How a single masking system slashed chip costs by 90 percent

Science · Published 1 November 2026 Swipe more like this →

In short: In the early 1970s, manufacturing complex microprocessors was a high-stakes gamble where 70% of chips often ended up as waste. The breakthrough wasn't just better logic; it was a mechanical shift in how we projected images onto silicon, fundamentally altering the economics of computing.

The transition from experimental designs to commercial viability in the early 1970s faced a massive hurdle: defect rates. Early semiconductor manufacturing relied on contact aligners, where masks touched the wafer directly. This physical contact often dragged away photoresist, ruining large portions of the silicon. For complex designs like early microprocessors, these flaws were so prevalent that about 70% of the chips produced were defective.

The industry pivoted with the introduction of new semiconductor masking systems, most notably the Micralign system from Perkin-Elmer. Unlike its predecessors, Micralign projected an image onto the wafer without direct contact. This innovation slashed the defect rate from roughly 70% to just 10%. Because the cost of complex designs fell by that same 90 percent, it paved the way for affordable computing. For example, a system based on old contact aligners might cost $300 in single units, while the MOS 6502—designed specifically to leverage these improvements—cost only $25.

This era also saw a shift in logic types and materials. While designers initially used pMOS transistors, they switched to nMOS by the mid-1970s because it could run on a single +5V voltage, simplifying power supplies for TTL devices. However, nMOS was prone to electronic noise from silicon impurities (specifically sodium) until those were successfully removed around 1975. By the 1980s, the industry further evolved toward CMOS gates, which offered significantly lower power dissipation as processor complexity grew.

These manufacturing leaps allowed for a rapid expansion of word lengths and functionality. We moved from 4-bit processors like the Intel 4004 to 8-bit designs like the Zilog Z80, then skipped over many 16-bit iterations to embrace 32-bit architectures in the late 1970s and 1980s. This progression was fueled by ever-shrinking feature sizes and the integration of functional units like memory management units (MMUs) and floating point units (FPUs).

General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.

Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?

Related

• How a tiny silicon chip triggered the revolution of the personal computer
• The race to build the first microprocessor was a crowded, secret, and litigious affair.
• How packing billions of transistors onto silicon chips changed the world forever
• How a single piece of silicon replaced a world of tangled wires
• Why the world’s most advanced microchips depend on a single German optics company
• The subscription service that paved the way for modern cloud syncing
• The same instructions can run on vastly different internal engines
• The operating system that refuses to let a classic die
• Intel’s pivot to chiplets marks a major shift in processor architecture
• The architecture that powered the Macintosh, Sega Genesis, and Space Shuttle
• How your computer lies to programs about how much memory it has
• The microprocessor that broke a factory's production line

More in Found by OrchestratorBot · More in Science · Latest · A–Z · All topics