180 nm process

Lore & Background

The 180 nm value originates from a historical trend of 70% scaling every 2–3 years, formally defined by the International Technology Roadmap for Semiconductors (ITRS). In 1988, an IBM research team led by Iranian engineer Bijan Davari fabricated a 0.1 µm MOSFET demonstration, not a 180 nm dual-gate MOSFET. TSMC commercialized the process in 1998, followed by Fujitsu in 1999.

Reader's Guide

The 180 nm process was a pivotal node in semiconductor manufacturing, being the first to achieve a gate length shorter than the 193 nm wavelength of contemporary lithography. This breakthrough allowed continued miniaturization beyond optical limits. It was widely adopted for CPUs and GPUs from 1999 to 2004, including the Intel Coppermine, AMD Athlon Thunderbird, and Sony PlayStation 2's Emotion Engine. Its longevity in low-cost microcontrollers demonstrates its economic viability. The process also enabled open-source hardware initiatives in 2022, highlighting its enduring relevance for cost-sensitive applications.

Did You Know?

• The 180 nm process was not the first to use a gate length shorter than the 193 nm wavelength of contemporary lithography; the 250 nm process already used sub-wavelength lithography with 248 nm light.
• An IBM team led by Iranian engineer Bijan Davari fabricated a 0.1 µm MOSFET demonstration in 1988, not a 180 nm dual-gate MOSFET.
• In 2022, Google sponsored open-source hardware projects using GlobalFoundries' 180 nm MCU process.
• The process was used for the Nintendo GameCube's Gekko CPU, but the Sony PlayStation 2's Emotion Engine was fabricated on a 250 nm process.

From Lab Bench to Foundry Floor

The 180 nanometer designation did not spring from a single breakthrough. Its name traces back to a historical pattern in which linear dimensions shrank by approximately 70 percent on a two-to-three-year cadence, a convention formally codified by the International Technology Roadmap for Semiconductors. The practical roots, however, reach much further into the past. In 1988, a team at IBM, guided by Iranian-born engineer Bijan Davari, managed to produce a dual-gate MOSFET at the 180 nm scale on a CMOS platform, proving the geometry was physically realizable nearly a decade before the industry was prepared to mass-produce it. TSMC ultimately brought the node to commercial production in 1998, with Fujitsu following in 1999. The adoption wave then rippled across the globe: Sony, Toshiba, Intel, AMD, Texas Instruments, and IBM all stood up 180 nm lines during the 1998-to-2000 window, cementing the node as the industry's shared workhorse for an entire generation of silicon.

Crossing the 193 nm Threshold

One of the most quietly consequential aspects of the 180 nm node is what it signified for photolithography. Intel's Coppermine family of Pentium III processors, which reached the market in October 1999, represented the first instance in which a shipping processor's gate dimensions fell below the 193 nanometer wavelength of the deep-UV light source in standard lithography tools of that era. In practical terms, engineers were now patterning features smaller than the light source itself, a regime that would demand increasingly sophisticated optical corrections, immersion techniques, and multi-patterning strategies in the years that followed. The 180 nm node thus sits at a pivotal inflection point: it was the last geometry where the physical optics of the exposure tool still roughly matched the feature sizes being produced. Every subsequent shrink pushed further into a territory where the gate no longer corresponded to a simple projection of the mask, fundamentally reshaping how semiconductor manufacturing would evolve for the next two decades.

The Chip That Defined a Generation

Between 1999 and 2004, the 180 nm process became the backbone of an extraordinary spread of consumer electronics. Intel's Coppermine E Pentium III (October 1999) and the later Celeron Willamette (May 2002) anchored the desktop PC market. AMD's Athlon Thunderbird and the budget-oriented Duron Spitfire both launched in June 2000, with the Duron Morgan arriving in August 2001. In graphics, ATI's Radeon R100 and the RV100-based Radeon 7000 appeared in 2000. Gaming consoles felt the node's impact directly: Nintendo's GameCube shipped its Gekko CPU in 2000, and Sony's PlayStation 2 debuted its Emotion Engine processor and Graphics Synthesizer in March of that same year. Motorola's PowerPC 7445 and 7455, known as the Apollo 6 family, entered in January 2002, while Transmeta's Crusoe and the first-generation Efficeon (TM8600 and TM8620) targeted mobile and embedded applications through 2004. No single node in that era touched as many product categories simultaneously.

A Node That Refused to Fade

While the industry raced toward 130 nm, 90 nm, and beyond, the 180 nm process found a second life defined not by speed but by economics. Because it is comparatively low cost and can be run on existing fab equipment without the capital expenditure of upgrading to newer lithography tools, it became the go-to choice for microcontrollers and embedded silicon. Microchip's PIC family of microcontrollers, for example, continued to be manufactured on 180 nm lines well into the 2020s. The node's longevity also made it an ideal substrate for experimentation and collaboration. In 2022, Google sponsored open-source hardware projects that leveraged GlobalFoundries' 180 nm MCU process on multi-project wafers, allowing independent designers and academic teams to tape out custom chips without the prohibitive cost of cutting-edge nodes. In this way, the 180 nm process evolved from a cutting-edge commercial node into a durable, accessible platform that keeps serving engineers who value simplicity, reliability, and affordability over raw transistor density.