Sunlin Chou: A tribute

Sunlin Chou passed December 5th, 2018. While his legacy had faded since he retired in 2005, his career included filling Gordon Moore’s shoes as a key technologist in the second generation of semiconductor technical leadership. Sunlin’s Chou last position was in running the Technology side of Intel’s Technology and Manufacturing Group. He was in that group of shadow warriors who run R&D. They seldom get any public light and only the people in their company really know what they do. Yet they bear a tremendous burden. They must decide among the myriad of technologies, which ones really deserve attention, funding, etc. They must ply the foggy seas of the tech triangle that lie between customer need, realization in product, and manufacturability. It is a place where ship wrecks occur all too often. For Sunlin, the challenge was even greater. He had to fill Gordon Moore’s shoes. That meant not only being right, but also being above industry politics. You can learn a lot from someone who has successfully sailed these seas.

At the core of Sunlin’s values was the belief that all problems should be dealt with systematically. He was one of the most systematic individuals I ever met, exemplary in many ways: every hair was always in a precise place, clothing all in alignment, shoes polished, and a poker stare that came from years of making many calculated gambles on future technology. This is exactly how he dealt with the issue of reviving Intel’s prowess after the 1985 debacle.

Sunlin’s role in this was to develop a new business model for research and development. While many people were looking for an outside solution, like consortia; Sunlin was looking inside to see how the process could be streamlined. Most knew the model wasn’t working well. Research often never led anywhere for the companies that funded it (Xerox Parc being the best example). It was a money sink, because few innovations ever made it to the business side of an organization. Yet few did anything about it.

Sunlin looked at these problems and saw opportunity.

Sunlin’s most important contribution is his R&D Pipeline model, where he applied the first systematic approach to the whole development cycle. Central to his method was a strong understanding of Moore’s Law and how it impacted product life cycles. It led to his decision to set technical direction based on its relentless progression. This led to node planning. What came next was truly innovative. First, he covered the horizontal time line with a flow model that started with the research infrastructure outside Intel; then feeding it inside to research; then to path finding, on to development; and finally, to manufacturing – all synchronized to the nodal clock.

To do this he had to create a copy-exactly infrastructure for R&D that had never existed. Plus, everything had to be tightly coupled at the hand-off points, which is where most fail. This was done with exacting business processes. Second, he also covered the vertical technology line, integrating the test, assembly, and process roadmaps into a single cohesive strategy/roadmap. It doesn’t get more systematic than this. The result was that Intel was able to shift into high gear, running on a 24-month nodal clock.

Another thing that always amazed me was the clarity with which he thought – especially when it came to transferring technology into manufacturing or avoiding falling into the trap of implementing new technology for technology’s sake. I remember once telling him that Intel was behind because they were not implementing copper at 180nm. He said, “We’re not behind. We can implement it, but we have chosen not to.” While I was thinking, ‘yeah . . . sure’ – then remembering that it was Mark Bohr’s famous graph that kicked off the copper race in the first place – I asked “Why?”

Sunlin: “Implementing copper at 180nm is easy. But the real difficulty won’t come until 130nm. We have chosen to postpone it a node to understand its manufacturability better.” Two years later, Intel introduced copper into production at 130nm without skipping a beat. Many others, who had gained false confidence at 180nm ran into serious roadblocks at 130, slowing them down and marring their reputations. He had been right on a point so subtle that most others missed.

He had seen 130nm as a critical juncture for the semiconductor industry, as that node would be the first where physical gate lengths would be below 100nm, the official demarcation point between microtechnology and nanotechnology. While university and corporate researchers were loudly declaring the death of a semiconductor industry that would soon be displaced by nanotechnology, Sunlin had quietly been preparing to break the nanotechnology barrier and make semiconductors the first large-scale nanotechnology.

He knew crossing the nanotechnology barrier would not be easy, nor cheap. He had foreseen a collision between the end of simple Dennard Scaling and the technology smorgasbord from IBM Research and Bell Labs that had always been free for the taking. Sunlin had architected his Research Pipeline to avoid this collision, putting Intel in a perfect position to make the leap from Dennard Scaling to a new era of Materials Enabled Scaling, which Intel would lead. 

Intel’s 130nm went off without a hitch, while those researchers’ visions of disruptive nanotechnology industries proved to be mostly smoke and mirrors. The true disruptions were coming from companies enabled by Sunlin’s brand of nanotechnology: Facebook, Apple, Amazon, Netflix, and Google — a group of disruptors that would come to be known as FAANG.

Finally, Sunlin’s highest values were credibility and integrity. For as long as I knew him, he never gave way to an easy PR gain. One thing he was always strict on was his metaphors. Whenever they didn’t match the technology he was trying to explain closely, he tossed them. So he never got into trouble.

While metaphors are essential to communicating highly technical issues to a non-technical world, a wrong metaphor can haunt you later. Would-be competitors can weaponize a bad metaphor against you. The following provides a good example to explain why it is so important to get them right. The then-current media roasting of Intel over power started with just such a wrong metaphor laid out a few years before, when an Intel executive used the metaphor of a nuclear power plant in a presentation to explain the growing power problem. The metaphor and the chart were powerful mindshare grabbers, but far short of accurate. First, power had always been a problem at the leading edge of performance. Second, the issue isn’t raw power dissipation. Nor is it dissipation at the transistor level, which is always dealt with via design and packaging. The issue is power density, which is dissipation-per-area. Third, there was no way that a chip would ever have the power density of an entire nuclear reactor. It was exponential extrapolation gone wild.

But the metaphor was powerful. Others would take it out of context to make their points that semiconductors were heading into a dead end. If you were in nanotech and needed funding, what better way to get money than to show that chips were coming to an end by presenting something as powerful as an Intel’s nuclear power plant slide as proof? Soon, people were overlaying images of nuclear power plants on the graph and the pro-Apple crowd had animations on the Internet showing a beige box with a Pentium roasting a chicken. Silly, because you can’t cool a nuclear power with a $5 fan and one can’t roast a chicken with a 100-Watt output. But it was highly effective and highly damaging.

In the end, it was Sunlin who took up the challenge of fixing the misperception. He was the first Intel executive to address the power issue in a coherent and comprehensive manner. It was done in his usual, albeit dry, way: with facts and figures. It is important to have credibility, but Sunlin commanded credibility, which is why the issue faded away so fast. He did this by being factual and dead serious.

So, there you have it, if you approach problems systematically, have clarity of vision, and never trade on your credibility, you will go far. Thanks, Sunlin, for being a shining example.

Here are links to two interviews with Dr. Chou:

       https://www.chiphistory.org/66-intel-s-r-d-pipeline

       https://www.chiphistory.org/430-power-performance-cost-and-moore-s-law