If you ever designed a circuit you used SPICE … or you never really designed a circuit.
When we think ‘open source’ we think Linux or Android or some other recent operating system endeavors. But imagine for a second that not just these systems but the hardware used to run them is all derived from circuits designed and analysed using an open source program created as a student project back in 1969!
Here is a quick summary showing why SPICE is so important to electrical engineers, which is inscribed on a plaque at UC Berkeley’s Corey Hall which was dedicated on February 20th:
SPICE (Simulation Program with Integrated Circuit Emphasis) was created at UC Berkeley as a class project in 1969–1970. It evolved to become the worldwide standard integrated circuit simulator. SPICE has been used to train many students in the intricacies of circuit simulation. SPICE and its descendants have become essential tools employed by virtually all integrated circuit designers.
Simulating a circuit with SPICE is the industry-wide standard way to verify operation at the transistor level before manufacturing an IC. The program has become so ubiquitous that engineers often say they are going to “SPICE a circuit” when they are about to test one.
To mark the 40th anniversary of SPICE—the Simulation Program with Integrated Circuit Emphasis—IEEE has designated its creation an IEEE Milestone in Electrical Engineering and Computing.
As circuits evolved from simple transistors and diodes to integrated circuits, tools were needed to help designers understand system performance:
The process of making an IC is expensive, and probing the behavior of internal signals is difficult. SPICE checks the integrity of circuits and predicts how they will behave before anything is manufactured.
Simulating a circuit involves using mathematical models to replicate its behavior. Simulation can determine problems in the design early on, leading to a better final design and significant cost savings. Almost all IC design relies heavily on simulation.
IC designers aren’t the only ones using SPICE, though. It is part of engineering curricula to help teach students the fundamentals of circuit design. In fact, the classroom is where SPICE was developed.
Here is some background on its creation:
The program began as a class project in 1969 for a circuit design and analysis course taught by Ronald A. Rohrer, a professor at the University of California at Berkeley. Rohrer, an IEEE Life Fellow, said the best way for his students to learn the subject was to do hands-on work, so he assigned them a project to develop a comprehensive IC simulator.
The students were told they would pass only if their program impressed Donald O. Pederson, a professor of electrical engineering and computer science at the university, and a pioneer in transistor research. Pederson, for whom the IEEE Donald O. Pederson Award in Solid-State Circuits is named, was instrumental in establishing the first IC fabrication lab at UC Berkeley in 1960. At the time, IC fabrication labs were found only in high-tech companies, so the Berkeley lab was the first to give students hands-on experience.
At the end of the course the students had indeed developed a simulator and gave it a rather unusual acronym: CANCER (computer analysis of nonlinear circuits, excluding radiation). The name—particularly the words “excluding radiation”—were purposely defiant, because at the time, circuit simulation programs were typically funded by the government for defense purposes and were required to include a feature to test a circuit’s resistance to radiation.
“The name CANCER was a bold statement that this program never would simulate radiation and was not funded by the defense industry,” according to Laurence W. Nagel, one of the students in that Berkeley class. (See “The Origins of SPICE”.)
CANCER was the first simulator to use sparse matrix techniques, which made it possible to simulate greater orders of magnitude. Pederson was impressed with the students’ efforts, and they all received passing grades. But Nagel’s work on the program didn’t end there. He focused on improving CANCER for both his master’s and doctoral theses. Rohrer was his advisor for his master’s thesis, and Pederson became Nagel’s doctoral advisor.
Pederson’s goal for Nagel was to redevelop CANCER to make it a public-domain, general-purpose circuit simulator—and to change its name. The result was SPICE, the first program to combine the circuit-analysis capabilities of DC analysis, AC analysis, and transient analysis.
SPICE was made publicly available in 1971 so that chip designers could modify it—an early example of open-source software. Two years later, SPICE became well known after it was described in a paper by Pederson at the 16th Midwest Symposium on Circuit Theory, in Waterloo, Ont., Canada.
During the next few years, developers around the world began using and modifying SPICE, leading it to become the industry standard it is today.
“What happened was truly phenomenal,” Nagel wrote in “The Origins of SPICE.” “Within a few years, SPICE had achieved acceptance at almost all electrical engineering schools [for use in teaching] and had [spawned] a cottage industry to supply SPICE derivatives to the rapidly expanding integrated circuit industry.”
It has been more than a few years since I needed to use SPICE tools as my career has headed in different directions since those days, but it remains a foundation of how I think about problems – using structured, logical, analytical tools to test hypotheses. Even during the years I did use SPICE the tools evolved rapidly as computers became exponentially more powerful – but they never lost sight of the core mission of circuit design and analysis for young electrical engineers.