Thanks for visiting Imaging and Machine Vision Europe.

You're trying to access an editorial feature that is only available to logged in, registered users of Imaging and Machine Vision Europe. Registering is completely free, so why not sign up with us?

By registering, as well as being able to browse all content on the site without further interruption, you'll also have the option to receive our magazine (multiple times a year) and our email newsletters.

MIC-770 industrial computer and ROM-5620 SMARC 2.1 module

Share this on social media:

Advantech, a leading provider of intelligent systems, is happy to announce its GPU edge computing solution with NVIDIA® T4 GPU NVQual Validation: a MIC-770 compact industrial computer equipped with an MIC-75M20-B expansion i-Module. Compatible with the NVIDIA® T4 GPU Card, the GPU edge computing solution supports up to 2,560 CUDA cores and 16GB GDDR6 GPU memory. With compact size (127 x 192 x 230 mm) and a high-scalable design, plus the longevity of the Tesla T4 GPU card, it is an excellent AIoT solution for deep learning and inference applications such as factory automation, automated optical inspection (AOI), AI prediction in medical equipment, and smart sensing in autonomous vehicles.

Featuring a dedicated server-grade air duct design that facilitates enhanced thermal management of the Tesla T4 GPU, the solution maintains a GPU IC operating temperature below 80 °C/176 °F in environmental conditions higher than 40 °C/104 °F. This design significantly reduces the operating temperature to deliver superior GPU performance and clock frequency. Equipped with PCIe Gen3 x16, the edge solution is certified by NVQual for fast, wide-ranging bandwidth and a reliable data processing solution.

Featuring 127 x 192 x 230 mm dimensions and an Intel® 9th Gen Core™ i CPU, the powerful open-source x86 platform provides fast development for integrating the GPU solution. The innovative modular design for PCIe expansion and flex I/O in the front panel makes it possible for scalable configuration regarding multi-application requests. By adding PCIE-1674 vision frame grabber card, it can connect to 4-ch cameras for autonomous vehicles. Adding 4x GbE Flex I/O (98910770300), the controller connects to LiDAR for sensing. It can detect distance and shape in its surroundings before passing on the data for T4 GPU to build 3D models for graphics analysis.

Key Features

Qualified by NVIDIA® T4 GPU NVQual Validation
Compact (127 x 192 x 230 mm) and ruggedized design
8th/ 9th gen Intel® Core™ processor
Supports PCIe slot expansion for I/O card
Supports Flex I/O expansion
Supports Windows 10 and Linux OS

SMARC2.1 module ROM-5620 powered by NXP i.MX 8X family for automation and industrial control

Advantech has also announced the ROM -5620 SMARC 2.1 module, its first Arm® Cortex® -A35 based on the NXP i.MX8X application processor. The ROM-5620, features an ultra-low-power consumption design and extended operating temperature, ideally suited for automation equipment and HMI devices. The ROM-5620 adopts an automotive grade SoC and related industrial grade components to provide long life cycle support and wide operating temperature from -40°C to 85°C. Equipped with AIM-Linux and WISE-DeviceOn software services, ROM-5620 provides long-term maintenance along with industrial-focused apps and SDK add-ons within the Linux BSP for better time-to-market support. Two GbE Ethe rnet controllers, 2 x CAN bus, software configurable display solutions with dual channel LVDS or MIPI-DSI and a wide range of I/O interfaces like PCIe 3.0, USB 3.0 and 4-Lane MIPI-CSI2 are default supported on our ROM-5620 compact size SMARC 2.0 Computer-On-Module.

The ROM-5620 is characterised by its high efficiency for almost all of the industrial control applications as it provides 40% higher performance and 10% lower power consumption comparing to the Cortex-A7 platform. ROM-5620 is powered by NXP's i.MX 8X SoC, with 2 to 4 core energy-efficient Cortex-A35 processor for mid-range automation and industrial market segments, one Cortex-M4F core for real-time processing, and one Tensilica® Hi-Fi 4 DSP for efficient audio and voice codec execution. For graphic performance, the Vivante GC7000 Lite GPU, provides 4K H.265 capable decoder, and dual 1080P60 display controller.

Following the SMARC 2.0 standard, ROM-5620 leverages high-bandwidth 2GB LPDDR4 memory, 16GB eMMC and dual GbE LAN controllers. It also provides a flexible display solution for 1 x dual channel LVDS or 2 x 4-lane MIPI-DSI through software configuration. I/O expansion choices include USB3.0, 2 x USB2.0, PCIe3.0, SDIO and 3 x UARTs for wireless module connectivity. These expansion choices together with 1 x 4-Lane MIPI-CSI2, 4 x I2C, 2 x SPI, 2 x CAN bus and 12 x GPIO interfaces, cover all the essential requirements for automated control and ruggedized application.

Advantech supports an Allied, Industrial and Modular (AIM) framework for Linux applications that help accelerate software development via flexible, long-term support. AIM-Linux services offer verified embedded OS platforms and industrial-focused apps and SDKs through which users can easily select the embedded software tools they need to focus on their application software development. ROM-5620 is also armed with WISE-DeviceOn for securely automated device onboarding, remote device monitoring and software over-the-air updates.

Key features

  • NXP i.MX 8X DualXPlus/QuadXPlus 1.2GHz Processor with 2 to 4 Cortex-A35 and Cortex-M4F
  • 64-bit LPDDR4 2GB, 16GB eMMC and 8MB QSPI NOR Flash
  • 4Kp30 H.265/H.264 decoder, 1080p30 H.264 encoder capability
  • Ultra-Low-Power Consumption with 4.45W(Max) @5V
  • Dual channel LVDS or 2 x 4-Lane MIPI-DSI by software configuration
  • Dual GbE LAN / PCIe 3.0 / 1 x USB 3.0 and 2 x USB2.0 / 3 x UART / 2 x CAN bus / MIPI-CSI
  • Multi OS support in Yocto2.5 AIM-Linux & Android BSP
  • Operating Temperature: 0~60°C / -40~85°C 

"The ROM-5620 delivers all the performance of the i.MX 8X in a SMARC 2.1 module form factor that will enable fast design-in and rapid adoption with a wide range of customers across the industrial and automotive markets," said Robert Thompson, Director Global i.MX Ecosystem at NXP Semiconductors.

Company: 
Other tags: