What is the difference between interpretation and compilation in computer programming?

Answer 1

Interpretive languages compile blocks of code into machine code, execute them, and then move onto the next block. The blocks may be as little as a single statement, but once each statement is executed, the machine code is lost. This means that functions must be recompiled every time they are called. Ultimately, performance suffers because every block must be interpreted before it can be executed. Moreover, the program must always be executed within the interpreter; you cannot create standalone programs.

Compiled languages pre-compile the entire program to produce a standalone machine code executable. As a result, compiled programs execute many times faster than interpretive languages.

C++ and Java are both compiled languages however Java programs are compiled to byte code rather than machine code. The byte code must then be interpreted by the Java virtual machine in order to execute. As a result of this interpretation, Java programs do not perform well compared to equivalent programs written in C++.

Answer 2

Interpretation essentially requires each line of code to be converted into machine code and executed, one at a time. This means the program can only be run within a suitable interpreter -- it cannot produce standalone executables.

Compilation (and linking) produces machine code for the entire program, which can then be run without any further interpretation. The program is completely standalone.