Advanced Micro Devices, Inc. (AMD) is a semiconductor company headquartered in Santa Clara, California, that develops CPUs, GPUs, and FPGAs. Incorporated on May 1, 1969, the company serves data center, gaming, and embedded markets, having surpassed Intel in market capitalization for the first time in 2022, according to Wikipedia.
Diversified Semiconductor Segments
AMD operates through three primary business segments: Data Center, Client and Gaming, and Embedded, according to Yahoo Finance. The company produces a variety of hardware, including artificial intelligence (AI) accelerators, microprocessors, and graphics processing units (GPUs), which it sells as standalone devices or integrates into chipsets, accelerated processing units, and professional GPUs.
The company’s processor portfolio includes several distinct brands. For consumers and professionals, AMD provides the Ryzen, Ryzen AI, Ryzen PRO, Ryzen Threadripper, Ryzen Threadripper PRO, Athlon, and PRO A-Series brands, according to Yahoo Finance. In the server market, the company offers microprocessors under the EPYC brand.
In August 2024, AMD released the Ryzen 9000 series based on the Zen 5 architecture, which the company claims provides a 16% average increase in instructions per clock (IPC) over the Zen 4 generation. For the mobile market, the Ryzen AI 300 series features an integrated Neural Processing Unit (NPU) delivering up to 50 TOPS (trillions of operations per second), a specification designed to meet Microsoft’s hardware requirements for Copilot+ PCs.
Graphics capabilities are divided between the Radeon graphics and Embedded Radeon graphics for general use, and the Radeon Pro graphics brand for professional applications. For hyperscale providers, AMD provides AI and general-purpose compute infrastructure, including data center graphics under the Instinct accelerators and Radeon PRO V-series brands, according to Yahoo Finance.
The Instinct MI300X accelerator, launched in late 2023, features 192GB of HBM3 memory and is positioned as a competitor to Nvidia’s H100 GPU for large language model (LLM) inference. CEO Lisa Su has stated that the MI300 series is designed to provide higher memory capacity and bandwidth to handle larger AI models on fewer GPUs compared to previous generations.
Strategic Pivot to Fabless Manufacturing
AMD’s operational model underwent a fundamental shift in 2009. According to Wikipedia, the company outsourced its manufacturing after spinning off GlobalFoundries. This transition moved the company away from its initial model of manufacturing its own processors.
The company currently utilizes Taiwan Semiconductor Manufacturing Company (TSMC) as its primary foundry partner, employing 5nm and 4nm process nodes for its latest Ryzen and EPYC products. This fabless model is supported by a “chiplet” architecture, which uses the Infinity Fabric interconnect to link separate compute dies (CCDs) and I/O dies (IODs), a method that improves yields compared to traditional monolithic chip designs.
This structural change coincided with a period of volatility. Wikipedia reports that the company struggled to compete with Intel during the late 2000s and early 2010s. However, the company regained market share in the late 2010s by implementing a penetration pricing strategy for its Ryzen processors. This approach allowed AMD to offer lower costs and competitive performance compared to Intel microprocessors.
The Xilinx Acquisition and FPGA Integration
In 2022, AMD expanded its product capabilities through the acquisition of Xilinx, according to Wikipedia. This move allowed the company to offer field-programmable gate array (FPGA) products.
The resulting FPGA portfolio includes several specific brands, as listed by Yahoo Finance: – Virtex-6, Virtex-7, and Virtex UltraScale+ – Kintex-7, Kintex UltraScale, and Kintex UltraScale+ – Artix-7 and Artix UltraScale+ – Spartan-6 and Spartan-7
Additionally, the company provides adaptive SOCs under the Zynq-7000, Zynq UltraScale+ MPSoC, Zynq UltraScale+ RFSoCs, Versal HBM, Versal Premium, Versal Prime, Versal AI Core, Versal AI Edge, Vitis, and Vivado brands. The Versal Adaptive SoC line specifically integrates AI Engines (AIE) to accelerate machine learning workloads. Compute and network acceleration board products are offered under the Alveo and Pensando brands, according to Yahoo Finance. The Pensando integration provides distributed services platforms and data processing units (DPUs) for cloud-scale networking and security.
Historical Competition and Market Capitalization
AMD was formally incorporated on May 1, 1969, by Jerry Sanders and seven colleagues from Fairchild Semiconductor, according to Wikipedia. The founding group included Ed Turney, John Carey, Sven Simonsen, Jack Gifford, Frank Botte, Jim Giles, and Larry Stenger. Sanders, a former director of marketing at Fairchild, started the firm after becoming frustrated with the lack of flexibility and opportunity at his previous employer.
The company’s initial product focus was on memory chips and other computer components. AMD entered the microprocessor market in 1975, establishing a long-term rivalry with Intel, according to Wikipedia.
The company saw significant growth in the early 2000s, driven by its position in the PC market and the success of its Opteron and Athlon processors. Under the leadership of CEO Lisa Su, who assumed the role in 2014, the company pivoted toward high-performance computing (HPC) and data center growth. This trajectory eventually led to a milestone in 2022, when AMD’s market capitalization surpassed that of Intel for the first time, according to Wikipedia.
