Fourteen billion smartphones, 3.6 billion computers, and many more consumer electronic devices all share a common lineage: a young man’s fascination with the French horn.
One of the most celebrated and innovative IBMers across the company’s long history, Robert “Bob” Dennard hails from humble beginnings. Born in 1932 in Terrell, Texas, Dennard spent his early years on a farm with no electricity, attending classes in a one-room schoolhouse. In high school, Dennard established himself as a promising student but harbored no grand ambitions. When a guidance counselor suggested he pursue the growing field of electrical engineering, Dennard replied that he planned to join his friends at the local junior college.
Then a band recruiter from Southern Methodist University discovered Dennard’s talents with the French horn and offered him an alternative path. Dennard accepted a scholarship to the university, where he studied electrical engineering. It was the first step toward what would become his crowning achievement: the invention of dynamic random-access memory, better known as DRAM, the revolutionary memory cell inside nearly every computer, server and consumer electronic device on the market today.
In 1958, Dennard completed his PhD in electrical engineering at the Carnegie Institute of Technology, now known as Carnegie Mellon, and joined IBM as a researcher. It was a time of technological ferment. A Texas Instruments engineer had just demonstrated the first integrated circuit, which laid the groundwork for Dennard’s own research. Engineers everywhere were focused on advancing computer memory and logic, and no one more so than Dennard. He spent his early days at IBM studying metal-oxide-semiconductor field-effect transistor (MOSFET) designs and circuit applications.
In 1964, Dennard joined a microelectronics team intent on replacing the large, slow and power-hungry computer memory system — known as magnetic-core random access memory, or RAM — currently in use. They developed an alternative system that used six MOS transistors to store 1 bit of information on an integrated circuit. While promising, the design was still too complex and its output too slow. Dennard wanted something faster, more elegant and cheaper.
One night while lying on the couch and pondering a presentation given by his peers earlier in the day, he had a notion: What if he could store a bit of information in a single transistor? The insight was the catalyst for DRAM — Dennard’s most important innovation. Once the new technology came to market, it wiped earlier magnetic technologies off the map almost instantly and spearheaded dramatic progress in memory capacity and speed. Today, Dennard’s invention has permeated modern life.
DRAM is not Dennard’s only contribution to the evolution of 21st century technology. In 1972, he outlined a theory that set the framework for computers to get smaller, faster and more efficient every year.
Known as Dennard scaling, or Dennard’s law, the theory was based on Moore’s law, which posits that the number of transistors that can be fit into a given space will double approximately every two years. Dennard scaling predicted that the performance of each watt would increase at roughly the same rate. In other words, power consumption would drop as transistors were made smaller, enabling chipmakers to increase processor clock speed without increasing power draw.
Thanks to visibility gained from Dennard’s law, consumers came to expect rapid-fire advancements in product performance and capability, but the theory couldn’t sustain the demand. As engineers whittled dimensions, transistors began to leak, causing the chip to heat up and consume more energy. By 2005, Dennard’s law broke down. But its impact on the ecosystem of newer, smaller and ever-more-powerful memory chips endures.
Dennard’s work ethic and ingenuity shaped the culture at IBM’s Thomas J. Watson Research Center in Yorktown Heights, New York, a long-standing intellectual hive of top scientists and electrical engineers. Dennard became a cornerstone of the institution, where he hatched 75 US patents.
Russ Lange, an IBM Fellow and a former IBM vice president of technology, spent decades working alongside Dennard and recalled his intrepid commitment to innovation. “Bob and I would always have lively discussions about whether there was going to be an end to the scaling,” Lange said. “And he would say, ‘Yes, there’s an end to scaling. But there’s no end to creativity.’”
Throughout his half-century career at IBM, Dennard won a slate of awards in recognition of his seminal contributions. He was declared an IBM Fellow in 1979, received the US National Medal of Technology from President Reagan in 1988, was inducted into the US National Inventors Hall of Fame in 1997, and received the prestigious IEEE Medal of Honor in 2009. Most recently, in 2019, the Semiconductor Industry Association bestowed upon Dennard the Robert N. Noyce Award, the industry’s highest honor.
Though Dennard retired the French horn that paved his way, he never gave up his love of music. Throughout his retirement, Dennard continued to fuel his creativity through choral singing and Scottish dancing, which he practiced with his wife, Jane. Dennard died on April 23, 2024, at the age of 91.
A memory architecture sparked a revolution in computers, electronics and mobile phones
IBM has always provided its scientists the freedom to explore and discover solutions to society's greatest problems
IBM’s research into ‘spintronics’ is closing in on the creation of racetrack memory