• ​Leverages power of GPU technology
• Reduces lines of code required by a factor of five
• Includes reference examples, algorithms for even greater developer productivity

Abaco Systems has announced Release 2.0 of its powerful, flexible ImageFlex  image processing and visualisation toolkit at the GPU Technology Conference. Leveraging the enormous power of GPU technology, ImageFlex provides an easy-to-use API framework to considerably speed and simplify the development, optimisation and maintenance of advanced AI applications – especially those targeted at autonomous vehicles.
ImageFlex enables developers of image/video processing and visualisation applications on GPUs to be substantially more productive by hiding the complexity of the underlying software layers, while maintaining high performance. By providing an OpenGL® abstraction layer (no OpenGL experience is required) it can reduce the number of lines of code required by a factor of five, radically reducing the effort and time needed in order to create, test and maintain the application. This means faster time-to-market as well as lower development cost.

New features for ImageFlex Release 2.0 include:

• Tools and reference examples enabling AI-based applications to be deployed on Abaco’s NVIDIA®-based GPU products.
• Provision of a reference target tracking example - a core building block for tracking applications.
• High quality, GPU-optimised image stabilisation.

ImageFlex is highly complementary to Abaco’s NVIDIA GPU-based GVC1000 and GVC2000 hardware platforms, which use the NVIDIA Jetson supercomputer on a module for AI computing at the edge. This allows the creation of complete solutions for Degraded Visual Environment (DVE), 360° situational awareness, helmet mount sight processing, target identification and tracking and other EO/IR processing applications. It is portable across a range of graphics processing architectures and operating systems, and is potentially safety certifiable. ​

• Reduces slot count by one fourth for comparable performance
• Very high density with eight ADC (4GSPS) and eight DAC (6.4GSPS) synchronised channels
• FPGA fabric and ARM CPU tightly connected to RF front end
• Adds to Abaco’s extensive 3U VPX ecosystem

Abaco Systems announced the VP430 Direct RF Processing System, the first 3U VPX COTS solution to feature the all new Xilinx® ZU27DR RF system-on-chip (RFSoC) technology. It is also one of the densest analog FPGA DSP boards available, with eight ADC and DAC synchronised channels, and features the ability to synchronise multiple boards for even larger system applications.

When combined with Abaco’s extensive portfolio of leading edge 3U VPX multi-architecture processing boards including single board computers and graphics processor units, the VP430 enables complete systems to be built from high performance, interoperable COTS (commercial off-the-shelf) components.
Designed for advanced electronic warfare applications including MIMO (multiple input/output), beamforming, sensor processing and radar signal processing, the VP430 is a major step forward in performance and density, enabling the use of fewer boards and much less power while delivering increased processing throughput. In fact, for comparable capability, it occupies one fourth of the slots that earlier generation solutions would require.
The VP430 includes the option to be built with a FireFly™ 8-channel VITA 66.4 fiber optic interface for transfers of greater than 12 GBps for applications in which the throughput of the native VPX PCI Express® Gen3 data plane is inadequate or when a remote client requires streaming data.​

The SA4-COUNTRY is a CompactPCI® Serial carrier for a low profile PCI Express® Card. The board is provided with a PCI Express® x8 connector and accommodates a PCI Express® card with maximum dimensions of up to 176mm (length) x 68.9mm (height). The card I/O bracket is covered by a metallic hood at the SA4-COUNTRY front panel.
The SA4-COUNTRY is equipped with backplane connectors for support of up to eight PCI Express® lanes when inserted into a CompactPCI® Serial fat pipe slot. For optimum signal integrity, PCI Express® Gen3 redrivers are provided on-board.
The SA4-COUNTRY is useful for system integration, e.g. rapid prototyping and special function applications.
More Information.

The new Type E37 encoder boasts a unique combination of ruggedness and a wide range of options while providing an excellent indexing feel. It is a high-quality, cost-effective dual concentric encoder available with or without push-button. 

• Extended operating temperature range: -40 to +85°C
• MIL-STD-202G vibration test compliant
• Gold plated contacts
• Rotational life: Up to 1,000,000 revolutions
• Standard resolution 16 or 32 detents plus excellent indexing feel with 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5 or 4.5 NCM switching torque 
• Swiss Click Indexing System provides positive tactile feedback, secure activation and consistent torque over the lifespan
• Space saving integrated push button option
• Optional IP68 front panel sealing for water tight performance and resistance to dust
• Various options and customisation

More information.

In addition to the proven CAN-PMC/FD, Janz Tec is expanding its portfolio to include the new CAN-PCIe/FD, which is available with up to 4 CAN channels. Supported operating systems include various Linux distributions as well as Windows 7 and 10.
 The CAN-PCIe/FD is a CAN interface for the PCI Express socket. It makes it possible to connect embedded systems to as many as four CAN FD networks.
A feature which sets the CAN-PCIe/FD apart is its guaranteed data transfer rate of up to 8 MBit/s, which makes it possible, for example, to distribute updates to control devices more quickly. With an expanded operating temperature range from 0°C to +70°C, the board is optimised for the industrial environment and can be used in a broad range of applications.
 The board has two independent CAN-FD channels, which can be supplemented by up to two additional CAN-FD-expandable interfaces via transceiver modules - along with guaranteeing a higher bandwidth for the transfer of data than standard CAN.

More information.

• Eliminates need for CPU to ‘throttle back’, maintaining optimum performance
• Delivers greater performance consistency for applications requiring real time determinism

Abaco Systems announced a significant breakthrough in delivering optimum performance from a 3U VPX single board computer (SBC). By implementing an innovative cooling strategy, the 12-core Intel® Xeon® D-1500 processor featured on the SBC347D can now operate at 100% of its capability at temperatures up to 75°C in a cold wall (conduction cooled) environment.
Traditional solutions that are widely implemented on other SBCs see the CPU being throttled by up to 50% of its core frequency, causing a substantial deterioration in performance. This throttling is in order to maintain the processor within its rated temperature range to avoid failure.
“The advantage this unique cooling architecture gives users of the   is not only the higher performance that is required by many advanced applications, but also the repeatability and consistency of that performance almost regardless of operating temperature,” said Richard Kirk, Product Manager, Abaco Systems.  “The SBC347D will operate at its maximum frequency at all times - not just when it’s cool enough to do so. That can be highly advantageous in, for example, applications that require real time determinism.”
Abaco expects that the new cooling architecture will be progressively implemented across its SBC product lines.

• Designed for mission critical defence electronic warfare applications
• Delivers increased bandwidth, performance at lower power, smaller size
• Provides simple, cost-effective upgrade for existing users

Abaco Systems announced the VP889 high performance FPGA processing board, which features Xilinx®’s latest Ultrascale+™ device, together with Zynq® Ultrascale+ technology for advanced security. In line with Abaco’s commitment to provide customers with simple, cost-effective upgradability/technology insertion that enables them to reach new levels of performance, the VP889 delivers a form, fit and function upgrade for the VP881.
The VP889 is designed for the most demanding, mission critical defence applications. These include electronic warfare/ DRFM, radar/sonar processing, satellite communications systems, multi-channel digital transmission/reception and advanced digital beamforming.
Next generation signal processing systems require more bandwidth and more processing capability at lower power and smaller size. The VP889 helps customers meet these demanding requirements on an open COTS platform. With FMC+, customers have up to 300 Gbits/second digital serial bandwidth to and from modular I/O devices. The VP889’s combination of high performance processing and high density I/O delivers the flexibility to leverage Abaco’s market-leading FMC and FMC+ (VITA 57) product portfolio.

he Vision4ce CHARM-100 is a standalone video processing board for embedded video and image processing applications which is designed to host the Vision4ce VMS software. The VMS software includes the field proven DART video tracking software. The CHARM-100 is based on a NVIDIA Jetson embedded processor which incorporates a multicore ARM processor and powerful GPU. Video interfaces are provided for HD and SD video in both analog and digital formats.
 
The CHARM 100 processes two concurrent streams of standard definition or high definition video selected from the multiple analog and digital video inputs. The output video is genlocked to the input video and is comprised of low latency video from the selected input combined with processed video and graphics from the Jetson.
A rugged enclosure is available for the CHARM-100. The enclosure includes an input power supply which allows the CHARM-100 to be powered directly from vehicle and aircraft power sources.

The FMC134 provides customers with a high bandwidth and high channel count I/O solution that is ideal for multi-channel receivers in a wide range of applications.
With four 12-bit analog-to-digital channels up to 3.2GSPS or two analog channels up to 6.4GSPS, the FMC134’s design is based on the Texas Instruments™ ADC12DJ3200 and is mechanically and electrically compliant with the FMC+ standard (VITA 57.4), the next generation of small form factor expansion I/O for high performance FPGA-based systems.
The A/D JESD204B data interface connects to an FPGA using 16 of the 24 high speed serial lanes available on an FMC+ high serial pin count (HSPC) connector. The unused eight high speed serial lanes are passed to the top of the board, enabling the potential for FMC stacking.
The sample clock can be supplied externally through a coax connection or supplied by an internal clock source (optionally locked to an external reference) for multi-board and multi-channel synchronisation.
The board also includes an external trigger input and two reference outputs that can be enabled for multi-board synchronization.
The front panel I/O can optionally be populated with MMCX or SSMC coaxial connectors. Analog I/O is AC-coupled.
When paired with the latest FPGA carrier cards with FMC+ sites - such as the Abaco 3U VP880 or the 6U VP868 with Xilinx® Ultrascale™ technology - customers can implement high performance algorithms on an industry standard platform.

More information on the FMC134 can be found here.
Other FMC Cards can be found here.
More information on Abaco's FPGA cards can be found here.

​Designed with a scalable architecture, the TuffServ640 (TS 640) recorder / reproducer supports a range of applications, from vehicle and weapons testing to intelligence gathering.
Designed for Speed: The TuffServ 640 delivers well over 1000 Megabytes per second of write throughput for hours, with read performance that saturates a 10 gigabit Ethernet link. That’s the performance delivered over NFS, after all the network and protocol overhead is accommodated.
Designed for Capacity: Writing data at over 1000MB/s requires huge amounts of storage: nearly 4 terabytes per hour, every hour. The TS 640 has a high capacity Removable Storage Array (RSA-2) that comfortably supports up to 32TB of storage now, with an option for 80TB. Each RSA-2 contains two storage controllers and up to 16 solid-state devices; 256 bit AES encryption is available on request.
Designed For Growth: Everything in the TS 640 has been designed to provide a flexible platform for current and future requirements. The interface between the system components including the RSA-2 storage module runs at a massive 8000MB/s, so as the storage price/performance curve shifts, so can the TS 640’s. XMC and two miniPCIe sites provide opportunities to expand and adapt the platform for new or unique requirements, as well as provide options such as MIL-STD-1553 control and GPS time synchronisation.
Designed For Control: In addition to the dual standard 10 gigabit/s optical Ethernet interfaces, the TS 640 offers a standard gigabit Ethernet connection for lower bandwidth tasks, and a 100Base-T Ethernet for control and monitoring. Serial ports are also available, as are GPIO signals for embedded applications, such as unmanned vehicles and pods.
Designed For Use: The TS 640 is a truly open system, running mainstream COTS software, including the AMPEX TuffServ recording application; it can also capture raw 10 gigabit Ethernet at line rate. This architecture delivers a solution that is accessible to third-party developers, so that application-specific software can be added to the system by integrators and/or end users.
Information assurance and cyber threat prevention strategies are available on the TuffServ platform, facilitating its acceptance in critical or sensitive environments.

Find more information here.