The Wire That Changed How I Think About Time in Computers

Most people have no idea what time actually means inside a computer. I certainly didn't until recently.

A while back, I walked into an Ace Hardware and asked for exactly 11.8 inches of wire. The guy behind the counter blinked. "We only sell by the foot." I told him I needed exactly 11.8 inches. He looked confused, but eventually measured it out and handed it over.

Why 11.8 inches? Because that is how far light travels in a vacuum in a single nanosecond. In copper wire, it is a little less, but close enough. That length represents one nanosecond of travel time. One billionth of a second. It is hard to imagine something that small, which is the point.

I got the idea from Professor Brian Stuart, one of the best teachers I have had. He had mentioned how Grace Hopper used to hand out wires this exact length to help people physically grasp what a nanosecond actually is. She would show up to talks with a handful of them and pass them out so people could take home a billionth of a second.

You can say a nanosecond is one billionth of a second, but that does not mean much until you are holding one.

Once you have that frame of reference, the other units start to make a little more sense. A microsecond is a thousand nanoseconds. That is over one thousand feet. If a nanosecond is under a foot of wire, a microsecond is three football fields. You can actually buy a 1000 foot spool of 22 gauge wire on Amazon for about one hundred dollars. That is one microsecond.

A millisecond? Multiply that again. That is over 300 football fields. One second is a million microseconds. One billion nanoseconds.

The real takeaway here is that different parts of the computer operate on entirely different time scales, and most people treat them like they are the same. They are not.

The CPU internally works on timescales smaller than a nanosecond. Reading a register, making a branch decision, even something like pushing a value onto the stack — these are all happening in under a nanosecond. When the CPU accesses memory, that takes a few nanoseconds. Still fast, but noticeably slower. And if it needs to talk to the disk, that is measured in milliseconds. Not micro. Milli. That is not just slower. That is a completely different universe.

This is why pretending that disk can be used as memory is such a bad idea. Memory and disk are not interchangeable. You cannot swap between them and expect things to work the same. You are comparing feet of wire to football fields. You are giving up billions of potential CPU actions every time you wait on the disk.

The wire helped me understand that. We are terrible at imagining time at this scale. Our brains are not built for it. We cannot see or feel a nanosecond. But we can hold 11.8 inches of wire. That gives it weight. It makes it real.

Ever since then, I think differently about how my code uses resources. When something wastes CPU cycles or stalls the pipeline or hits the disk unnecessarily, I think about that wire. I think about how much time I am really wasting.

We live in a world built on abstraction. But underneath every abstraction is physics. Real wires. Real voltages. Real timing. And those layers matter more than we think.