code page 437

code page 437 is the original character encoding of the IBM PC, introduced with its first model in 1981. It is an 8-bit extension of the ASCII standard, preserving the standard 128 characters while adding 128 additional graphics characters for creating simple user interface elements. This encoding became the default for English-language versions of MS-DOS and early Microsoft Windows systems, establishing a foundational visual language for the personal computer era. Its distinctive set of box-drawing characters, symbols, and block elements was integral to the appearance of early text mode software and games.

History and development

The encoding was developed by a team at IBM led by engineers for the original IBM Personal Computer, which required a robust system for displaying both text and rudimentary graphics. It was designed to support the capabilities of the Monochrome Display Adapter and Color Graphics Adapter, providing the necessary characters for creating borders, charts, and simple sprites within the constraints of text mode. The selection of characters was influenced by earlier mainframe computer terminals and the need for a versatile set that could serve both business applications and nascent video game development. Its release coincided with the launch of the IBM PC 5150, cementing its role as a core component of the platform's BIOS.

Character set

The upper 128 positions supplement the standard ASCII set with a diverse collection of glyphs. These include an extensive array of box-drawing characters for creating lines and corners, which were essential for designing text user interface screens in programs like Norton Utilities. The set also features common mathematical symbols, currency symbols like the British pound and Japanese yen, and a subset of Greek letters used in scientific notation. Notably, it contains many block elements and shading characters of varying density, which were used for creating simple graphics in ASCII art and early roguelike games such as Rogue (video game).

Usage and adoption

This encoding saw ubiquitous use throughout the 1980s and early 1990s as the default for English-language DOS systems from IBM, Compaq, and other IBM PC compatible manufacturers. It was the native character set for the COMMAND.COM shell and was used by thousands of applications, including WordPerfect, Lotus 1-2-3, and the BASIC programming language interpreter. The distinctive appearance of its characters defined the visual style of BBS systems, file archive directories, and the startup screens of many MS-DOS games. While later supplanted by code page 850 and Windows-1252 in international contexts, its legacy persists in terminal emulator settings and retro computing communities.

Technical details

Technically, it is an 8-bit, single-byte encoding where code points 0 through 127 map identically to ASCII. Code points 128 through 255 are defined by the hardware font stored in the ROM of the original IBM PC video adapters. The encoding does not conform to the ISO/IEC 8859 family of standards, as its primary design goal was graphical utility rather than Latin alphabet coverage. In modern systems, it is often referenced by the IANA registration "CP437" and can be selected in environments like the Microsoft Windows command prompt or the Linux console.

Variants and related code pages

Several official variants were created by IBM and Microsoft for different markets or hardware. Code page 850 was developed as a "multilingual" replacement, substituting many graphical characters with additional accented characters to better support Western European languages like French and German. For the IBM PC AT, code page 852 was created for Central European languages. Other graphical OEM code pages include code page 866 for Cyrillic script used in Russian versions of DOS, and Mazovia encoding for the Polish language. In the Unicode standard, all characters from this encoding have direct mappings, ensuring preservation in digital archives.