EIZO Rugged Releases 6-channel XMC Graphics/GPGPU Card

EIZO Rugged has just announced a game-changing, all-in-one solution with the ability to capture, process, encode, decode, display, and stream video and audio data, all with exceptionally low latency.

Rugged and packed with features, it's a thing of beauty - our team is excited to talk to you about it...

Based on the NVIDIA® Quadro® P2000 GPU with NVIDIA Pascal™ architecture, the Condor NVP2102AxX card also efficiently supports CUDA and OpenCL based GPGPU computing, AI processing (artificial intelligence), deep learning and H.265/H.264 encoding/decoding, with all of the processing done in hardware.

High-speed data transfer and exceptionally low latency levels are achieved with the Condor NVP2102AxX graphics processor by optimising NVIDIA’s GPUDirect RDMA (remote direct memory access) feature. This enables the captured video data to be sent over PCIe, bypassing system memory, directly to GPU memory for analysis/processing/display, almost in real time.

In GPU memory, the applications can do processing such as image analysis, image enhancement, 360-degree video stitching, sensor fusion, and target detection using the GPGPU (CUDA/OpenCL™) technology. The card supports multiple outputs including 3G-SDI, DisplayPort and DVI.

With air or conduction cooled options available, as well as a number of input/output configurations please contact one of our engineers to discuss the correct configuration for your application.

Accelerate your 5G and NFV Workloads with Intel PAC and Netcope P4

Did you know, Netcope P4 supports Intel® FPGA PAC N3000?

Netcope P4 compiles the packet processing program written in open-standard P4 language into 100 Gbps FPGA firmware, allowing unprecedented flexibility and rapid development of high-speed network applications.

Netcope P4 Flyer 

It offers a revolutionary way for programmable network adapters to be easily programmed using P4 language without expert knowledge of hardware architecture, which can significantly save money when hiring dedicated firmware designers. It also shortens time-to-market, avoiding lengthy firmware development cycles.

Using the Arria 10 FPGA, now in mass production, Intel® FPGA PAC N3000 is the platform of choice for the acceleration of many bandwidth and latency-sensitive virtualised network applications, such as vRAN, vEPC, vBNG and others. Netcope P4 brings the possibilities for this card to a whole new level. Instead of tedious and error-prone HDL coding, the card can be programmed in the open-standard P4 language.

Netcope P4 is now specifically optimised to take advantage of the N3000 unique features, offering large capacities of lookups in off-chip DDR memories and achieving line-rate packet processing with low latency and resource utilization. With Netcope P4 and N3000, your fully custom 100 Gbps packet processor built with COTS hardware is just hours away.

Sarsen Technology is Netcope's UK partner. For more information on P4 or FPGA acceleration please contact us.

Watch Video - Intel® FPGA PAC N3000

We Take a Look at the Xilinx RFSoC

Way back in October 2017 Xilinx announced the delivery of their Zynq® UltraScale+™ RFSoC family. We've recently seen an upturn of enquiries about these devices, and with the Early Access/ silicon samples phase out of the way customers are very excited about the potential this technology can offer.

Photo: Xilinx.com

To create the RFSoCs, Xilinx built the analog directly into the popular MPSoC Zynq device, in the form of  high performance ADCs and DACs that feed directly into the Zynq interconnect fabric.

This means that the physical footprint and chip-to-chip latency is greatly reduced, and it also completely removes the external digital interfaces between the FPGA itself and the mixed-signal convertors.

Putting this RF quality analog function on an MPSoC opens up the playing field for a wide range of  commercial applications including test & measurement and high performance RF applications, with 5G wireless deployment right at the top of the list.

As SWaP is such a critical set of factors for the majority of defence applications military customers also benefit because the Zynq RFSoCs provide a combination of power efficiency, hardware flexibility and a reduction in required space, making them ideal for radar, early warning detection systems and man-portable communications equipment.

Based on 16nm UltraScale+ MPSoC architecture with an ARM Cortex-A53 processing subsystem and the highest signal processing bandwidth in a Zynq UltraScale+ device, the All Programmable RFSoCs monolithically integrate RF data converters for up to 50-75 percent system power and footprint reduction, and soft-decision Forward Error Correction (SD-FEC) cores to meet 5G and DOCSIS 3.1 standards.

There are now a number of proven Zynq UltraScale+ RFSoC solutions available on the market. Please get in touch if you'd like to discuss an application with one of our engineers - info@sarsen.net or +44 1672 511166.

www.sarsen.net

BittWare partners with Tamba Networks to provide ethernet MAC+PCS IP cores

BittWare, the provider of Altera-based FPGA COTS boards, has entered into a Value-Added Reseller (VAR) agreement with Tamba Networks, a developer of low-latency, high-speed Intellectual Property (IP) cores for networking and financial applications.

The Tamba Networks 1 to 400 Gigabit Universal Ethernet IP Core is a low latency and gate count solution which enables a 10 GigE "FIFO+MAC+PCS" of approximately 25 ns in an Altera Stratix V FPGA. This decreases the round-trip delay wire-to-wire. By narrowing their focus to the ethernet MAC and PCS - designing their own soft PCS core as opposed to the typical solution which uses the hard PCS core - Tamba Networks has optimised the design of both, bringing an extremely efficient solution to market and saving many nanoseconds of latency in the process.

“Many of our customers are looking for the lowest possible latency,” stated Ron Huizen, Vice President of Systems & Solutions at BittWare. “Therefore, we are always seeking out industry-leading technology that can help. Tamba’s MAC+PCS IP cores enable us to provide our customers with unprecedented wire-to-application ethernet latency.”

The Tamba Networks IP cores are available today on all of BittWare’s PCIe-based Altera Stratix V family of boards which include the following:
S5PE-DS Dual Altera Stratix V GX/GS PCIe Board with Quad QSFP+, DDR3, QDRII+, and RLDRAM3
S5PH-Q Altera Stratix V GX/GS Half-Length PCIe Board with Dual QSFP+/SFP+, DDR3, and QDRII+
S5PE-F Altera Stratix V GX/GS PCI Board with VITA 57 FMC Site
S5PE Altera Stratix V GX/GS PCI Board with Dual QSFP+

For more information please contact Sarsen Technology on +44 1672 511166 or email the team - info@sarsen.net