I have a fairly strong understanding of how code like C is compiled into machine code and how transistors form gates which can be put together to compute lots of operations which can eventually become a computer. Say you have a machine instruction that says to add the contents of register 3 and register 5 and put the result in register 6. What is physically putting the voltages on the transistors in a way that achieves the desired result given that nothing is flipping a mechanical switch?
In: 2
Transistors are like a three-layer semiconductor cake. Between two layers of one type of silicon is a “doped” layer that responds to external electrical excitation by admiting electron flow between the other two layers.
So when a transistor is connected across a voltage potential, no current flows. But if you apply a voltage to its third pin, current flows in proportion to the voltage applied to this third pin. Integrated circuits such as your CPU scale this component down to nanometer scale and repeat it millions of times to produce the transistor networks that machine languages can operate, and which assembly languages can control.
Quite literally, an electric signal directly determines if a transistor will pass current or not, which is super convenient, especially in computers where it’s function depends entirely on precisely organizing and controlling electrical signals.
Many types of transistors exist, some of which operate a little differently than I describe. But the essential point is that transistors let an electric signal control the flow of another electric signal.
Latest Answers