UCS text can be encoded in a number of different ways. The most basic method is to represent each character as a two-byte integer equal to its assigned UCS value. This form of encoding is known as UCS-2, and is what OS/2 Universal Language Support uses for Unicode text processing.

Using UCS-2, up to 65,536 unique characters may be represented. This range corresponds to the UCS Basic Multilingual Plane.

Note: Unicode values beyond this range cannot be represented using UCS-2 encoding. Although this effectively limits OS/2 to the 65,536 characters within the BMP, this should be sufficient for comprehensive character support for almost any modern language.

ULS represents a Unicode character using the UniChar data type, which is a two-byte integer value:

    typedef unsigned short UniChar;

A UniChar is also sometimes referred to as a "UCS code element".