For 10 years, Cecil Alford was designing supercomputers here at Georgia Institute of Technology that he could not afford to build. Then came the Strategic Defense Initiative (SDI), aimed at researching a space-based defense against nuclear missiles.
His design, centering around 32 parallel processors, could solve 64 differential equations at once in real time. That means, for example, that the computer could model the bounce and flex of an auto tire speeding along on a bumpy road at the same speed it happened.
But Dr. Alford could buy only six of the 32 processors. Last June, he won a $21.3 million SDI contract. He immediately built the rest of his 32-processor computer, began tying it to three others to make a 128-processor, high-speed powerhouse. He is starting work on a new concept that will eventually put from 800 to 1,000 processors in parallel.
To Alford, SDI means a comfortable budget to do what he was already working on anyway. This is largely what SDI has meant in the academic research community. Few speak of the kind of national mission that inspired young engineers in the early 1960s to land a man on the moon.
To many here at Georgia Tech, which, together with the Massachusetts Institute of Technology and the University of Texas, is among the top universities to win SDI research contracts, ``star wars'', as SDI is popularly known, means a luxuriant new source of money.
But Andrew Zangwill, an associate professor of physics here, has a sharply different view. No pacifist, Dr. Zangwill is a consultant to Lawrence Livermore Laboratory, which conducts nuclear weapons research. He has refused to accept any SDI contracts, because he says the program is wildly impractical, diverting money and defense policy from more achievable propositions.
Zangwill is part of a nationwide group of scientists, largely physicists, that is boycotting SDI research. More than 3,100 senior faculty members have signed onto the boycott, based on the arguments that President Reagan's goal of an umbrella-like defense for the nation is not technically achievable, that it promotes the arms race by encouraging Soviet overkill capacity, and that it makes arms control more difficult.
``It's hard in the academic community to find real advocates of star wars,'' says boycott leader John Kogut, a physics professor at the University of Illinois at Champaign-Urbana. ``More often, they need the money for their research.''
At Georgia Tech, star wars has some strong advocates, and the political issues raised by boycotters have aroused little interest. Not everything proposed in SDI will work, says Robert Cassanova, director of the SDI program at Georgia Tech. The point of research is to understand by the 1990s just what the possibilities of a ballistic-missile defense system are.
The strongest skepticism has centered on the computer software that could handle the immensely complex battle management tasks for an SDI system.
Even Dr. Alford says the large, overall system ``raises a lot of questions about `will it work?' But my program will work.'' Further, he says the supercomputers he is working on could change the way the battle management software problem is handled.
Georgia Tech has won about $25 million in multiyear SDI contracts. Does that attract students to SDI projects? Alford says he does a brisk business in eager, inquiring students: ``They don't see it as SDI research but as computer research. There's nothing else like it on campus.''