Introduction

AV200, EBX rugged system is a powerful system that is driven by Intel® 7th generation Kabylake CPU and chipset soldering onboard, integrated with Nvidia GPU GTX 1660S/1050Ti that supports 2 independent DisplayPort. Processor i7-7820EQ plus Intel® QM175 chipset supports clock speed 3.0GHz, up to 3.7GHz. Quad cores, turbo up to 8 cores to cope with enormous data computing. Apart from computing power, the system is able to survive in harsh environment from -40 to 60°C, and ruggedness for vibration and shock is proven to pass MIL-810G standard. With the combination of high CPU computing power and rich display inputs, AV200 is a perfect portable solution for rugged consoles & workstations, transportation, defense control room to share instant and accurate information on huge screen for in-time solutions.

Thermal solution for fanless system design

7starlake designs a unique enclosure that is able to stack together both horizontally and vertically. The aluminum heat sink enclosure allows dual-sided heat dissipation. 7starlake exclusively adopts special heat radiating material and combining with special CNC cutting, further forged into a lavish sophisticated metal. Apart from the special heat sink enclosure, AV200 innovatively adopts two kinds of copper heat spreader. The shape and the size of the spreader are tailor-made based on the heat sources placement of CPU module and graphic module. In view of the gap difference between motherboard and heat sink, 7starlake builds two heat spreaders in different thicknesses. Combining all these exclusive thermal designs can alternatively replace traditional fan, and also ensures high reliability and stability while working under wide range temperature from -40°C to 60°C.

 

Support OpenVINO™

The  AV200 is an AI hardware ready system ideal for deep learning inference computing, AV200 support Intel OpenVINO™(Open Visual Inference & Neural Network Optimization),Intel® architecture-based processor (CPU) and Integrated GPU and deep learning accelerator (FPGA, Movidius ™ VPU) deep learning acceleration chip to enhance vision system functions and performance. It's also runs great with the Intel Neural Compute Stick , and helps developers to create cost-effective and robust computer vision applications. It enables deep learning inference at the edge and supports heterogeneous execution across computer vision accelerators.

 

System main board: EBX SBC-OXY5740A

1. Intel® Core i7 CPU soldering onboard

AV200 is based on EBX SBC—OXY5740A, powered by Intel Kabylake i7-7820EQ quad core processor plus QM175 chipset soldered onboard. With CPU soldering onboard, there is less conduction of heat and high density interconnection between the motherboard and the components, which reduces the MB from the crisis of overheating. Apart from overheating, soldering onboard also provides the best level of shock and vibration protection, removing the unnecessary concern about poor connection that CPU socket type may bring.

2. Two Independent DisplayPort

AV200 has added on NVIDIA graphic card through PCIe/104 interface, supporting 2 independent DisplayPort which is driven by GPU GTX1050Ti. GTX1050Ti itself has 768 CUDA Cores, representing its high computing performance and consumes power at 75W maximum. Connecting by PCIe/104 interface which enhance the resistance of shock and vibration as the connector allows nearly seamless connecting.

3. Wide Range DC input

For military and heavy duty vehicle applications, instable voltage always brings headaches about damaging the electric components. AV200 supports 9V to 36V DC-in, the flexible acceptance of inconsistent power input makes it capable of surviving tough engine cranks and transient over-voltage situations.

Supporting NVIDIA GTX 1660S/1050 Ti 

 

Powered by NVIDIA, CUDA® is a parallel computing platform and programming model developed by NVIDIA for general computing on graphical processing units (GPUs). With CUDA, developers are able to dramatically speed up computing applications by harnessing the power of GPUs. In GPU-accelerated applications, the sequential part of the workload runs on the CPU – which is optimized for single-threaded performance – while the compute intensive portion of the application runs on thousands of GPU cores in parallel.

Related Products

SK1660-NVIDIA GTX1660SUPER MXM 3.1 Graphics Module

• Powered by NVIDIA GeForce GTX 1660 SUPER
• NVIDIA CUDA technology with 1408 Cores
• MXM 3.1 Type B Support DirectX 12, OpenGL 4.6, Vulkan API
• Support for 4 outputs
• 192-bit width, 6GB , GDDR6 High-Speed Memory
• Power Consumption: 120W

Related Products II

SK1050-NVIDIA GTX1050Ti MXM 3.1 Graphics Module

• Powered by NVIDIA GeForce® GTX1050Ti
• MXM 3.1Type-A Module
• High-speed 4GB GDDR5 Memory
• 768 new-gen. Pascal architecture CUDA cores
• Outputs 4 Channel Support DisplayPort 1.2 Certified.
• DisplayPort 1.3/1.4 Ready
• Support NVIDIA CUDATM, OptimusTM, DirectX® 12, OpenGL® 4.5