to get started, including a video demo of GPU processing. Have a look at our GPU solutions page and contact us with your image processing queries.