February 8, 2005

Commentary: Cell chip story may be a window into future of IT

Author: Chris Preimesberger

We're barely into 2005, and we're already seeing another landmark innovation in information technology: the new Cell chip, which could revolutionize our entire industry with its tremendous clock speeds of more than 4 GHz. It's been in R&D since 2001 and was developed by a consortium of companies:
IBM, Sony, and Toshiba.
The new chip, designed to power the upcoming PlayStation 3, integrates 234 million transistors and is fabricated with 90-nanometer SOI technology. Theoretically, it will be able to handle 16
The new Cell chip, aside thumb tacks.

trillion floating point operations per second. The Cell's multi-core architecture and ultra high-speed communications capabilities enable a vastly improved real-time performance -- often 10 times the speed of the newest PC processors.

Yes, these three conglomerates normally do compete with each other. So why did they hold hands to put this new superchip together? Were they trying to gang up on the establishment -- Intel, AMD, and the like? Possibly, but there's a larger vision involved here, I believe.

I wasn't around in the 1940s, but I am a student of history. This unusual IT co-op reminds me of World War II, when several ultra-competitive aircraft companies collaborated to build a long-range bomber that could perform as well at sea level as it could at 28,000 feet. Because the war effort required team players in a big way, the four companies (Consolidated [originator of the aircraft and based in San Diego, Calif.], Ford Aircraft [Willow Run, Mich.], Douglas [Tulsa, Okla.], and North American [Dallas]) decided the bigger motivation -- winning the war -- was far more important than individual company profits. So they worked together to produce the classic, twin-tailed B-24 Liberator, the four-engine bomber that went on to knock out the Ploesti oil fields in then-Rumania, the daring Aug. 1, 1943 low-level attack that crippled the Nazi war effort in a big way and probably shortened the war by a year. That aircraft, along with the B-17 bomber and the P-51 fighter, became the key military air tools in defeating the Axis.

The Consolidated B-24 bomber, c. 1943.

A total of 19,256 B-24 bombers of all variants eventually would be produced -- more than any other U.S. warplane of any era. This was truly an open source effort; the B-24 plans and materials were shared among all participating companies in order to build a superior tool for the common good. Proprietary gain was secondary; all hands worked toward the common goal.

I'm not saying that this is a war, or that IBM, Sony, and Toshiba aren't looking for profits in this Cell chip venture. Of course they are, and they will eventually profit big time. The aircraft manufacturers also made their fair share of money from government defense contracts. But a big upgrade was overdue in the printed circuit business -- just as in the bomber-building business in the '40s. The Cell chip may be that innovation, and competing businesses shared the specs to build it. Like the B-24, it took more than one company to finish the job.

The spirit of corporate cooperation was the key in both instances. All parties benefited in the long run.

Is this a peek into the future? As we move on, designing and building things that work and scale well get more complicated and take more human resources. Will it take new cooperative efforts by competing IT companies in other areas (your market name here) to really do the job well? Quite possibly. And will the open source model work -- possibly in amended ways -- to help get these important quantum leaps done?

You can bet on it.

Click Here!