Lesson 1
Lesson 2
Lesson 3
Lesson 4
FAQ
Assignments
Take Quiz
Links
PowerPoint

The machine language of a computer is actually hardwired into the microprocessor of that machine. The microprocessor at the right is shown about actual size. Yet, it contains many miles of circuitry and millions of transistors and other electrical components (The Intel Pentium IV contains 42,000,000 transistors ). The transistors act as switches which can be set by the program to cause electrical current to flow along different paths thereby affecting the behavior of the machine. These modifications to the circuitry of the microprocessor are the "instructions" that the machine "understands". Since various families of computers use different microprocessors, they use different machine languages. The machine language of a PC is different from that of the Sun workstation which is different from Apple Macintosh. Yet each of these computers is able to execute a program written in C++ or FORTRAN or Java. This is possible because other applications can translate a program written in one of these high-level languages (source code) into the computer's own low-level machine language (object code).

            There are three general categories of translator programs: assemblers, interpreters, and compilers.

            Assembler languages use a set of abbreviated words (mnemonics) to represent the instruction set of the microprocessor. An assembler language program must be developed for each different microprocessor since their instruction sets are different. For example, every microprocessor contains a circuit that connects to a special part of the computer's memory called the accumulator. An assembler language for that microprocessor must then contain a mnemonic that generates the binary code that activates that particular function of the microprocessor. In the simulation program we will use later, that mnemonic is the LOD (pronounced LOAD) instruction. The programmer must memorize the mnemonics that are associated with the particular assembler language for a specific microprocessor. Once a program is written, the assembler translates the mnemonics into the binary instructions that control the microporcessor.

            Interpreters are programming languages that remove the programmer even farther from the microprocessor. Interpreted languages allow the programmer to use his/her own language to write programs. A single line of code may generate many machine language instructions. As you will see later, the single JavaScript statement total = cost + tax; would generate at least three machine language instructions. Interpreters generate the machine language instructions for a statement each time they encounter that statement. If a set of statements is contained in a loop, each statement is reinterpreted each time through the loop.

            Compilers are similar to interpreters in most respects. However, they differ in one important way. Compilers translate the entire program into machine language and then store those instructions in a separate file. Once a program has been compiled without error, it never needs to be translated again (unless the programmer changes the program).

            The following table compares the three types of translators.

Top