Graphics XMC minimises latency for mission-critical projects, says Abaco
Leveraging Nvidia Pascal GPU technology, the NVP2102A is a rugged, high bandwidth XMC Graphics and general purpose graphics processor unit (GPGPU) card, introduced by Abaco. It is based on the Nvidia Quadro Pascal (GP107) P2000 GPU architecture and provides up to 2.3TFLOPS of peak performance.
It is intended for high-end graphics capabilities or Cuda support when performing general purpose processing and that require raw video capture and display.
The NVP2102A has four 3G-SDI inputs, which is twice typical number, claims Abaco. There are also two NTSC/PAL video inputs and two audio inputs. Video output is via two 3G-SDI and two DVI or DisplayPort interfaces.
Support for direct video capture to GPU memory reduces latency, minimising glass-to-glass time so that actionable information is delivered at speeds that are closer to real-time, explains Abaco. The input resolution of incoming video is automatically detected and raw video frames are transferred directly to GPU memory (or host memory). In GPU memory, processing such as image analysis, image enhancement, 360-degree video stitching, sensor fusion and target detection takes advantage of the Nvidia GPGPU (Cuda/OpenCL) technology to deliver output faster than would otherwise be possible, Abaco says. The captured data can also be encoded using the GPU-native H.265 (HEVC) or H.264 encoders directly in GPU memory. Windows or Linux drivers and API are available for x86 systems.
“With the NVP2102A, we’re delivering the kind of performance and functionality on an XMC card that, until recently, would have required a full size card – occupying a valuable chassis slot,” said Peter Thompson, vice president, product management at Abaco Systems. “Its architecture enables it to deliver reduced latency – addressing a significant challenge for developers of advanced video and graphics processing applications,” he added.
The NVP2102A is offered in conduction cooled or air cooled variants with rear XMC I/O on Pn6. The rear XMC pin-out is compatible with VPX systems that follow VITA 46.9 x12d+x8d+24s. 3U VPX and PCIe form factors are supported with carrier boards.
The NVP2102A supports Abaco’s AXIS ImageFlex. This is an image processing and visualisation toolkit enabling rapid development of high performance image processing, visualisation and autonomy applications aimed at size, weight and power (SWaP) sensitive platforms. It is focused on high performance GPU processing and graphics with interoperability with other programing paradigms such as OpenGL, OpenCL, CUDA and OpenCV.
