ASCII Character Code Reference

[Copyright 1975,1979,1983,2002,2003,2004,2005,2007,2008,2011 Frank Durda IV, All Rights Reserved.
Mirroring of any material on this site in any form is expressly prohibited.
The official web site for this material is:  http://nemesis.lonestar.org
Contact this address for use clearances: clearance at nemesis.lonestar.org
Comments and queries to this address: web_reference at nemesis.lonestar.org]

In this table, the code or symbol name is shown on the first line, followed by the decimal value for that code or symbol, followed by the hexadecimal value. The binary value can be computed based on the row and column where the code or symbol resides, or directly from the hexadecimal value. For example, the character "+" has the binary value "010 1011", with "010" taken from the row and "1011" taken from the column. Similarly, the lowercase letter 'p' has the binary value "111 0000".

The background color for each code or symbol indicates the category that the code resides in. Red indicates control (non-printable) codes. Orange indicates basic punctuation and symbols. Yellow indicates numeric digits. Green indicates the uppercase letters. Blue indicates lowercase letters. (part of the extended character set). Purple indicates punctuation and symbols that are in the extended character set. (If color viewing is not available, the following table gives these categories as numeric ranges.)

ASCII Code Divisions and Categories

The extended printable character set was deliberately arranged so that if a symbol was received in this range and could not be displayed due to limitations of the printing or display device, the symbol in the basic printable range exactly 32 (0x20) positions earlier could be substituted and would provide reasonable results. In such situations, "{" and "}" would be displayed or printed as "[" and "]", while lowercase letters would be displayed or printed in uppercase.

ASCII Control Codes

Codes shown in Yellow are used for physical level synchronous link idle fills, editing, and DCE command functions. Orange is for codes primarily used in Synchronous transmission protocols, such as SDLC. Green codes are used to direct printer (or VDT display) paper and print head (cursor) non-printing movement. Blue codes are peripheral operator alert controls. Red codes divide data in higher level protocols, including some file system and multi-track tape format structures.

It should be mentioned that the operating systems for computers made by the late Digital Equipment Corporation (DEC), particularly the PDP-8, PDP-11, and PDP-10/DECSystem-10/DECSsystem-20 systems, had a profound and lasting influence on the uses of ASCII control characters that were employed by users at client terminals, directing operating systems and applications.

Virtually all the control character uses and conventions that were used in the DEC PDP-11 operating systems (including RT-11 and RSTS-11) were copied into the Digital Research CP/M operating system, which itself was later copied by other vendors to create a CP/M-clone that ran on the Intel 8088/8086 processor, an operating system which Microsoft Corporation bought, renamed PC-DOS, and licensed to IBM. (PC-DOS was later renamed again to MS-DOS.)

The earliest versions of UNIX were developed on DEC systems, and the Bell Labs programmers elected to use many of the same character codes that DEC had already defined for various functions in their operating systems.

Finally, additional control code uses from the DECSystem TOPS, TWENEX and ITS (aka "Incompatible Time Sharing") operating systems can be found today in BSD-derived versions of UNIX terminal drivers and applications. One specific duplication of the TOPS environment can be found in the "set filec" mode of BSD csh shell. (I am talking about the real BSD 4.x csh, not the Convex tcsh one commonly renamed as "csh" today. The "set" command with no parameters will show you if it is the real csh or not.)

Evolution of ASCII

There have been several versions of the ASCII coding system. There were formal versions in 1963, 1965, 1967 and the ANSI version in 1968. The following list details many of the changes made to the coding system during this period.

• Code 0 (now the control code NUL) is also known as NULL on some equipment.
• Code 1 (now the control code SOH) was the control code SOM prior to 1965.
• Code 2 (now the control code STX) was the control code EOA prior to 1965.
• Code 3 (now the control code ETX) was the control code EOT prior to 1965.
• Code 5 (now the control code ENQ) was the control code WRU prior to 1965. Both names had the same meaning. Also known as "HERE IS" or "ANSWERBACK" on some equipment.
• Code 6 (now the control code ACK) was the control code RU prior to 1965.
• Code 16 (now the control code DLE) was the control code DC0 prior to 1965.
• Code 21 (now the control code NAK) was the control code ERR prior to 1965.
• Code 22 (now the control code SYN) was the control code SYNC prior to 1965.
• Code 23 (now the control code ETB) was the control code LEM prior to 1965.
• Code 24 (now the control code CAN) was the control code S0 prior to 1965.
• Code 25 (now the control code EM) was the control code S1 prior to 1965.
• Code 26 (now the control code SUB) was the control code S2 prior to 1965.
• Code 27 (now the control code ESC) was the control code S3 prior to 1965.
• Code 28 (now the control code FS) was the control code S4 prior to 1965.
• Code 29 (now the control code GS) was the control code S5 prior to 1965.
• Code 30 (now the control code RS) was the control code S6 prior to 1965.
• Code 31 (now the control code US) was the control code S7 prior to 1965.
• Code 35 ("#", known as the Number sign if it precedes a numeral and known as the Pound sign if it follows a numeral) is used for the British sterling pound currency symbol ("£") in the ISO version of ASCII (ISO 646). The British sterling pound currency symbol was never part of ASCII. However, the British sterling pound currency symbol does exist in the ISO 8859-I character code set as Code 163. (HTML display systems that are able to display ISO 8859-I will display the British sterling pound currency symbol in response to the HTML code "£". The results on systems that do not support ISO 8859-I are not defined, so this code should be avoided or used with caution.)
  
• Code 39 (now the apostrophe "'") was assigned to the acute accent ("´") before 1965.
  
• Code 64 (now the at-sign "@") was assigned to the "`" character between 1965 and 1967 on some equipment.
  
• Code 92 (now the backslash "\") was assigned to the "~" character between 1965 and 1967 on some equipment.
  
• Code 94 (now the circumflex "^") was assigned to a fully-formed arrow pointing up prior to 1965. There are no arrow symbols (or indicators) in modern ASCII.
  
• Code 95 (now the underscore "_") was assigned to a fully-formed arrow pointing to the left before 1965. There are no arrow symbols (or indicators) in modern ASCII.
  
• The extended printable character set (Codes 96 thru 123) was first defined in 1965.
  
• Code 124 (now the vertical bar "|") was the control code ACK before 1965. Between 1965 and 1967, it was used for the NOT symbol ("¬").
  
• Code 125 (now the close brace "}") was a control character with no specific assignment prior to 1965, although some manufacturers used it as "ALT MODE".
  
• Code 126 (now the tilde "~") was the ESC control code prior to 1965 and between 1965 and 1967, it was used for the vertical bar "|"). In subsequent years, some equipment treated this code and the newer code 27 for ESC as being identical.
  
• Code 127 is also known as "RUB OUT" or "DELETE" on some equipment. On paper punching systems, this code was commonly used to over-punch an incorrect punch on a tape, with systems configured to ignore this code when received. Both code 0 and 127 were used as timing idle characters on various platforms, transmitting one or more following a carriage return code, as Teletypes and other mechanical printers that existed during the early ASCII period took more than one character time to position the print head to column 1, and took additional time to advance the paper.

Symbols not included in ASCII

There are numerous symbols that do not exist in ASCII but might seem logical to have. Some do exist in other character sets, but these are not part of ASCII.

ASCII has only 94 code combinations that can be used to produce printable characters. Since 52 codes are consumed by the alphabet, and another 10 are consumed by numeric digits, this only leaves 32 codes for punctuation and other symbols. And in those 32 codes, 5 had to contain similar looking characters to characters found in the other set of 5. For example, eg '[' and ']' is in one set of 5 with '{' and '}' in the other set of 5. and so on. This design of ASCII was intentionally organized to allow simpler display devices to be produced that only had to print 62 of the 94 ASCII printable codes and could substitute something "close" when asked to display an ASCII character that the device was incapable of producing, such as using the uppercase letter when the lowercase letter could not be printed.

Subsequently, many symbols that might be desired are just not present in ASCII. For example, to provide all accent marks commonly used in European languages on vowels, as many as twelve codes per vowel would be needed, requiring perhaps sixty codes. The ASCII character coding system just doesn't have the space to include these symbols.

Here is a list of codes that people frequently inquire about that do not exist in ASCII.

• Currency symbols other than the US Dollar symbol ($). That would include the US "cent" symbol, the British sterling pound currency symbol, shillings, Yen, the recently-created "Euro" symbol, etc.
• Copyright symbol (letter 'C' in a circle). The US Patent and Trademark office accepts "Copr." or "Copyright" spelled-out as substitutes when the copyright symbol is not available for use. The three symbol sequence '(C)' is not considered an acceptable substitute for the actual copyright symbol or in place of using one of the two text phrases shown.
• Registered trademark symbol (capital letter 'R' in a circle).
• Mathematical division symbol (computer languages usually substitute the forward slash '/', sometimes called a Virgule, which does exist in ASCII).
• Numerous other mathematical symbols, including "not equal", "plus-minus", "pi", "integration", etc. (Computer programming languages use '!=' or '<>' to represent "not equal", while the others have no standardized substitutes.)
• Language vowel accent marks, used in numerous European languages.
• Non-Roman characters as found in Asian, Eastern Europe and Middle-Eastern languages.

Some display systems also offer special "fonts" that include symbols specific to certain occupations or world regions, but these typically re-use the same numerical code values that ASCII uses for its printable and extended printable characters. Because of the overlap, in order to mix special character codes and normal ASCII characters together, the special font must be activated, the special character selected, and then the special font deactivated. This must be repeated each time a character from the set not currently selected is desired, and in some equipment, only one set can be displayed at a time.

The typical World Wide Web browser is able to display both ASCII and ISO-8859 characters simultaneously because their numerical codes do not overlap. However, since most keyboards can only produce ASCII characters, the display of ISO-8859 characters is achieved by using HTML escape codes that are entered using ASCII codes. For example in the HTML language, the ASCII character sequence '&cent;' in an HTML document will display the ISO-8859 character '¢' on most systems.

ASCII printable characters and color, font type and font size issues

It should be understood that no ASCII code specifies the font type, font size or color of the ASCII printable characters. These and any other additional attributes of printable text are optionally specified at a higher coding level, usually by preceding the target characters with an escape sequence, followed by instructions specifying how printable characters from this point forward should be displayed.

For example, in the HTML language, the HTML tag sequence <FONT COLOR="#FF6666"> specifies that subsequent characters should be displayed in the same font type and size previously used, but that when displayed on a device capable of displaying colors, the color of the subsequent characters should be displayed in the specified shade of red. You can see some color tables and the HTML values needed to produce them in the appendix of this document: The Use and Misuse of Color in Web Pages

A note about BREAK and Modem BREAK signaling

When the BREAK key is pressed on a real communications terminal or similar device, the asynchronous serial transmission line begins to send continuous Spacing, the opposite of the "rest" state of continuous Marking. While the BREAK signal condition is present, there are no start, data, stop or parity bits being sent.

If the duration of a continuous spacing condition exceeds 1.6 seconds, it is usually considered to be a Modem BREAK indication. Traditionally, a Modem BREAK indication directed the local and distant modems to drop carrier and end the call. Some teleprinters also turned their motors off in response to this signal, while some "party line" networks used a BREAK signal to attract the attention of the network controller.

In "current-loop" transmission systems, the Spacing condition is equivalent to a lack of current present on the loop, and if the condition persisted it was treated as a "break" in the circuit.

On time-sharing dial-up systems, a BREAK signal is usually interpreted by the receiving communications controller or computer as a directive to stop the current operation, usually by behaving as though it had received some ASCII control character that normally performed the interrupt function, like CTRL-C.

On keyboards with computers attached or integrated (such as all modern PCs), the BREAK key (if present) likely is translated from a keyboard scan code into some ASCII code (such as CTRL-C), rather than producing an actual BREAK line condition. If an actual BREAK signal was to be sent out a serial port, this would have to be done by a special instruction to the tty/serial driver of that operating system. (POSIX environments use an ioctl() call to do this.)

Early Uses of ASCII and alternate coding systems

One of the earliest 7-bit ASCII devices was an improved line of electro-mechanical printers made by the Teletype corporation. With an operational speed of up to 10 characters per second, these devices were used worldwide for message transmission by Western Union, various news wire services and the military. Later, these devices found new uses as input/output devices connected to computer systems that also communicated using the ASCII character set.

The most widely-manufactured Teletype model was number 33, which was sold under a variety of model names such as the KSR-33 and ASR-33. These devices could only print the basic printable character portion of the ASCII character set (64 characters). This limited these devices to uppercase letters, numbers and most punctuation characters as shown in the table above. Some early video terminals and computers (such as the Digital Equipment Corporation VT50 and the Radio Shack TRS-80 Model I) supported only the basic printable set of characters, despite being designed and manufactured years after the ASCII extended character set was adopted. Some manufacturers did offer upgrades that allowed for the display of all ASCII printable characters.

Prior to the introduction of the ASCII-based teletype printers, the Teletype corporation produced teleprinters that used Baudot or "5-Level" character codes, operating at speeds between 40 and 75 baud. These were widely used for over thirty years, but were largely removed from service by the mid 1960s.

IBMs earlier mainframe computers (notably the IBM 360 and 370 families) did not use ASCII. Instead, they used an alternate character coding system called EBCDIC which was devised by IBM as a way to ensure that any peripherals to be connected to IBM computers were also made by IBM. IBM eventually lost this battle and by the late 1970s, it was common to see IBM systems that used EBCDIC internally, but had external communication processors that translated transmissions between IBMs EBCDIC and what other equipment makers were using, which was ASCII.

Related Topics

Baudot (5-Level) Character Code Reference (HTML)

SIXBIT Character Code Reference (HTML)

RADIX50 Character Code Reference (HTML)

Baudot (5-Level) Character Code Reference (HTML)

Return to the Telecommunications Reference Index (HTML)

[Copyright 1975,1979,1983,2002,2003,2004,2005,2007,2008,2011 Frank Durda IV, All Rights Reserved.
Mirroring of any material on this site in any form is expressly prohibited.
The official web site for this material is:  http://nemesis.lonestar.org
Contact this address for use clearances: clearance at nemesis.lonestar.org
Comments and queries to this address: web_reference at nemesis.lonestar.org]

Visit the nemesis.lonestar.org home page and index