Divisions of a Second
ACCURACY
If the three functions named above are all computers can do, then why have these machines become so widely used? Computers are useful because they are very fast, their circuits are reliable, and they can store vast amounts of data. In other words, computers are used because of their speed, accuracy, and storage capabilities.
The speed of a computer is controlled by a number of physical factors. First, the switching speed of its electronic components, such as switching the state from on to off or switching the direction the current travels, affects the speed of the computer. A second factor is the distances that electric currents must travel within the circuits. By packing circuits closer together and increasing the switching speed, engineers have been able to increase the speed of computers vastly. Other factors that affect computer speeds are the programming language (that is, type of code) used in writing programs, the amount of data a computer can handle at one time, and the amount of data and instructions readily available in the computer's memory.
Modern computers can perform millions of
instructions per second. This is why computer speeds are sometimes
discussed in terms of millions of instructions per second (or MIPS).
Their speed is fast reaching the physical limitation of the speed of
light, 186,000 miles per second. Computer speed is also measured in
terms of nanoseconds and other small units (see Figure below).
Today's processors can perform from 35 million to 150 million
instructions per second.
Divisions of a Second
ACCURACY
The accuracy of a computer refers to the inherent
reliability of its electronic components. The same type of current
passed through the same circuits yields the same results each time.
We take advantage of this aspect of circuitry every time we switch on
an electric device. When we turn on a light switch a light comes on,
not a radio or a fan. The computer is reliable for the same
reason--its circuitry is reliable. A computer can run for hours,
days, and weeks at a time, giving accurate results for millions of
activities. Of course, if the data or programs submitted to the
computer are faulty, the computer will not produce correct results.
The output will be useless and meaningless, illustrating the human
error involved. This concept is called garbage in-garbage out
(GIGO) and is fundamental in understanding computer
''mistakes."
Besides being very fast and reliable, computers can store large amounts of data. Some data is held in memory for use during immediate operations. The amount of data held in memory varies among computers. Some small computers hold as few as 16,000 characters, whereas large computers can hold billions of characters. Data can also be recorded on magnetic disks and tapes (see Figure below); this storage makes a computer's "memory" almost limitless. Storage holds data that is not im-mediately needed by the computer.
Tape Storage
The ability of the computer to store, retrieve, and process data--all without human intervention--separates it from a simple calculator and gives it its power and appeal to humans. So while humans can perform the same functions as the computer, the difference is that the computer can reliably execute millions of instructions in a second and store the results in an almost unlimited memory.
