Capabilities
Rugged Video Modules
WOLF modules are designed and manufactured specifically for use in the harsh environments encountered in military and aerospace applications. They have been designed to pass MIL-STD-810 and DO-160 environmental tests. They have been manufactured to IPC-A-610 CLASS 3 and IPC 6012 CLASS 3 for high reliability electronic products. They are compliant with IPC J-STD-001 soldering standards.
WOLF designs include optional air cooled or conduction cooled heat plates. Heat is a critical concern when it comes to processing video data, so unlocking the best performance requires the best cooling capability. WOLF’s advanced cooling technology is designed to move heat using a low weight, high efficiency pipeline which moves heat away from the GPU die or FPGAs.
WOLF’s designs follow best practices for ensuring reliability in tough environments. Component derating meets or exceeds NASA and Rome Labs specifications for reliability. Environmental considerations guide our PCB material selection, connectors, and mounting hole placement. Conformal coating options are also available.
WOLF is certified for ISO 9001:2015 quality management system, and compliant with AS5553 counterfeit electronics part avoidance, detection, mitigation, and disposition. WOLF has been AS9100D certified since 2019.
Capture
WOLF designs dedicated capture/conversion boards which can be used to get video data into a system and to convert it to a format that can be used by other GPUs/processors in the system. The boards can also be used to convert video for output to a display device.
At the heart of WOLF’s capture boards is the WOLF Frame Grabber eXtreme (FGX). It supports low latency capture, transmit and video conversion of current and legacy analog and digital interfaces. This flexible technology can be upgraded with WOLF firmware, providing an easy development path to enable new interfaces, increase channel density and add other new features. Many MCOTS solutions can be provided via firmware update only.
Formats supported on current COTS/MCOTS products:
- SD/HD/3G SDI
- 6G and 12G SDI
- CoaXpress CPX-6
- ARINC 818
- DisplayPort
- HDMI
- DVI
- Analog CVBS
- STANAG 3350
- PAL/NTSC
- VGA
Process
Modern GPUs have the processing power to handle a wide range of operations that benefit from parallel processing, such as color correction and enhancement, video stabilization, filtering, terrain analytics, 3D visualization of geospatial data, and tracking. Thanks to the available processing languages/frameworks, such as NVIDIA CUDA™ and OpenCL, a wide range of applications can take advantage of GPU processing.
GPUs can also be used for general purpose data processing, and have found use in fields as diverse as analyzing scientific data for molecule identification to financial analysis of large systems to cryptocurrency mining. This is often referred to as GPGPU processing (where the first GP stands for General Purpose).
When considering a board for its ability to process images or data the most important considerations are the number of floating point operations they can perform per second (FLOPS) as well as the amount and the speed of the memory. WOLF makes a wide range of high performing boards which can be used for video and data processing.
AI Inference
In practical terms, artificial intelligence means giving a machine the ability to analyze data without specifically programing every criteria used to perform that analysis.
The first phase in developing an AI device is called training. The system is normally presented with a large dataset, and a program with a basic model to allow the system to analyze the data and to determine if it has achieved the success criteria. The system analyzes the data and makes modifications to the model used to arrive at the final analysis in order to improve the success rate. The training phase is very processor intensive, and normally takes place in a lab or data center with access to a large number of processors.
Once the system has been trained the model that was developed in the training phase can then be loaded onto another device to analyze new data. Using the model to infer something about new data is known as the inference phase. AI inference generally takes far less processing power than training, so inference models can be found running on smaller, lower power devices that are operating in the field, analyzing real world data.
Devices using AI inference have been seen in self-driving vehicles, speech analysis, natural language translation, chat-bots, document analysis (e.g., legal and medical documents), robotics, and predictive video encoding, to name just a few uses.
Both the training and inference phases normally benefit from parallel processing and matrix processing. Since GPUs are already based on a parallel processing technology, using GPUs for AI was a natural extension of the existing technology. NVIDIA has taken that even further by implementing Tensor Cores in its latest generation of GPUs. Tensor cores are similar to normal graphics processing cores, but they allow mixed precision computations. Machine learning models do not always require the higher precision of dedicated graphics cores, making Tensor cores a more effective use of the available processing power.
WOLF offers a number of NVIDIA GPU-based modules which include Tensor Cores as well as specialized accelerator circuits for deep learning inference, machine vision, audio processing, and video encoding. These modules can also benefit from NVIDIA’s rich set of AI tools and workflows.
WOLF modules which include Tensor Cores include all modules with NVIDIA Turing, Ampere, and Ada GPUs. Newer generations have more advanced Tensor cores which can support more data types, sparsity, and other advanced features.
Encode
Video encoding has also been a standard feature of GPU processors for many years, and have found widespread use for H.264 (AVC) and more recently as H.265 (HEVC) encoders. They are often used for live streaming, and with the recent advances in the GPU technology they have been used to encode high quality, real time, HD and 4K streams that are comparable with the best CPU solutions.
Video encoding can also be accomplished using dedicated devices, especially in cases where the low power use of a dedicated device is more important than the flexibility of a GPU or a CPU based solution.
WOLF encoding solutions include GPU-based, APU-based and dedicated encoder solutions.
Display
WOLF video display boards support multiple outputs from a single module, typically from 3 to 6 outputs per module, with double that capability for 6U VPX modules. Native GPU formats are supported, including DisplayPort, HDMI and DVI. For modern GPUs the board may also include WOLF technology to support analog formats, such as CVBS and VGA. Some modules also include a low power WOLF video conversion FGX, providing support for other output formats, such as SDI, STANAG 3350 and ARINC 818.
Video output modules support different resolutions, with support for up to HD being standard, and UHD, 4K and 5K being supported on the newer modules. Select modules can also support 8K video output. These newer modules support High Dynamic Range (HDR) video and 10-bit color depth.
