New Chip-Architecture Design Program Relies on 'Intuition'

Over recent years, designing the best possible chip architecture – forcing as many transistors and connections on an ever-smaller microchip in an optimum manner – has become so complex, that many designers have simply outsourced their job to supercomputers. The machines browse through billions of arrangement possibilities, and select the best possible ones, but they lack something humans have plenty of, namely intuition and visual-pattern recognition capabilities. And, while these traits may not seem like much in an exact field of science, they are oftentimes the things that make a difference.

Indeed, experts agree that it's intuition and ingenuity that allow the best integrated-circuit designers to sometimes simply beat computers at coming up with the most advanced type of circuitry. To make use of these traits, experts at the University of Michigan have recently developed a new, online logic puzzle called FunSAT, which is meant to help designers make the best of their own work. According to its creators, UM experts and Computer Science Researchers Valeria Bertacco and Andrew DeOrio, the system can harness the designers' abilities to strategize, visualize and understand complex systems.

“Computer games can be more than a fun diversion. Humans are good at playing games and they enjoy dedicating time to it. We hope that we can use their strengths to improve chip designs, databases and even robotics,” Bertacco, who is also an associate professor in computer science and engineering at the university, explains. DeOrio is a doctoral student in Computer Science and Engineering, and will be in charge of presenting the team's work to the public on July 30th, at the Design Automation Conference, held in San Francisco.

“When solving these [satisfiability] problems, humans can use their intuition and visualization skills. For instance, by just glancing at the neighborhood map they can gain an intuition of where to begin in the case of the postal route. FunSAT can leverage these human skills that computer-based solvers do not have,” Bertacco adds.