IBM Carbon Nanotubes May Boost CPU Performance Tenfold

Moore's Law is one of the great old laws that is frequently quoted yet sometimes misapplied and occasionally misunderstood. Even Moore himself said that his law could not continue indefinitely, that eventually, limits would be reach on the overall ability to miniaturize transistors. But the law that claims, essentially, that computing power doubles roughly every two years, is about to get a little extra help from IBM (News - Alert) and a recently demonstrated new line of carbon nanotubes.

IBM's newest demonstration, which featured a new way to place the carbon nanotubes, was set to not only make processors significantly smaller, it was also set to make them significantly more powerful. While most current desktop PCs pack a 22nm chip, with 2014's Broadwell CPU from Intel (News - Alert) set to drop that number down to 14nm, the carbon nanotube chips from IBM dropped that clear down to just 10nm, with a set of possible solutions to explore to cut even that tiny number in half and drop it down to a mere five.

The problem, as mentioned earlier, was that eventually the limits of physics as we knew them, are going to get in the way and limit the size to which a chip could be shrunk in the first place. But the introduction of the carbon nanotubes would force back the laws of physics, at least for a time, and allow for the construction of chips that offered between five and 10 times the performance of current-model silicon chips.

Restraining that, however, is the comparative difficulty involved in producing the carbon nanotubes, which were generally only manageable in a laboratory setting and limited to a couple hundred being able to be placed at once. Considering that Intel's line of Ivy Bridge processors were putting 1.4 billion transistors on a chip that required processor makers to think bigger. IBM's newest model allows for just over 10,000 carbon nanotubes to be placed; a big step to be sure, but still nowhere near ready for prime time. However, the gain also shows a potential new way to make more carbon nanotubes, and thus shore up the numbers sufficient to make them the replacement of the not too distant future.

Getting the production methods in place for the chips of tomorrow is no easy feat, but it requires careful consideration and plenty of experimenting. Improving chip power and capability, while also reducing size, yields a lot of benefits that can actually be seen currently. Consider the rise of the smartphone; would this have been possible without improvements in processor capability? Everyone who uses a computer--or anything that comes in contact with one--gets a firsthand look at the sheer value inherent in improved processors.

IBM's method may well bring about the processors of both the near future and the future for some time to come, but it will be a while before we see this start to hit desktop rigs and other systems. Still, carbon nanotubes are likely to be a major part of the computing landscape, and fairly soon.