The Shrinking Transistor Is Getting Too Small

When researchers first started mucking around with digital computers in the late 1930's and 1940's (yes, that long ago), the "switches" that made up the calculating part of the computer were comprised of vacuum tubes. 

If you are a baby boomer like me - or older - you may remember a time when TVs and audio equipment still had tubes (they would light up like a weak-looking light bulb as they worked). 

The "WITCH" computer, with vacuum tubes

Because of the physical size of the tubes, computers of that era were, well, pretty huge. Room-sized, in many cases - even though they would probably have had the equivalent computing power of a basic pocket calculator.

Transistors helped shrink this down quite a bit; transistors are the "solid state", silicon-based version of the vacuum tube. Now, computers were only the size of a big wardrobe.

Since then, the manufacturing techniques we have available have continued to shrink transistors so that a single consumer-grade computer CPU (like an Intel i7 in your laptop or desktop PC) might have 3/4 of a billion transistors within it. Specialized CPUs can have transistor counts in the tens of billions.

Cool, so we can just keep shrinking everything down and get better and faster, right? Not so fast, skippy. 

The silicon-based fabrications we currently have are so small (with transistor "gates" getting down to around 10 nanometers), we are bumping into the practical limits and starting to encounter quantum effects. That's right, the things are so microscopically small that the quantum world is coming into play.

Sounds a bit grim, but there are things to be done: we can further optimize computer performance with solutions like better artificial intelligence in the programming. We could utilize other materials that avoid some or most of the quantum effects, such as some forms of carbon nanotubes.

Lastly, we might move away from the electronic systems and move to light-based computers, using photons rather than electrons to do the math for us.

What a time to be alive.

No comments :