NVIDIA MXM compatible motherboard for AIoT applications

IBASE Technology has announced a new NVIDIA MXM compatible motherboard designed for AIoT applications. The MT800M-P supports the latest 8th Generation Intel Core processor family, enabling customers to benefit from optimised processing and graphics performance and enhance I/O capability.

With the increasing demand of AI-enabled inference platforms, IBASE has developed the MT800M-P with an MXM socket for NVIDIA’s graphics cards (up to 190w) with GPU acceleration to harness its cutting-edge performance and responsiveness to power currently prevalent AI services such as speech recognition, image analysis, visual search and media processing.

“Combined with the latest MXM 3.0/3.1 GPU modules, the MT800M-P leverages the advantages of the NVIDIA’s GPU technology to deliver parallel processing performance with unmatched power efficiency,” said Archer Chien, Director of IBASE Solution Product Planning Department. “The MT800M-P is suitable for artificial intelligence applications in the retail, banking and transportation industries.”

In addition to four display interfaces and 32GB system memory support, MT800M-P offers rich network connectivity with four Gigabit Ethernet and high speed I/O peripheral expandability through its six USB 3.1, two USB 2.0, two SATA III, and six serial ports. Expansion is provided by a fast PCIe x16, a Mini PCIe and an M.2 (M-key) slot.

The board measures 270 by 220mm and has 24V power input. It runs on both Windows 10 and Linux Ubuntu operating systems.

MT800M-P features:

  • 8th Gen Intel Core i7/i5/i3 processors
  • Two DDR4 SO-DIMM, Max. 32GB
  • Four Intel Gigabit LAN
  • Two DisplayPort, one DVI-D, one VGA
  • Six USB 3.1, two USB 2.0, two SATA III, six COM
  • One PCI-Ex16 (three PCI-E four signal, optional riser card), one Mini PCI-E socket, one M.2 (M-key)
  • One MXM socket supports NVIDIA MXM GPU card (Up to 190W)

Keep Up to Date with the Most Important News

By pressing the Subscribe button, you confirm that you have read and are agreeing to our Privacy Policy and Terms of Use
Previous Post

Fully integrated power management IC saves board space

Next Post

Support for standalone C++ and mixed-language embedded software