WASHINGTON -- The Federal Communications Commission announced last week that it would begin to test six rival high-definition television systems in April, but the competition has been thrown into turmoil.
Stunned in June by the surprise entry of an all-digital system proposed by General Instrument Corp., a maker of signal-scrambling equipment, three of the leading rivals are rushing to overhaul their proposals in time to meet the agency's deadlines.
The last-minute efforts mark a significant upheaval in the race to develop a technical standard for high-definition television, or HDTV.
Until several months ago, all of the systems submitted to the FCC would have transmitted the advanced television signals largely as analog waveforms, much as television is broadcast today.
In June, however, General Instrument announced that it had developed an all-digital system. Digital broadcasting transmits programming entirely in the 1's and 0's of computer code.
It would set the stage for the development of televisions that function like computer workstations, capable of storing, retrieving and manipulating video material.
As recently as a year ago, the conventional wisdom among broadcasters was that fully digital television could not be developed soon enough to compete in the global race to develop high-definition television.
Indeed, advocates of fully digital technology say they were routinely dismissed.
That is no longer true.
On Tuesday, a consortium formed by NBC, Thomson SA of France, Philips NV of the Netherlands and the David Sarnoff Research Center in Princeton, N.J., announced that it had made two key advances and that it planned to replace its current proposal with a fully digital HDTV system.
Similar optimism was voiced by the head of the Massachusetts Institute of Technology's advanced television research program, another rival in the race.
John I. Taylor, a spokesman for Zenith Electronics Corp. was more cautious, saying only that it was "possible" his company would replace its proposal with an all-digital system.
Indeed, the only competitor that ruled out replacing its existing proposal was NHK, Japan's national broadcasting company.
"The whole world is going digital," said D. Joseph Donahue, senior vice president for technology at Thomson Consumer Electronics.
"It can be done," agreed Professor Jae S. Lim, director of MIT's research program, which is proposing a "hybrid" high-definition system that uses both analog and digital processing.
Mr. Lim, who had been skeptical about a fully digital approach as recently as six months ago, said his laboratory had developed computer simulations that changed his mind.
"We have been making rapid progress in the past two to three months," he said. "Lab studies show, in fairly impressive detail, that the technology is here."
The goal of all high-definition television systems is to produce pictures that have the crispness of film, the sound quality of compact discs and the wide-screen dimensions of movies.
The key question is whether the different competitors can produce a new system in time to meet government deadlines.
Determined to adopt a system before the United States is eclipsed by programs in Japan and Europe, the FCC chairman, Alfred C. Sikes, is insisting that advanced-television proponents meet a tight schedule under which government-sanctioned laboratory tests would be completed by the spring of 1992 and a final decision would be made by June 1993.
The system that is selected for the standard would have the right to the underlying broadcasting technology.
Even tougher requirements face companies that want to change their proposals now.
Richard E. Wiley, chairman of the FCC's advisory committee on advanced television, said last week that companies would have to advise him formally of any basic changes by Dec. 31, just seven weeks from now, and they would have to supply detailed technical information by Feb. 28.
At this point, only General Instrument has produced a detailed all-digital proposal. Officials at the Advanced Television Research Consortium, the venture headed by Philips and Sarnoff, which has also proposed an enhanced-definition system, acknowledged they had not yet come up with a final proposal.
But they said they had developed a new modulation technique that allows digital information to be transmitted over the airwaves at low power levels, so the signals do not interfere with adjacent television channels.
In addition, the consortium said it had developed a technique for compressing the huge amount of data in a high-definition picture into a signal that can fit onto a standard television channel.
MIT, meanwhile, said it had completed successful computer simulations for an all-digital system of its own. But the university needs to find additional financing, probably from a corporate partner, to develop a system that can be tested.
Zenith officials were more cautious about their effort to develop a fully digital plan. "There are serious technical challenges that need to be addressed," said Mr. Taylor, the Zenith spokesman.
But Mr. Taylor said it was "possible" the company would propose a fully digital approach.
General Instrument has already demonstrated its system in public, using computer simulations, and it has generally impressed most experts who have viewed it.
But skeptics, including some government officials, are concerned that the signal cannot be transmitted for more than about 10 miles without becoming garbled.
Television signals today must be able to reach 80 to 100 miles.
By all accounts, however, General Instrument has forever changed the race.
"The debate is over," said Nicholas Negroponte, director of MIT's media laboratory and a champion of end-to-end digital systems. "In the United States, every serious thinker I've encountered believes the future of television is digital."