The English version of Nikkei Tech-On is featuring the interview with Ken Kutaragi about the Cell project. It's basically the English version of the interview referred in this thread before. Currently only pt.1-3 are online but pt.4 may be online in a week or so if we're lucky.
http://techon.nikkeibp.co.jp/english/NEWS_EN/20050407/103542/
http://techon.nikkeibp.co.jp/english/NEWS_EN/20050407/103542/
"Because It's an Once in a Lifetime Challenge-" Interview with Sony's Ken Kutaragi
Apr 11, 2005 09:30
By Naoki Asami, Editor-in-chief of Nikkei Electronics
Is the prototype chip that was unveiled close enough to the first image of Cell you had in your mind at the initial stage of development? How much of your targets do you think you have achieved?
It's a miracle, indeed. The chip came out far beyond my expectations. When this product was launched, I set out a number of propositions, like "I want distant cells on earth connecting and communicating with each other organically," "there should be object-oriented ideas," "it's got to be secure," and so on. The team was more than just marvelous in reacting to these demands. They did such a fantastic job that I'd feel ashamed to even have a thought about rating their achievement.
When did you yourself start forging the Cell initiative?
I started thinking of the idea before the launch of the PlayStation 2 (PS2), around the summer of 1999. There were a mound of things needing to be done about the PS2 in terms of nurturing it as a business, but in terms of technological development, it was already a thing of the past to me. So, I was dying to think about what's next. The idea of a computer architecture that emulates living organisms struck me as I was swimming through a sea of ideas. A network made of cells that works like a single computer? I felt this wild urge to try out that sort of thing.
Looking back, the paradigm of the computer has not changed in essence since the birth of ENIAC in 1946. From CISC to RISC to VLIW, more and more advanced technology had been mounted one after the other on semiconductors, but we have come to an end now of this path, so to speak. We have accomplished the height of this development. So, I wanted to bring in completely different ideas and make a revolution in existing computer architecture.
The history of the computer took a turn in 1995, when PCs were connected to the DARPA Internet. The Internet itself had already been around for a while, but from the moment it became available to PCs of general consumers, dramatic changes occurred. PCs were no longer a mere calculator but became a means to access information. This naturally gave rise to demands for a new, more suitable computer architecture.
The network of today still remains a means to access information. The time will come, however, when the network itself becomes a computer. So far, we have pursued how to improve the performance of a single chip, but we are now becoming aware of the limits that loom there. Once you suppose that the network is a computer, what you have to do is to improve the computing performance of the network as a whole. This is s paradigm shift. Be it on-chip, on-board, or on-network, chips will be infinitely connected -- and that is what this "Cell" is all about.
One Word Brought the Team Together
Apr 14, 2005 18:18
By Naoki Asami, Editor-in-chief of Nikkei Electronics
Did you have any concerns in carrying out this project?
We all have our own realities, such as quarterly goals to meet or solving the most recent product issues. We cannot help but being shortsighted in thinking that way. Honestly, I was not really confident that I could find engineers who could embrace such far-fetched, bold ideas and share this dream of ours. To me, this was not just an empty wish, but something that was logically sensible. Well then, would I be able to win the empathy of top-level engineers? I wasn't sure. This was the kind of a project that could create a new history, and so I needed to have everything in place -- the right people, a substantial amount of time for a project that could last for some years ahead, and drawing out management decisions from the participating companies to pour management resources -- all needed to be there at the global level and transcend the framework of a single company.
And you chose Toshiba and IBM as your partners.
I was rather optimistic in getting support out of Toshiba, as I knew a lot of people there through the PS2 project. In addition to Toshiba, I wanted IBM to be a part of this project, as this project was all about setting off a paradigm shift. Some say I went after IBM because they had the SOI technology, or because they had PowerPC technology, but these weren't the reasons. These two technologies are undoubtedly important in realizing Cell, but what mattered most to me was to work with a partner that could really share with us the strong determination to change computing history. We all know that IBM is the leader of the computer industry. The company created the era of the mainframe with System/360. As the word "PC" once almost meant exclusively "IBM PC," PC technologies originated from IBM technologies. In other words, IBM had always been there in the history of computers. And so, IBM was indispensable for this project to create the next computing paradigm.
Was IBM really cooperative from the beginning?
They really were; they orchestrated an all-star team for us. It was like lining up an Olympic athlete-class of engineers. I was really thrilled by their enthusiasm.
It must be hard to mesh the vectors of the different companies.
I myself had initially thought it would be. I thought there would be some kind of strains, in one way or another. But the engineers consolidated the instant they shared a sense of purpose. Actually, there's a story to this. One day, when we were having a meeting, one of IBM engineers - and he was among the top star architects - got so excited that he began to speak very passionately. It happened to be an extraordinary scene for the rest of the IBM engineering team as well. Then, somebody said, "Why are we so hot about this?" and somebody else simply said, "Because it's a once in a lifetime challenge." That stroked the hearts of twenty engineers. It was at this very moment that I realized their vectors got aligned in one direction. Just think of the Apollo program that sent human beings to the moon, or Celera's project to analyze the human genome. Not every engineer has the chance to participate in such projects that can impact the course of the history. Cell is exactly this kind of project. In the day-to-day world, engineers design microprocessors with model numbers. In this project, however, you are to create a chip that has a proper name. You never know, thirty years from now, some members of our team may end up being looked up to just like Shockley or Moore. Wouldn't that be just exciting?
The Power of Two Is an Aesthetic
Apr 26, 2005 16:30
By Naoki Asami, Editor-in-chief of Nikkei Electronics
On the other hand, it takes real technologies to realize a dream.
I intend to stick to 90nm technology. I know some people who suggested going into mass production using such outrageous technologies as 65nm or 45nm, but you can't make something only by dreaming. In my management role, I'm determined to start mass production with the 90nm. Unlike in the past, it has become extremely difficult to establish a single process technology. What used to take six months after the delivery of manufacturing equipment now takes two or even three years. In other words, once technology uses fine scale processing, you now have to face ten or twenty new problems that bubble up along the way, where in the past, one action was enough. It takes strong conviction and confidence to move on. That is why a stable process technology is so important for the mass production of Cell. Whether we like it or not, we cannot slow down the adoption of fine scale processes in the future, and that is a fact. In this sense, the significance of "PSP" is enormous. Through mass production of PSP chips, Sony succeeded in stabilizing 90nm manufacturing technology. And with that, we were fully prepared-including the injection of new material - to take on new challenges for Cell.
Do you have any idea of what the yield will be?
I'm not going to give a green light unless we attain a certain yield rate level. Just think of the amount of chips we are to produce - 20 million for game consoles alone, an estimated 10% of TV sets, as well as home servers. At this scale, brute strength alone won't do. We are not aiming to produce 100,000-yen chips. As Mr. Masatoshi Shima, who had designed the "4004" chip with Intel once wrote, Intel usually starts with the yield of 40% and then brings it up to 85%. That is a good curve. I believe there is universality in these figures that stands up regardless of time and use.
Cell has 8 embedded "SPE" CPU cores. What is the basis for this number?
Because it's a power of two, that's all there is to it. It's an aesthetic. In the world of computers, the power of two is the fundamental principle - there's no other way. Actually, in the course of development, there's this one occasion when we had an all-night, intense discussion in a U.S. hotel. The IBM team proposed to make it six. But my answer was simple - "the power of two." As a result of insisting on this aesthetic, the chip size ended up being 221mm2, which actually was not desirable for manufacturing.
In terms of the one-shot exposure area, a size under 185 mm2 was preferable. I knew being oversized meant twice the labor, but I on the other hand, I thought these problems of chip size and costs would eventually be cleared as we go along. But in this challenge of changing the history of computing, I could not possibly accept any deviation from the rule of the power of two. For example, the world of communication also has gone to the rule of the power of ten. Ethernet, which started with 10M bit/s, has gone through stages of 100M and 1G, and 10G is certain to come next. You won't go with, say, 4G just because 10G is technologically difficult. It is my belief that real technological innovation is born from such persistence.
To be continued.