Synopsys and Samsung Foundry have expanded their collaboration to advance the design of next-generation AI, HPC, and multi-die systems on Samsung’s most advanced nodes, including SF2 and SF2P. The companies’ joint efforts are helping customers accelerate tape-outs of complex SoCs and multi-die designs using Synopsys’ AI-driven EDA tools, IP portfolio, and 3DIC Compiler, optimized for Samsung’s latest process technologies.
The latest milestone includes a successful customer tape-out of an HBM3 design on Samsung’s SF2 process and I-CubeS 2.5D packaging, with Synopsys’ 3DIC Compiler reducing turnaround time by 10x and improving eye opening by 6% for enhanced performance. Synopsys’ certified digital and analog flows on SF2P, powered by the Synopsys.ai full-stack EDA suite, are helping customers achieve superior PPA results while accelerating the migration of analog IP to advanced nodes. Additionally, a broad portfolio of Synopsys IP—including 224G, UCIe, PCIe 7.0, MIPI, LPDDR6, and USB4—is now optimized for Samsung’s SF2P, SF4X, and other leading-edge processes.
The expanded collaboration is driving innovation for a wide range of applications, from AI inference engines for data centers to ultra-efficient edge AI devices and advanced HPC platforms. The combined design technology co-optimization (DTCO) and certified IP portfolio enable Samsung Foundry customers to reduce design risk and time-to-market for complex multi-die and AI-driven architectures.
- Synopsys 3DIC Compiler and Samsung I-CubeS tech enable faster multi-die tape-outs
- AI-driven flows certified on SF2P, delivering improved PPA and faster analog IP migration
- Successful HBM3 design on SF2 achieved 10x faster turnaround and 6% better eye opening
- Broad IP portfolio includes 224G, UCIe, PCIe 7.0, MIPI, LPDDR6, USB4, and more
- Collaboration supports advanced AI, HPC, mobile, automotive, and IoT designs
“Together with Samsung Foundry, we’re enabling the most advanced AI processors across a broad spectrum of use cases, from high-performance AI inference engines to ultra-efficient Edge AI devices,” said John Koeter, senior vice president, Synopsys IP Group.