MOS SRAM

By 1964, when John Schmidt built the first metal-oxide-semiconductor static RAM at Fairchild Semiconductor, the invention was the inevitable convergence of three distinct technological streams in Silicon Valley. The foundation began in 1918 when William Eccles and F.W. Jordan invented the flip-flop circuit using vacuum tubes—an electronic toggle maintaining stable on or off states. For decades, flip-flops existed as laboratory curiosities, too bulky and power-hungry for practical memory.

The substrate arrived in 1959 when Mohamed Atalla and Dawon Kahng at Bell Labs invented the MOSFET—a transistor switching states using minimal power while occupying far less space than vacuum tubes. The MOSFET's metal-oxide-semiconductor structure controlled current through electric fields rather than physical contact. But in 1959, nobody knew how to manufacture MOSFETs at scale.

The third stream emerged from Fairchild's development of the planar process. By 1963, Robert Norman at Fairchild had invented bipolar SRAM—memory cells using bipolar transistors in flip-flop configurations. But bipolar cells consumed significant power and generated heat. Schmidt's 1964 invention substituted MOSFETs for bipolar transistors. The six-transistor cell design—two cross-coupled inverters for storage, two access transistors for control—became the standard persisting today.

The final piece arrived in 1968 when Federico Faggin developed silicon-gate MOS technology enabling finer feature sizes. By 1969, Intel introduced the 3101 bipolar SRAM as its first product while developing the 1101 MOS SRAM. The 1101 lost the race to market but won the war: MOS memory was cheaper, denser, and more scalable. SRAM's speed advantage—no refresh cycles, sub-nanosecond access—made it ideal for processor cache memory enabling modern computing's multi-level cache hierarchies.