While IBM has demonstrated the possibility of creating a chip with seven-nanometre transistors – down from the current 14 – it’s still in the advanced research stage and is not consumer ready.
Advertisement
It is more powerful than any chip now in existence, and could be in machines within two years.
The lowered nanometer count is a big deal as in general a smaller “nm” number, means manufacturers can fit more transistors, which are tiny switches that act as the building blocks of processors, onto the chip thus increasing its power and performance.
Creating next-generation non-silicon chips has been an ongoing project for IBM.
The Wall Street Journal notes that IBM announced last year its plans to spend $3 billion on semiconductor research over the next five years, conducting that work mainly at facilities operated by the State University of New York in Albany. ‘The question is not if we will introduce 7 nanometre technology into manufacturing, but rather how, when, and at what cost, ‘ explained John Kelly, senior vice president at IBM Research, at the time.
The development of 7-nanometer transistors will permit the semiconductor industry to continue the trend toward increasing miniaturization and performance of computing chips known as Moore’s Law.
Previous attempts to make a 7nm chip haven’t been efficient enough and have needed too much power to run, the New York Times points out.
These new micro-sized transistors mean that IBM and its partners will be able to produce microprocessors containing 20 billion transistors, boosting processor power and energy efficiency significantly. For instance, companies like Intel boast of manufacturing some of the smallest of chips in the markets.
While semiconductors based on 7nm technology are still a long time off from being manufactured, the breakthrough shows that many technical advances can still be wrung out of existing semiconductor technology. The company is now commercializing 10-nanometer chips.
Advertisement
Intel said earlier this year that it is still researching promising options for 7nm process technologies and, while the firm has also been investigating EUV lithography, senior fellow Mark Bohr indicated that he believed Intel would not need to use this technology to produce 7nm chips.
