When I was a child, I didn't quite know what I wanted to be, but I liked fiddling with machines and enjoyed making simple radios using common electronic kits.I also liked carrying out chemical experiments, so I joined the chemistry club and carried out organic synthesis in junior high school and high school.When I entered university, I planned to specialize in chemistry or biology.In the end, I was influenced by my seniors and majored in physics.At that time, there was a way of thinking about physics as wide-ranging, that is, physics exists in the field of chemistry as well as in the field of biology.In fact, there was this thinking that we could specialize in any field after we had studied physics; I had, therefore, thought that if I studied any new field for about one year, I could probably get along somehow.However, on moving to biology from physics, developments in biology are too fast to follow.I am carrying out my research with my deficiencies covered with the help of my colleagues.

In graduate school, I carried out research on laser spectroscopy and performed experiments on how rhodopsin protein changes upon applying a laser to the eyes of an octopus.I was sometimes puzzled myself over the interpretations of my experimental results, and thought I wish I could interpret them on the basis of the protein structure, but at that time, the structural analysis of large proteins was not yet advanced.Today, it is like a dream to see the protein structure revealed one after another in projects such as National Project on Protein Structural and Functional Analyses (Protein 3000 Project). Since that time, I have wanted to simulate molecules based on their structure.In graduate school, this research was my main theme.Meanwhile, as a hobby, I developed a computer exclusively for physics calculations.I think this was the first completely specialized computer for this purpose in Japan.After graduating graduate school, I became a research assistant of the College of Arts and Science, TokyoUniversity.There, developing specialized computers, which had been my hobby, became my main task.

Subsequently, I held a position in the Institute of Statistical Mathematics.There, I created a random number generator (RNG) that uses hardware to revise JIS and developed a specialized computer for matrix calculation.Because protein is a very complex substance, the interpretation of results is difficult even by computer simulation and statistics becomes highly useful.Statistics is the study of the way of analyzing things, and it is necessary to consider how experimental or computation results can be converted to various kinds of knowledge.Having learned the concept of statistics there, it is now coming in useful.Also at that time, I was developing a specialized computer, and wanted to create a larger, more practical one.

I was then invited to join the RIKEN project. Because my research was on protein in graduate school, I tried to develop a specialized computer for the simulation of the dynamics of biological molecules. I have been with this RIKEN project since 2002, and I developed the LSI chip that is the heart of MDCGRAPE-3, the spealized computer for molecular dynamics simulation in 2004.This computer with the LSI is now in practical use at Peta Computing Institute Limited, and is marketed at by the agent, SGI Japan.Using this chip, I plan to continue the development of faster application-specialized computers and aim to achieve the speed of 1 PETAFLOPS* in the spring of 2006.

*FLOPS is an abbreviation of FLoating Point Operations Per Second.It is one of the units of the processing speed of computers.

---------- I gained the impression that all things that Dr. Taiji has carried out up to now has made him the outstanding researcher that he is today. (From the interviewer)