Portwell, Inc., a global innovator in industrial and embedded computing solutions, announced the launch of PCOM-B65C, a high-performance COM Express® Type 6 Compact module powered by the NPU-integrated AMD Ryzen™ Embedded 8000 Series processors. Designed in compliance with the PICMG COM Express® R3.1 specification, PCOM-B65C combines robust computing, graphics, and AI acceleration with integrated Out-of-Band (OOB) remote management in a compact 95 x 95 mm form factor. As industrial and embedded applications increasingly demand real-time AI inference, advanced visualization, and long-life platform support, PCOM-B65C provides an ideal platform for edge AI computing, industrial automation, medical imaging systems, machine vision, and digital signage deployments.
At the heart of PCOM-B65C is the AMD Ryzen™ Embedded 8000 Series SoC, featuring up to 8 Zen 4 CPU cores, integrated RDNA™ 3 graphics, and an AMD XDNA™ Neural Processing Unit (NPU) delivering up to 16 TOPS AI performance. The module supports up to 96GB DDR5-5600 ECC/non-ECC memory, PCIe Gen4 expansion, dual SATA interfaces, optional onboard PCIe NVMe storage, and up to four independent displays through DDI, eDP/LVDS, and VGA support. These capabilities enable developers to build high-performance systems for graphics-intensive and AI-driven workloads while optimizing power efficiency across configurable 15W to 54W TDP profiles.
“Industrial systems today require more than just processing performance—they must also handle AI workloads efficiently while remaining easy to manage remotely,” said Austin Tsai, Portwell Product Manager. “PCOM-B65C leverages AMD XDNA™ AI acceleration to deliver a balanced combination of CPU performance, RDNA™ 3 graphics, and efficient AI processing for edge computing applications. In addition, the module integrates Out-of-Band Management, enabling remote power control, system reboot, and BIOS updates while helping customers reduce maintenance costs and improve operational efficiency.”
The combination of CPU, GPU, and NPU resources makes PCOM-B65C particularly well suited for demanding edge applications. In factory automation, the module supports real-time machine control, AI-assisted inspection, and predictive maintenance by processing data locally with low latency. For medical imaging and healthcare systems, its advanced graphics capabilities and embedded longevity support diagnostic imaging, patient monitoring, and clinical visualization applications. In digital signage and information display applications, support for up to four simultaneous displays and high-resolution graphics enables immersive content delivery across retail, transportation, and public information environments.
“The market is rapidly moving toward AI-enabled edge systems that can process data closer to where it is generated,” added Tsai. “By combining efficient AI acceleration, multi-display graphics capability, and industrial-grade reliability in a compact COM Express module, the PCOM-B65C allows system designers to accelerate development while addressing the evolving requirements of automation, healthcare, smart retail, and machine vision applications.”
With its powerful AMD Ryzen™ Embedded 8000 architecture, integrated AI acceleration, advanced graphics capabilities, extensive I/O connectivity, and the Out-of-Band remote management functionality, the PCOM-B65C provides a versatile foundation for next-generation embedded systems. Whether deployed in intelligent manufacturing, medical imaging, edge AI computing, machine vision, or digital signage environments, the module enables OEMs and system integrators to shorten development cycles while delivering scalable and future-ready solutions. Through the launch of the PCOM-B65C, Portwell continues its commitment to empowering customers with reliable, high-performance embedded computing platforms designed to meet the growing demands of AI-driven industrial markets.
Para ofrecer las mejores experiencias, utilizamos tecnologías como cookies para almacenar y/o acceder a la información del dispositivo. Dar su consentimiento a estas tecnologías nos permitirá procesar datos como el comportamiento de navegación o identificadores únicos en este sitio. No dar o retirar el consentimiento puede afectar negativamente ciertas funciones y características.