the design and internal structure of digital computers.

Fundamentally, a computer is a machine that can store instructions and execute them. Thus, it consists of two major parts, memory and the central processing unit (CPU), which communicate through a set of parallel electrical connections called the bus (Figure 65). The bus also connects to input-output devices such as a screen, a keyboard, and disk drives.

The CPU spends its time retrieving instructions from memory and doing whatever those instructions say. Each instruction is a pattern of bits (binary ones and zeroes, represented by electrical on and off signals). When the instruction reaches the CPU, the CPU must decode recognize) it and activate the appropriate functional unit within the CPU in order to carry out the instruction. Functional units include adders, multipliers, circuits to compare bit patterns, etc., all of which are built from logic gates (for an example, see binary addition).

The CPU contains registersto hold data that is being worked on. For example, in order to add two numbers, the CPU will typically retrieve the two numbers from memory into registers, perform the addition, place the result in another register, and finally store it back into memory. Figure 66 illustrates the parts of the CPU. In most presentday computers, the CPU is a single integrated circuit (IC) called a microprocessor.

Every location in memory has an address (i.e., a bit pattern, binary number, that identifies the location).To retrieve the contents of memory location 011000011, the CPU places the bit pattern 011000011 on the address portion of the bus, activates the "read memory" line, and waits a specified length of time. The memory places the contents of that location onto the data portion of the bus so that the CPU can read it. To put data into memory, the CPU puts both the address and the data onto the bus and activates the "write memory" line. Some computers also include "read port" and "write port" lines, which are like the lines used for accessing memory except that addresses are understood as applying to input and output devices (printer ports, etc.) rather than memory.

Most computers use a Von Neumann architecture, which means that programs and data are stored in the same kind of memory. Some microcontrollers use a Harvard architecture, with separate memories for program and data (mainly because programs are kept permanently recorded in ROM, but data must be changeable).

Programmers normally do not write CPU instructions. Instead, they write programs in a high-level language such as BASIC, C, or Pascal, and use a compilerto translate the programs into machine language. It is also possible to write programs in assembly language which translates into machine language more directly.

intermediate-level computer language that is less complex to use than a machine language. Assembly languages use abbreviations or mnemonic codes to replace the 0s and 1s of machine language (A for "add," C for "compare," and MP for "multiply"). A translator is required to convert the assembly language program into machine language that can be executed by the computer. This translator is the assembly program. Every command in assembly language has a corresponding command in machine language. The assembly language differs among computers, and thus these programs are not easily transferable to machines of a different type from the one on which they were written.