Leading Technology
The Abaco NVP2102A is a rugged XMC graphics and raw video capture card based on the NVIDIA® Pascal ™ (GP107) Quadro® P2000 GPU architecture.
The NVP2102A leverages the high performance of the GPU with its 768 CUDA cores and 4GBytes of GDDR5 memory to deliver 2.3 TeraFLOPs of peak performance with CUDA™ and OpenCL™ support.
High Performance Applications
Target applications for the NVP2102A include those that require very high-end graphics capabilities or CUDA® support when performing general purpose processing, and that require raw video capture and display. It is ideal for GPGPU applications such as radar and video surveillance/analysis, situational awareness and signals intelligence. The NVP2102A makes it an optimum fit for GPGPU applications including machine learning and autonomy as well as offering additional support for legacy interfaces and peripherals.
The NVP2102A leverages the high performance of the GPU with its 768 CUDA cores and 4GBytes of GDDR5 memory to deliver 2.3 TeraFLOPs of peak performance with CUDA™ and OpenCL™ support.
Glass-to-Glass Latency
The NVP2102A supports direct video capture to GPU memory. This significantly reduces latency, minimizing glass-to-glass time – enabling the delivery of actionable information in closer to real time. 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, target detection and so on takes advantage of the extraordinary capabilities of NVIDIA GPGPU (CUDA/OpenCL™) technology to deliver output to the user far faster than would otherwise be possible.
AXIS Software Toolkit
The NVP2102A supports Abaco’s AXIS ImageFlex is an image processing and visualization toolkit enabling the rapid development of high performance image processing, visualization and autonomy applications aimed at size, weight and power (SWaP) sensitive platforms. It is focused on high performance GPU processing and graphics and is interoperable with other programming paradigms such as OpenGL, OpenCL, CUDA and OpenCV.
Benefits
- Lower SWaP COTS video ingest and graphics solution
- Most recent NVIDIA chip-down Pascal technology: do more with less
- Direct video capture to GPU memory and output for minimal glass-to-glass latency
- Significant video ingress and egress capability
- Support for legacy interfaces and peripherals
- Increased frequency and resolution for improved operator displays
- Increased overall system performance in a small form factor
- Potential to lower overall system, spares and maintenance costs
- Reduced system bottle necks for overall increased system performance
- Enhanced flexibility
- Decreased development costs through provision of key board level drivers and integration
- Accelerated image processing and manipulation with Abaco AXIS ImageFlex
Features
- HPC graphics output board
- XMC form factor
- NVIDIA Quadro Pascal P2000 GPU
- 2.3 TFLOPs peak FP32 performance
- 768 CUDA cores
- 4GB GDDR5
- 4x 3G-SDI Inputs
- Direct capture into GPU memory
- 2x CVBS (NTSC/PAL) inputs
- Direct capture into GPU memory
- 2x Audio inputs
- 2x 3G-SDI outputs
- 2x Rear DP 1.2 or 2x rear SL-DVI outputs
- PCI Gen 3 – x4 lanes
- GPU capable of H.264/H.265 (HEVC) encoding/decoding
- Windows and Linux drivers on x86
- NVIDIA CUDA 10, OpenCL 1.2 and NVIDIA software suite
- AXIS ImageFlex support
- Air-cooled and conduction-cooled
Please contact systerra for ordering options
|
