Skip to the content.

Forth crash course

If you are new to Forth, it’s good to cover some of the basics of the language before diving into gbforth. This crash course helps you read and follow (on a very basic level) the code in the examples and documentation.


Forth is a simple stack-based language: There is essentially no syntax, just numbers and words. Numbers are literal values that are pushed on the parameter stack. Words are subroutines that operate on this stack. For example, consider the following Forth code:

4 dup *

The system starts interpreting this code. First it pushes 4 to the stack. Upon reading the word dup, the value on top of the stack is duplicated. After that, the word * will pop the top 2 values off the stack, multiply them, and push the result back on the stack.

At the end of this program, the parameter stack will have a single value on it, which is 16.

Defining words

You can also add new words to the Forth dictionary:

: square dup * ;

Once the word : is read, the system parses the next word and uses it as the name for your new definition.

It then switches to compiler mode, which means that it stops interpreting code, but instead add words (and numbers) to the body of the new definition.

After some time the word ; is read. This words ends the definition and switches the system back to interpreter mode. With this new word defined, we can rewrite our earlier example:

: square dup * ;

4 square

This code will have the same effect as before: The parameter stack contains a single 16.

Take note of the concatenative style of Forth: Words are always evaluated from left to right, and concatenation is composition. Any series of words can be moved into a new definition without having to shuffle words around. In practice, this means you can safely replace all occurrences of dup * with square, so simplifying your code is very straightforward!


There are 2 styles of comments: In-line comments that are delimited between ( and ), and comments that start with \ and continue till the end of the line:

: square ( this is a comment ) dup * ;

\ and this too
4 square

Note that this is no special syntax: Both ( and \ are simply words that parse and discard all characters until a ) or line end is reached. Make sure to not forget to follow these words with a space before writing your comment.

It’s common to use in-line comments to document the stack effect of a word:

: square ( n -- n ) dup * ;

Further reading

If you want to learn about Forth on a deeper level, Starting FORTH is a great read!