Centaur Announces SoC with New x86 CPU Core and Integrated AI Coprocessor

An x86 SoC with Server-Class CPUs and AI Co-Processor

Austin, Texas based Centaur Technology (which is a subsidiary of VIA Technologies) has announced a new x86 processor that they claim to be “the industry’s first high-performance x86 SoC with server-class CPUs and integrated AI coprocessor technology”.

“Centaur’s microprocessor-design technology includes both a NEW high-performance x86 core AND the industry’s FIRST integrated AI Coprocessor for x86 systems. The x86 microprocessor cores deliver high instructions/clock (IPC) for server-class applications and support the latest x86 extensions such as AVX 512 and new instructions for fast transfer of AI data. The AI Coprocessor is a clean-sheet processor designed to deliver high performance and efficiency on deep-learning applications, freeing up the x86 cores for general-purpose computing. Both the x86 and AI Coprocessor technologies have now been proven in silicon using a new scalable SoC platform with eight x86 cores and an AI Coprocessor able to compute 20 trillion AI operations/sec with 20 terabytes/sec memory bandwidth. This SoC architecture requires less than 195mm2 in TSMC 16nm and provides an extensible platform with 44 PCIe lanes and 4 channels of PC3200 DDR4.”

Of interest in the above description is the reference to AVX512 support (credit Twitter user InstLatX64 for pointing this out), previously an Intel exclusive. (AMD’s Zen 2 architecture supports AVX and AVX2, but not AVX512.) Intel uses AVX512 technology for their “Deep Learning Boost” feature, found in the latest Xeon W-2200 and X-Series HEDT processors.

The full press release from Centaur is available at this link (PDF).

Could this new SoC from a relative unknown provide a legitimate competitor to AMD and Intel in the AI/DL space? We shall see. But any new x86 core is always interesting (and we shall pick Josh’s brain about this on the next podcast).