Playstation creator courts serious fun
CMP Media Inc
Mar 07, 1:07 am
Mar. 07, 1999 (Electronic Engineering Times - CMP via COMTEX) -- After nearly 25 years of hard work that included a product flop, a failed alliance and plenty of grunt-level design engineering, Ken Kutaragi has become something of a celebrity in Silicon Valley. Semiconductor executives and reporters hang on the words of the man who began his career as a rank-and-file engineer in Tokyo. They're eager to find out what move he'll make next with his Sony Playstation, the videogame machine that has sold an extraordinary 50 million units since it was launched in 1994, and those same industry watchers are now waiting for a much-anticipated refresh.
"He's had more impact on my career than any other customer I've ever known," said Brian Halla, chief executive officer of National Semiconductor Corp., who as head of the consumer division of LSI Logic helped broker the deal that gave LSI the design win for a chip set inside the original Playstation. "I tell him I have a piece of my home named the Kutaragi wing."
Not only did the landmark design win give the company's system-chip strategy a shot in the arm (along with its stock price, which jumped from 4 7/8 to 126, largely on the strength of the deal), it also colored Halla's goals for National in the PC and set-top world. "Everything I am trying to do at National stemmed from the [Playstation project]," he said. As proof, he pointed to National's own single-chip offering, the MediaPC, which it plans to roll out in June for set-top boxes.
The silicon behind the next generation of videogame consoles is having a similar impact on Toshiba Corp., which has embarked on a new media-processor design as well as a 128-bit embedded CPU family based on its work on the next Playstation processor. "We learned a lot from this project," said Mitsuo Saito, general manager of Toshiba's system ULSI engineering lab in Kawasaki, Japan.
But Kutaragi's work did not always generate such high-profile praise. Long before he became chairman of Sony Computer Entertainment America, Kutaragi's first design efforts were as obscure as one could imagine-developing dot-matrix LCDs and code for stripped-down controllers.
The 80 x 100 matrix LCD Kutaragi designed as a young engineer at Sony never got off the ground. "It was too early," he said. But the assemblers, debuggers and compilers he helped write for proprietary 4- and 8-bit Sony controllers was more of a success, albeit only inside the walls of Sony's engineering labs.
Kutaragi describes those processors as something less ambitious than a full-blown microcontroller, since they were streamlined and optimized around specific Sony systems such as tape decks and videotape players. The code itself ran on the CP/M operating system. "There were no Microsoft tools," he said. "I realized it was important to have integrated tools to help optimize those cores."
From that small success, Kutaragi went on to become one of a handful of project leaders on Sony's analog Mavica camera. He helped develop a 2-inch floppy disk that the camera used for storage. It rotated at a then-unheard-of 3,600 rpm. "It was similar in bit density and rotation speed to a hard-disk drive," he said.
The microfloppy was used in the analog Mavica and a few word processors of the day, but ultimately flopped as the 3.5-inch floppy became a standard. Nevertheless, Kutaragi said he learned a great deal about error-correction coding and modulation techniques from the project. "It was a very good experience that helped me learn about signal processing."
Kutaragi found his calling one day when he bought one of the first 8-bit Nintendo NES videogame consoles while working at a Sony information-systems research center, where he was one of about 100 digital engineers. "I was really impressed by this machine because it was totally program-driven," Kutaragi said. "The graphics were very sophisticated if you compared it with one of the TI computers of that time . . . but the sound was terrible. There was no frequency flexibility. It was just one or zero. I was frustrated that such a nice machine had such horrible sound."
Out of that frustration an opportunity was born that ultimately led Kutaragi to his current spot as "the father" of the Playstation. Kutaragi and a Sony salesperson met with executives from Nintendo to propose Sony apply its signal-processing prowess to Nintendo's next-generation console. The two companies quickly struck a deal.
"We designed a small chip and made an offer to Nintendo, and they picked it up in their 16-bit system, the Super NES, which offered PCM [pulse code modulation] audio," said Kutaragi. The work gave birth to a small team of about five designers, including Masakazu Suzuoki, who later became the core of the Playstation team.
"We realized that this was a nice growth area for us in digital entertainment, and driven by the evolution in semiconductors-Moore's Law-there would be a new level in entertainment," he said. Indeed, emboldened by his success, Kutaragi made another proposal to Nintendo in 1989: the two should work toward developing the first CD-ROM-based console. A year later, Nintendo agreed and the two partners were off to the races.
'Beat Nintendo'
"But at Summer CES [Consumer Electronics Show] in 1991, I had a surprise when Nintendo announced a realignment, and said they would work with Philips on the console and stop our project," Kutaragi recalled. "Our engineers had a good relationship [with theirs], but management decided to go another way."
Ironically, the Nintendo/ Philips console never got off the ground, but the deal had an unintended effect. "We decided to start our own [console] development, and we gathered up a team at Sony to create a new gaming system to beat Nintendo," said Kutaragi.
Sony's ambitious goal in May 1992 was to create the first CD-ROM console with real-time computer graphics powered by a 1-million-transistor system-on-chip. Kutaragi talked to every semiconductor company that would accept a meeting to find a partner for his plans. Some weren't interested, others said it couldn't be done. Ultimately, Halla seized the opportunity for LSI Logic and helped deliver the MIPS-based chip.
"Almost every night for two-and-a-half years we had conference calls on the project, with Kutaragi in attendance at most of them, mainly to go over engineering trade-offs," Halla said. "The rest of the meetings were highly animated philosophical discussions about pricing."
For its part, Sony took on the challenge of creating its own RTOS environment, libraries and third-party tools for game-title developers. Convincing title developers they should stop writing to low-level hardware registers where performance advantages can be gained and focus instead on delivering content quickly with the Sony libraries and tools was a tough job-one Microsoft Corp. still faces today with its Windows DirectX application programming interfaces.
Ultimately, Sony was able to roll out the tools in early 1994 and ship the console in Japan later that year. As many as 50 million Playstation consoles have been sold to date and 2,000 titles are available for the machine, reports Kutaragi. Still, since 1996 he has moved on to work on the Playstation's successor.
At the International Solid-State Circuits Conference in February, Kutaragi's colleague Suzuoki described a 128-bit media processor expected to form the silicon heart of that system (see Feb 22, page 4). At 10 million transistors, the chip's integration is an order-of-magnitude beyond that of the existing LSI Logic part.
Toshiba managers who participated in the project claimed they have learned enough from their work to roll a new line of 128-bit embedded processors for networking and a new media-processor design of their own. Last week, Kutaragi rolled out details of two more pieces of silicon inside the next Playstation-an ambitious graphics chip Sony has designed and an I/O processor from LSI Logic that will ensure the new machine is backward-compatible with the existing one.
The new DVD-based console, which Kutaragi said will hit the Japan market in the coming winter, will not only attack competition from Nintendo and Sega, but also will give the embattled home PC a run for its sub-sub-$1,000 market. However, Kutaragi said the new system will depend on Direct Rambus DRAMs and 0.18-micron process technology, neither of which will be in wide supply until 2000.
"My guess is they will position the Playstation II as being more of an information appliance," said Halla, who has staked his company on just such a vision. "It has to access the Web."
The bold ambition behind the new design is generating plenty of interest around Kutaragi and his Sony Computer Entertainment division. After the ISSCC paper in February, LSI's chief executive officer, Wilf Corrigan, strode through the crowd of engineers to personally congratulate Kutaragi on the new media-processor design. And last year at the Microprocessor Forum, keynoter Halla asked Kutaragi to stand up in the audience and take a bow as a pioneer of the system-chip trend.
Just how the next round of videogame wars will turn out for Sony remains to be seen. But one thing is clear: Kutaragi plans to keep the bar high and ride Moore's Law for all it's worth.
|
