NOT so long ago, the quaint New England towns around Boston bubbled with high-tech entrepreneurs and economic success.Officials from all over the world came to witness the miracle. The region, along with California's Silicon Valley, became the shining example of how entrepreneurs could make the United States competitive again. Today, the Massachusetts Miracle is in tatters. High-tech employment has fallen to its lowest level in nearly a decade. The New England economy has been hit by deep recession. While local entrepreneurs continue to create new technologies, the hopes that they will spark another high-tech boom glow only dimly here. "I do not see coming fast the industries that are going to provide the replacement of manufacturing jobs," says Ed Roberts, a professor of technology management at the Massachusetts Institute of Technology (MIT) in Cambridge, Mass. The main problem is that the economic boom along Route 128 (the highway that rings Boston on three sides) was built on a technology that became outdated in the 1980s. "This has traditionally been the minicomputer capital of the world - and no one wants to buy minicomputers anymore," says Richard Fiorentino, president of Wavetracer Inc., a supercomputer maker in Acton, Mass. Several of Route 128's biggest minicomputer makers, such as Digital Equipment Corporation and Wang Laboratories, have lost business to makers of smaller and cheaper desktop computers. Their failure to keep up with the market has exacted a heavy toll. Several area computermakers have gone under or been bought out. Digital is in reorganizing and shrinking its operations. Wang has virtually abandoned hardware to become a vendor of software. The restructuring on Route 128 has rippled throughout the state's economy. High-tech manufacturing jobs, which peaked at 261,300 in 1984, had by 1989 fallen to 229,900 - below the level of 1980. The number continues to fall. Nonmanufacturing high-tech industries have been more resilient. But they too saw the number of jobs fall between 1989 and 1990. Route 128 was as much victim as cause of the economic debacle. Nationally, the computer business has been in a slump for more than a year. The Federal Reserve Bank of Boston has called New England's recession the worst since the Great Depression. Total nonfarm employment in Massachusetts fell 4.1 percent last year. Prospects for a recovery have brightened a bit, economists say. "We think we're probably bottoming out or close to it," says Elliot Winer, chief of economic research at the Massachusetts Department of Employment and Training. But "we're not looking for any rapid recovery." For all its problems, entrepreneurship continues to thrive here, Professor Roberts says. In converted warehouses around Cambridge (the home of MIT) and elsewhere young entrepreneurs are busy creating the next generation of technological advances: new computers, telecommunications, advanced materials, biotechnology, and software. If there is a bright spot in New England's high-tech industries, it is software. A Coopers & Lybrand survey found that chief executive officers of software companies in Massachusetts make substantially more annually than their counterparts on the West Coast ($145,000 versus $85,000) and get bigger annual raises (10 percent versus 5 percent). "I think this is going to be the software mecca for the entire world by the end of this decade," forecasts George Colony, president of Forrester Research Inc., a technology research firm in Cambridge. The shift is already under way. The debacle in Route 128 has made venture capitalists very leery of investing in hardware-intensive startups. Companies that build their own hardware components are frowned upon. "If you have developed a proprietary system [with a hardware component], you almost have to apologize," says Desmond Pieri, marketing director of Cambridge-based Voice Processing Corporation, which makes its own computer board for its voice-recognition software. "I think it is collective hallucination to think that Massachusetts is attractive to companies establishing a labor- and capital-intensive industry here," says Jim Manzi, the head of Lotus Development Corporation, a large software company. But "we will create all sorts of knowledge jobs." These new knowledge-industries will probably not create great numbers of manufacturing jobs, as the minicomputer boom did. At Digital, for example, software sales account for such a goodly chunk of business they would rival those of some major software companies. And how many people will a biotechnology firm need to make a new drug that sells for $10,000 per capsule, Roberts asks. "It could be a somewhat disturbing economy," Colony adds. There will probably be very little room for blue-collar people and manufacturing jobs. "The economy is going to be defined by 40 smart guys in an office building." In suburban Acton, Mr. Fiorentino and some 30 men and women are busy creating and marketing massively parallel supercomputers. In its three years of existence, Wavetracer has pioneered the linking up of thousands of processors in a single machine. The result is a very fast computer that can solve difficult problems at a fraction of what it would cost to buy a single-processor supercomputer. In a very rapid two years since it received its first round of financing, Wavetracer managed to build and sell its first machine. "We're at least a year away from profitability," Fiorentino says. Here, as elsewhere, software has become at least as important as hardware. Wavetracer uses older 8-megahertz processors in its supercomputer, relying on its software to link the processors and create the needed speed.