Portwell Leverages Intel Core Ultra 200S Series for AI Inference at the Industrial Edge
August 20, 2026
News
Portwell introduced the Intel Core Ultra 200S Series processor powered PCOM-B887, a COM-HPC Client Type Size C module compliant with the PICMG COM-HPC Revision 1.2 specification. The module has an integrated Neural Processing Unit (NPU) and combines CPU, GPU, and dedicated AI acceleration into a single platform for AI inference at the edge.
The PCOM-B887 supports PCIe Gen5 x16, PCIe Gen5 x4, PCIe Gen4, USB4, DDR5 memory up to 192GB, SATA storage, and multiple display interfaces including DDI and eDP enabling engineers to build scalable platforms proficient in handling not only AI inference, but high-resolution graphics and real-time data processing while streamlining future upgrades.
"Industrial applications today require much more than raw computing performance," said Eric Lu, Product Manager at Portwell. "Customers are looking for platforms that combine AI acceleration, high-speed I/O, graphics capability, and long product life within a standardized modular architecture. PCOM-B887 delivers exactly that. By leveraging Intel Core Ultra 200S Series processors together with the COM-HPC ecosystem, customers can shorten development cycles while maintaining the flexibility to adapt to evolving application requirements."
Applications:
- Factory Automation - powerful CPU cores, high-bandwidth memory architecture, and multi-display ability allows responsive HMI control stations, industrial controllers, and intelligent manufacturing equipment requiring consistent real-time computing
- Machine Vision - the Intel Xe-LPG GPU provides high-speed image processing, while hardware-accelerated AV1 encoding and decoding supports low-latency, high-quality video streaming for AI inspection and vision analytics
- Medical Equipment - graphics performance for advanced CT, MRI, and diagnostic imaging systems, the socketed processor architecture supports upgrade flexibility and addresses long product lifecycle requirements
- Digital Signage - support for up to four independent 4K displays or a single 8K display for retail, transportation, and public information systems with support for SR-IOV GPU virtualization enabling multiple independent display content streams from a single computing platform
For more information, visit portwell.com.tw/products/pcom-b887/.
Chad Cox is the Production Editor at Embedded Computing Design. His responsibilities are centered around content creation, writing and editing, and article research and development. Chad covers industry news and events and is known to interact with various industrial leaders via on-premise visits and online interviews. He is responsible for the digital footprint and dissemination of news via social media posts, advertising creation and the production of newsletters including the Embedded Computing Design’s Daily.
He is well versed in many facets of industrial computing including Edge AI, IoT, Processing, Security, Open Source, and more.
Chad graduated from the University of Cincinnati with a B.A. in Cultural and Analytical Literature and holds a master’s in education.
