Introduction to Computing

Performance in Computing

Computer components are operated based on periodical clock signals that are generated by natural oscillation of various crystals when subjected to electricity. In each signal cycle, the chip components move from one state to the other determined by the current input, producing some other output. There may be logical gates embedded in the chip that combines binary inputs 0 or 1 and produce another binary output depending on its purpose. Binary inputs are generally fed into the chips as different levels of voltage. Each binary value used in computation is named as a binary digit (bit).


A simple logical OR gate may receive multiple binary values as input and produce 1 as output when at least one of the input values are 1. On the other hand, a logical AND gate will produce 1 as output when all of the input values are 1. There are multiple such basic building blocks in computer chips, e.g. XOR, NOR, NAND, XNOR, etc.


The capacity of a chip is basically determined by the speed that the internal gates can change their states by releasing their output to the destination and receiving new input. Although the amount of time required to do such operation is only a tiny fraction of a second, each calculation may require billions of such state transitions to generate the expected result to the end user.


Most of the modern end user computers are built based on 64-bit architecture where the CPU(s) can process 64 bits of information in any clock cycle. The bus (i.e. the communication link between the CPU and RAM) is also 64 bits. There are higher capacity computers, and other basic computer architectures still in use today that can process number of bits simultaneously, e.g. 32 bits, 16 bits, and 8 bits computers.


Computers built on multiple CPU cores are also capable of executing instructions from multiple programs within the same clock cycle using the concept of multi-threading.