Keeping up with Moore’s Law is getting harder
While the blistering pace of technology change has kept pace with Moore’s law, the end of Moore’s milestones may be inevitable as chip geometries continue to shrink exponentially.
Moore’s law, named after Intel co-founder Gordon Moore, is the theory that the number of transistors on integrated circuits doubles about every two years. This law has proved true since the invention of the integrated circuit in 1958, but may face an end in about 15 years.
“Moore’s Law is coming to an end — in the next decade it will pretty much come to an end so we have 15 years or so,” said Henry Samueli, chief technology officer of Broadcom Corp. “Standard CMOS silicon transistors will stop scaling around 5 nm and everything will plateau. I am comfortable we will get to terabit networking speeds, but I’m not sure I see a path to petabit speeds.”
Chipmaker AMD said its delay in transitioning from 28nm to 20nm chips reveals the impending end of Moore’s Law. “I’m saying you are seeing the beginning of the end of Moore’s law,” said John Gustafson, chief graphics product architect at AMD. “We want to also look for the sweet spot, because if you print too few transistors your chip will cost too much per transistor and if you put too many it will cost too much per transistor.”
Intel has also said that keeping up with Moore’s law will be much more challenging in the years to come, since manufacturing smaller chips with more features makes them open to a “wider class of defects.” The sensitivities and minor variations increase and much more attention to detail is required.
Intel has been at the leading edge with Moore’s law, adding strained silicon on the 90-nanometer and 65-nanometer processes to improve transistor performance. Intel also changed the way transistors are with its latest 22-nm chips, with transistors placed on top of each other to give it a 3D design.
“Are we closer to an end than we were five years ago? Of course,” said William Holt, executive vice president and general manager of Intel’s Technology Manufacturing Group. “But are we to the point where we can realistically predict that end, we don’t think so. We are confident that we are going to continue to provide the basic building blocks that allow improvements in electronic devices.”
Research into new ways to scale down chips is funded by chip manufacturers and is being conducted at universities, Holt said. Some of the research revolves around new transistor structures and new materials to replace silicon.
The U.S. government’s National Science Foundation is funding the research “Science and Engineering behind Moore’s Law,” which includes research on manufacturing, nanotechnology, multicore chips and emerging technologies like quantum computing.
The next big frontier for chip manufacturers is 450-mm wafers that allow more chips to be made at a cheaper cost, reported Computerworld. Intel invested $2.1 billion last year in the tool maker ASML, in order to enable smaller chip circuits and larger wafers. Samsung and Taiwan Semiconductor Manufacturing Co also invested in ASML.
Yet Holt said he couldn’t predict when Intel would create 450-mm wafers but is hoping it will be by the end of the decade.Tags: Productivity,Technology