DATE: The legacy of Mead and Conway

What is the legacy of the most famous textbook in the world of chip design: Mead and Conway’s Introduction to VLSI Systems?

Although some of the core assumptions of Carver Mead and Lynn Conway’s book were misses and the idea of the silicon compiler was outcompeted by other ideas, more than 30 years on it remains a key blueprint for the way that we think about IC design. But some of the core ideas at the heart of the book are even now being threatened by physics. Those are some of the outcomes of a panel at the DATE conference on Tuesday that sought to examine the relevance of a work that has outlasted the technologies that it accompanied.

Professor Alberto Sangiovanni-Vincentelli of the University of California at Berkeley said the core ideas of the methodology date back to 1970 – the concept finally crystallizing in courses taught in part by Mead and Conway at UC Berkeley, Caltech and MIT before the book was first published towards the end of the decade in late 1979.

The book’s direct legacy was education. “Many classes started out of this book,” said Sangiovanni-Vincentelli.

A key thrust of the Mead and Conway approach was to provide students with hands-on experience of chip design. “IC design was an expensive proposition. But putting multiple chips onto the same wafer allowed to do design at a cost that was not prohibitive.”

Bernard Courtois, director of CMP and one of the first to teach Mead and Conway’s methods in Europe, said: “One major follow-up was the establishment of service centers for manufacturing ICs, sharing the cost by sharing the wafers.”

Today, the multiproject wafer services provided by groups such as MOSIS, CMP and Europractice provide hundreds of universities, research institutes, and small and medium-sized enterprises (SMEs) to gain access to even advanced process technologies such as 28nm CMOS and fully depleted silicon-on-insulator (FD-SOI).

The work led to the idea of the “silicon compiler”. Although the commercial tools that built on this worked ultimately failed the idea of using a high-level language to generate chip layouts today lies as one of the cornerstones of modern design. The reason why it is possible is that the separation of design and manufacturing that Mead and Conway proposed has been upheld consistently well up to now, although what former head of IMEC Hugo de Man calls “physics hell” – the variabiilty and other effects caused by nanometre-sized features – is causing leakage between the two domains.

“The principle of design foundries was born then,” said Sangiovanni-Vincentelli. “The idea that you have to limit the freedom of design using regular structures using tools was another important principle.”

Professor Jan Rabaey of UC Berkeley said the imposition of rules maintained a clear separation between design and manufacturing but “maintained a degree of freedom to allow the designer to be creative”.

Antun Domic, senior vice president and general manager of the implementation group at Synopsys, said: “The separation of design from manufacturing can’t be underestimated. You were sending a layout to someone and chip would come back.

“There were misses,” said Domic, pointing to the book’s emphasis of the then leading technology NMOS. “CMOS was much closer than anyone thought. I remember that the first Vax processor was NMOS. The second was CMOS.”

Domic said MOSIS was in a position to offer access to CMOS manufacturing as early as 1983.

Sangiovanni-Vincentelli said the problem for silicon compilers lay in their results. “The designs were difficult to debug and the chips they produced were big. So they were not competitive with other design methodologies. What changed the picture? Optimization. That was what was missing. All the EDA companies we know today are all based on the use of optimization.”

Domic said: “There was too much emphasis on using custom generators. It was too complicated to use – designers used less optimization than one would have wanted.”

Luca Carloni of Columbia University said: “What has remained the same 30 years later is the importance of interdisciplinary approaches to R&D.”