AMD Low-Power Guru Addresses the Looming Electronics Power Challenge

Moore's Lobby · 25 Oct 2022 · 1h 13m · Episode 4

What this episode covers

Sam Naffziger, AMD's Senior Vice President and Corporate Fellow, details his 30-year career in processor design, explaining how he pioneered low-power techniques like adaptive clocking and boost modes at HP and Intel before bringing them to AMD. He discusses the transition from monolithic SoCs to chiplet architectures, highlighting the Rome server processor and 3D V-Cache as key innovations that address the looming power challenge in data centers and consumer devices. Naffziger emphasizes that future performance gains require a multidisciplinary approach integrating software, packaging, and silicon design to achieve AMD's '30 by 25' efficiency goal.

Other shows on the same ground, closest first.

  • EE Times Current · 41m

    Interview: AMD CTO Papermaster | Interview: Imagination CEO Black | Where to Invest in 2020

    AMD CTO Mark Papermaster discusses the company's transition to 7nm Zen architecture, chiplet-based designs like the Threadripper 3990X, and the integration of specialized accelerators for AI workloads. Imagination Technologies CEO Ron Black explains the company's strategic pivot toward automotive graphics, ray tracing, and RISC-V cores, highlighting the resumption of its partnership with Apple. EE Times editors analyze 2020 investment trends, emphasizing that data generation from apps is driving semiconductor growth, with specific focus on memory, RF front-ends, and power infrastructure in developing economies.

  • EE Times Current · 40m

    An All Inclusive Journey to HPC/Data Center Silicon Success

    Charlie Matar, Senior Vice President of Systems Solutions at Synopsys, discusses the escalating complexity of High Performance Computing (HPC) and data center silicon design. He highlights critical challenges including engineer shortages, the high cost of advanced nodes, and the shift toward chiplet architectures and 2.5D/3D packaging. Matar also explains how Synopsys is leveraging its DSO.AI platform to optimize power, performance, and area (PPA) while addressing reliability and security concerns in the semiconductor ecosystem.

  • The Circuit · 33m

    Episode 38: The Future of Transistor Design and Advanced Packaging

    Huiming Bu from IBM Research discusses the future of transistor design and advanced packaging with host Ben Bajarin. They explore the industry's transition from FinFET to nanosheet architectures and the emergence of 3D scaling via Vertical Transport FETs (VTFET) and Stacked FETs (SFET). Bu explains how chiplet technology and heterogeneous integration are accelerating compute density, allowing AMD's MI300X to reach 100 billion transistors seven years ahead of traditional scaling predictions. The conversation highlights the critical need for System Technology Co-optimization (STCO) and new EDA tools to manage the complexity of 2.5D and 3D IC designs.

  • The Circuit · 42m

    Episode 67: A Conversation with Aart de Geus on EDA and the Future of Silicon Design

    Aart de Geus, co-founder and executive chair of Synopsys, discusses the evolution of electronic design automation (EDA) and the future of silicon design. He explains how the industry shifted from analog to digital design, enabling the automation that drove Moore's Law, and details the transition to the Angstrom era with technologies like FinFETs and backside power delivery. De Geus also highlights the rise of systemic complexity through chiplets and vertical stacking, and the critical role of AI in optimizing these complex multi-die architectures.

  • EE Times Current · 37m

    Interview: Ampere CEO Renee James | Arm-ing Data Centers

    Renee James, former president of Intel and CEO of Ampere, discusses how Ampere is challenging Intel's dominance in the data center server market with its Arm-based Altra processors. She explains that while Intel dominated for decades, the shift toward cloud-native workloads, containers, and edge computing has created an opening for power-efficient, high-core-count chips. Analyst Kevin Krewell adds context on the Arm ecosystem's growth, noting that while Arm has long targeted servers, the convergence of mobile developer talent and improved process nodes (like TSMC's) has made Arm viable for high-performance computing.

The summary, chapters and topics are generated from the episode’s audio by this directory, not written by Moore's Lobby. The episode itself belongs to its publisher.