But despite IBM's stated plans to bring copper-wired chips to market this year, analysts say the technological and financial benefits of the new technology for IBM and the semiconductor industry are years away.
"IBM is talking about a few months away, but I don't think there's going to be any significant volume production until 1999," said A.G. Edwards analyst Chris Chaney. "This announcement has been way overhyped."
The use of copper will enable semiconductor makers to continue to shrink the chip's circuitry to widths previously impractical with aluminum, which becomes too slow as it gets thinner. Other companies have been testing non-aluminum conductors--including copper, gold, and silver--for some time, but IBM is the first to announce plans to deliver products based on copper circuitry.
The new technology is expected to yield faster and smaller chips that consume less energy and produce less heat than aluminum ones. Financial rewards for chip companies include greater production with more chips fitting on production lines and stepped-up demand for higher-performance chips.
IBM shares jumped up 4.7 percent to close the day at 103-7/8, up 4-5/8 from Friday.
But along with the benefits come costs. Factories will have to be retooled to produce chips with copper, a process complicated by copper's tendency to oxidize and to corrupt silicon.
"Initially it's going to cost a lot of money. It's not going to be very profitable at first," said Chaney. "But if they can keep their volumes up and their yields to match, they will have more output and cheaper costs."
But those costs will be "probably negligible," according to Peter Glaskowsky, senior analyst for the Microprocessor Report.
Industry analysts predict that the widespread use of copper is still years away.
"You won't see everyone instantly switching over," said Glaskowsky. "This is just part of a steady increase in technology over time."
Intel spokesman Adam Grossberg said his company was planning to move to copper within two generations, or as the company's circuits shrink from the current .25 micron linewidths to .18 microns and eventually to .13 microns.
[Intel is an investor in CNET: The Computer Network.]
"Copper is not the only way to do this, but we will most likely make the transfer over to copper," said Grossberg. "At this moment we don't see the need to make the switch."
Is Intel concerned that IBM got there first?
"No," said Grossberg. "It's a good announcement."