It’s not really a fad as far as Intel is concerned. Here’s why this shit is awesome: the transistor is the fundamental building block for processors in, well, everything. And they’re flat. Really. They are, at the risk of oversimplifying, printed like ink is printed on paper. Processor manufacturers have been able to shrink the transistor while simultaneously increasing the frequency (speed) and decreasing the power usage. A transistor can be as small as 32 nanometers. Intel is trying to push it to 22nm, but there’s a fundamental limit being reached at that size. The transistor is smaller, but you can’t get both higher speeds and use less power at the same time.
By not making the transistor flat and instead making it two intersecting fins can you make the transistor smaller (in terms of length and width, disregarding height) while being able to still reap the benefits of higher speeds (up to 37% faster) and use less power (up to 50%). Faster, lower-power chips that can be implanted in mobile devices. Or your brain. Whichever.
That’s some awesome shit. Speaking of awesome shit, those guys at Ars Technica write some pretty excellent articles:
These are some very significant jumps in performance and efficiency, and they’ll go a long way toward making Intel’s “x86 in smartphones at 22nm” dreams come true. Once again, Intel has proven that its semiconductor manufacturing prowess is without peer in the industry. Whatever you think of Atom vs. ARM, this is a major advance, and it puts Intel even further out ahead of the competition than it was earlier.
Check out the article for a more in-depth (but still accessible!) description, as well as pretty pictures and neat graphs.