An interview with the creator of stacked CMOS image sensor—logical thinking behind the epochal innovation

A miraculous journey of the development
Positive expectations and support propelled the project to mass production

The project of stacked CMOS image sensor had another major challenge. While the separation of the two components created the possibility of mounting various additional technologies onto the logic circuit section, there was no clients chosen for this product even when the mass-production phase was in sight. Preparations were ongoing to set up mass production lines at the Kumamoto and Nagasaki Technology Center（TEC） in 2011. At this stage, however, the lack of client was stirring a skepticism as to the legitimacy of mass-producing the product. There was even a talk about suspending the development project. When the mood was one of uncertainty, the accounting division expressed their support saying, “Don't worry about the development cost which has been spent. If you think the product is good enough to recover the mass-production cost, it’s worthwhile going ahead.” Umebayashi believes that this advice boosted the morale in the team and enabled them to take the image sensor to mass production. Eventually, it was decided that the product be implemented in Xperia Z and its subsequent models, creating an opportunity for the stacked structure to be widely recognized, leading to a huge market success.

Despite numerous difficulties, the project survived to reach the mass production stage. “I believe that everyone wanted this technology to succeed. When the initial PoC failed, normally the choice would be either to end the project or replace me with someone else, but both the project and I were allowed to continue. The process team also kept trying again and again. I also learned it afterwards that my bosses were helping us behind the scenes, too, with securing the budget and making necessary arrangements for the launch of mass-production lines,” said Umebayashi, describing the strong backing the project had. “We were at our lowest toward the end of 2009. Then came 2010, the sample finally produced some image. Boy, I was so happy to see it,” he smiled. This breakthrough led to a successful sample test at the Kumamoto TEC, and the tide was turned. What started as a discrete team effort was transformed into a research work that garnered the ‘internal attention.

Logical thinking behind actions and policies

What is the source of this epochal innovation? Umebayashi says that he had a great passion for mathematics at school. “I would tackle the math textbook as soon as I got my hands on it at the beginning of a school year. I would finish it before the second term, so I would ask my math teacher to give me the textbook for higher grades.” When he was only a child, he enjoyed solving mathematical problems with his father, which gave him the impression that he was playing puzzles or quizzes. “For me, math was not a study but more like playing a game,” he recalls. Perhaps the logical thinking which he developed through mathematical training was the foundation of his idea for the streamlined stacked structure. His personal philosophy of work is “to think logically to determine my actions and policies” and “understand the outcomes in terms of the system, cause and reasoning behind them.” A developer must have an attitude like this. He says, “It is important not to speculate uncertainly and to be honest. Meanwhile, for achieving something that is technically realizable, you must be bold. If you don’t take an action, possibilities for the future may never have a chance to be tested.” His genuine and honest attitude to technology, he believes, is what enabled him to persevere almost two years of struggles to come to the major success. When he finds himself at a dead end, however, he does not hesitate to consult a few select colleagues to discuss the matter to avoid being stuck with an idea that leads to nowhere. He often finds that discussions in small groups offer him new perspectives and help to find a solution. Meanwhile, English was not his favorite subject at school. As his jobs at Sony increasingly found him in English-speaking situations such as international academic conferences, Umebayashi was seeking an opportunity to improve his command of English. His choice was to take a post in the USA. He put his name forward for the overseas assignment even though his TOEIC score did not meet the requirement for the candidacy. He “wanted to take up a new challenge” and thought it would be “an opportunity to throw myself in at the deep end, and I would have no choice but to learn English.” For Umebayashi, it was an extension of his attitude to work, a decision drawn based on his logical thinking. His experience in the USA turned out to be positive. He learned the importance of flexible thinking as he found the ways of doing things were “loose in a good sense” and that they had a totally different approach from the Japanese ways to manufacturing semiconductors.

Focus not on outcomes
but the reasons behind them

Umebayashi set his mind on a career in research and development when he was at university. He liked working in a lab, and especially, experiments more than theoretical developments. He used to run experiments with his peers, staying in the lab often until late. Finding much “fun in research,” he became determined to have this career path. However, corporate research and development was very different from the academic pursuits. While the latter focused on principles, physical interpretation and other theoretical perspectives, the former took a more pragmatic approach to the development, such as how the product should be applied, what value it added, and what needed doing to make practical applications. He also recognizes that Sony’s corporate culture that respects individual freedom has shaped his attitude to work. He says, “During my first year at Sony, I noticed that my senior colleagues did not scrutinize the research or measurements I undertook. I learned from them that it was more important to think and understand the process behind the outcomes rather than verifying the validity of the results.” There is a long-standing golden rule at Sony: If you have a good idea, pursue it without waiting for boss’ approval — if it works, make it official; if not, you can leave it there. This rule epitomizes the workplace environment that allowed Umebayashi to take up and pursue new ideas until he succeeded, and it culminated in the stacked CMOS image sensor.

He says, “It is most satisfying in my work to see the products I worked on find many users. I will continue to work on commercializing products that the world has never seen and satisfy the desire of many people.” We wish him great success in his future career.

Plaques of the Prime Minister’s Award (2016 National Invention Awards) and 2020 decoration of the Medal with Purple Ribbon