Jump to content

PETSCII

From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by Washi (talk | contribs) at 15:11, 25 March 2024 (→Base 128). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

PETSCII
PETSCII (shifted and unshifted)
Language(s)English with pseudographics
Classification8-bit extended early ASCII
Based onUS-ASCII (1963 version)

PETSCII (PET Standard Code of Information Interchange), also known as CBM ASCII, is the character set used in Commodore Business Machines' 8-bit home computers.

This character set was first used by the PET from 1977, and was subsequently used by the CBM-II, VIC-20, Commodore 64, Commodore 16, Commodore 116, Plus/4, and Commodore 128. However, the Amiga personal computer family instead uses standard ISO/IEC 8859-1.

History

The character set was largely designed by Leonard Tramiel (the son of Commodore CEO Jack Tramiel) and PET designer Chuck Peddle.[1][2][3] The graphic characters of PETSCII were one of the extensions Commodore specified for Commodore BASIC when laying out desired changes to Microsoft's existing 6502 BASIC to Microsoft's Ric Weiland in 1977.[4] The VIC-20 used the same pixel-for-pixel font as the PET, although the characters appeared wider due to the VIC's 22-column screen. The Commodore 64, however, used a slightly re-designed, heavy upper-case font, essentially a thicker version of the PET's, in order to avoid color artifacts created by the machine's higher resolution screen. The C64's lowercase characters are identical to the lowercase characters in the Atari 800's system font (released several years earlier).

Peddle claims the inclusion of card suit symbols was spurred by the demand that it should be easy to write card games on the PET (as part of the specification list he received).[2]

Specifications

PETSCII Chart as displayed on the Commodore 64 in shifted and unshifted modes. (Not shown are control codes, as well as characters in the 0xC0–0xFF range, which are the standard uppercase keycodes returned from the keyboard, and which are duplicated to the range 0x60–0x7F.)

"Unshifted" PETSCII is based on the 1963 version of ASCII (rather than the 1967 version, which most if not all other computer character sets based on ASCII use). It has only uppercase letters, an up-arrow ⟨↑⟩ instead of caret ⟨^⟩ at 0x5E and a left-arrow ⟨←⟩ instead of an underscore ⟨_⟩ at 0x5F. In all versions except the original Commodore PET, it also has a British pound sign ⟨£⟩ instead of the backslash ⟨\⟩ at 0x5C. Other characters added in ASCII-1967 (lowercase letters, the grave accent, curly braces, vertical bar, and tilde) do not exist in PETSCII. Codes 0xA0–0xDF are allotted to CBM-specific block graphics characters—horizontal and vertical lines, hatches, shades, triangles, circles and card suits.

PETSCII also has a "shifted" mode (also called "business mode"), which changes the uppercase letters at 0x41–0x5A to lowercase, and changes the graphics at 0xC1–0xDA to uppercase letters. Upper- and lower-case are swapped from where ASCII has them. The mode is toggled by holding one of the SHIFT keys and then pressing and releasing the Commodore key. The shift can be done by POKEing location 59468 with the value 14 to select the alternative set or 12 to revert to standard. On the Commodore 64, the sets are alternated by flipping bit 2 of the byte 53272. On some models of PET, this can also be achieved via special control code PRINT CHR$(14) which adjust the line spacing as well as changing the character set; the POKE method is still available and does not alter the line spacing.[5]

Included in PETSCII are cursor and screen control codes, such as {HOME}, {CLR}, {RVS ON}, and {RVS OFF} (the latter two activating/deactivating reverse-video character display). The control codes appeared in program listings as reverse-video graphic characters, although some computer magazines, in their efforts to provide more clearly readable listings, pretty-printed the codes using their actual names in curly braces, like the above examples. This is unambiguous as PETSCII has no curly brace characters.

Different mappings are used for storing characters (the "interchange" mapping, as used by CHR$()) and displaying characters (the "video" mapping). For example, to display the characters "@ABC" on screen by directly writing into the screen memory, one would POKE the decimal values 0, 1, 2, and 3 rather than 64, 65, 66, and 67.[6][7]

The keyboard by default provides access to the lower half of the code page. Pressing Shift and a key gives the corresponding upper half code point. Some PETSCII code points cannot be printed and are only used for keyboard input (e.g. F1, RUN/STOP).

PET 2001 keyboard layout, illustrating PETSCII graphics characters

Character set

<?>
This article contains uncommon Unicode characters. Without proper rendering support, you may see question marks, boxes, or other symbols instead of the intended characters.

The tables below represent the "interchange" PETSCII encoding, as used by CHR$().

Control characters are defined in the ranges 0x00–0x1F and 0x80–0x9F, although which control characters are defined and what they are defined as varies between systems. The tables below exclude control characters—the encoding of control characters in discussed in § Control characters.

The ranges 0x60–0x7F and 0xE0–0xFF are duplicate ranges, although what they duplicate varies between systems. On the Commodore PET, they duplicate 0x20–0x3F and 0xA0–0xBF, respectively; on the Commodore VIC-20, 64, 16, and 128 they duplicate 0xC0–0xDF and 0xA0–0xBF, respectively.[6] While these characters are visually duplicates, they are semantically different; for example, on the Commodore PET, code points 0x2C and 0x6C both produce a comma character, but only 0x2C functions as a delimiter between input fields.[8]

Graphic characters are mostly identical across systems, with the exceptions of 0x5C (which is \ on the Commodore PET, and £ on other systems), 0xD0 (which is U+1FB95 CHECKER BOARD FILL on the Commodore PET and VIC-20, and U+1FB96 INVERSE CHECKER BOARD FILL on other systems), and the range 0x60–0x7F (which duplicates a different range on Commodore PET). Additionally, in Commodore PET 2001's shifted character set, uppercase and lowercase letters are swapped relative to other systems'.

Compatibility symbols for PETSCII characters were added to Unicode 13.0 as part of the Symbols for Legacy Computing block.[9]

Standard

The following tables represent the PETSCII encoding used on the Commodore VIC-20, 64, 16, and 128.

Unshifted

Unshifted PETSCII[6][10][7][11][12]
0 1 2 3 4 5 6 7 8 9 A B C D E F
0x
1x
2x  SP  ! " # $ % & ' ( ) * + , - . /
3x 0 1 2 3 4 5 6 7 8 9 : ; < = > ?
4x @ A B C D E F G H I J K L M N O
5x P Q R S T U V W X Y Z [ £ ] ↑ ←
6x ─ ♠ 🭲 🭸 🭷 🭶 🭺 🭱 🭴 ╮ ╰ ╯ 🭼 ╲ ╱ 🭽
7x 🭾 • 🭻 ♥ 🭰 ╭ ╳ ○ ♣ 🭵 ♦ ┼ 🮌 │ π ◥
8x
9x
Ax NBSP ▌ ▄ ▔ ▁ ▏ ▒ ▕ 🮏 ◤ 🮇 ├ ▗ └ ┐ ▂
Bx ┌ ┴ ┬ ┤ ▎ ▍ 🮈 🮂 🮃 ▃ 🭿 ▖ ▝ ┘ ▘ ▚
Cx ─ ♠ 🭲 🭸 🭷 🭶 🭺 🭱 🭴 ╮ ╰ ╯ 🭼 ╲ ╱ 🭽
Dx 🭾 • 🭻 ♥ 🭰 ╭ ╳ ○ ♣ 🭵 ♦ ┼ 🮌 │ π ◥
Ex NBSP ▌ ▄ ▔ ▁ ▏ ▒ ▕ 🮏 ◤ 🮇 ├ ▗ └ ┐ ▂
Fx ┌ ┴ ┬ ┤ ▎ ▍ 🮈 🮂 🮃 ▃ 🭿 ▖ ▝ ┘ ▘ ▚
  Differs between PETSCII variants.

Shifted

Shifted PETSCII[6][10][7][13][14]
0 1 2 3 4 5 6 7 8 9 A B C D E F
0x
1x
2x  SP  ! " # $ % & ' ( ) * + , - . /
3x 0 1 2 3 4 5 6 7 8 9 : ; < = > ?
4x @ a b c d e f g h i j k l m n o
5x p q r s t u v w x y z [ £ ] ↑ ←
6x ─ A B C D E F G H I J K L M N O
7x P Q R S T U V W X Y Z ┼ 🮌 │ 🮕/🮖[a] 🮘
8x
9x
Ax NBSP ▌ ▄ ▔ ▁ ▏ ▒ ▕ 🮏 🮙 🮇 ├ ▗ └ ┐ ▂
Bx ┌ ┴ ┬ ┤ ▎ ▍ 🮈 🮂 🮃 ▃ ✓ ▖ ▝ ┘ ▘ ▚
Cx ─ A B C D E F G H I J K L M N O
Dx P Q R S T U V W X Y Z ┼ 🮌 │ 🮕/🮖[a] 🮘
Ex NBSP ▌ ▄ ▔ ▁ ▏ ▒ ▕ 🮏 🮙 🮇 ├ ▗ └ ┐ ▂
Fx ┌ ┴ ┬ ┤ ▎ ▍ 🮈 🮂 🮃 ▃ ✓ ▖ ▝ ┘ ▘ ▚
  Differs between PETSCII variants.
  1. ^ a b This is U+1FB95 CHECKER BOARD FILL on the VIC-20; and U+1FB96 INVERSE CHECKER BOARD FILL on the Commodore 64 and Commodore 128.

Commodore PET

Unshifted

Unshifted PETSCII (PET)[6][15]
0 1 2 3 4 5 6 7 8 9 A B C D E F
0x
1x
2x  SP  ! " # $ % & ' ( ) * + , - . /
3x 0 1 2 3 4 5 6 7 8 9 : ; < = > ?
4x @ A B C D E F G H I J K L M N O
5x P Q R S T U V W X Y Z [ \ ] ↑ ←
6x  SP  ! " # $ % & ' ( ) * + , - . /
7x 0 1 2 3 4 5 6 7 8 9 : ; < = > ?
8x
9x
Ax NBSP ▌ ▄ ▔ ▁ ▏ ▒ ▕ 🮏 ◤ 🮇 ├ ▗ └ ┐ ▂
Bx ┌ ┴ ┬ ┤ ▎ ▍ 🮈 🮂 🮃 ▃ 🭿 ▖ ▝ ┘ ▘ ▚
Cx ─ ♠ 🭲 🭸 🭷 🭶 🭺 🭱 🭴 ╮ ╰ ╯ 🭼 ╲ ╱ 🭽
Dx 🭾 • 🭻 ♥ 🭰 ╭ ╳ ○ ♣ 🭵 ♦ ┼ 🮌 │ π ◥
Ex NBSP ▌ ▄ ▔ ▁ ▏ ▒ ▕ 🮏 ◤ 🮇 ├ ▗ └ ┐ ▂
Fx ┌ ┴ ┬ ┤ ▎ ▍ 🮈 🮂 🮃 ▃ 🭿 ▖ ▝ ┘ ▘ ▚
  Differs from standard PETSCII.

Shifted

Shifted PETSCII (PET)[6][16]
0 1 2 3 4 5 6 7 8 9 A B C D E F
0x
1x
2x  SP  ! " # $ % & ' ( ) * + , - . /
3x 0 1 2 3 4 5 6 7 8 9 : ; < = > ?
4x @ a b c d e f g h i j k l m n o
5x p q r s t u v w x y z [ \ ] ↑ ←
6x  SP  ! " # $ % & ' ( ) * + , - . /
7x 0 1 2 3 4 5 6 7 8 9 : ; < = > ?
8x
9x
Ax NBSP ▌ ▄ ▔ ▁ ▏ ▒ ▕ 🮏 🮙 🮇 ├ ▗ └ ┐ ▂
Bx ┌ ┴ ┬ ┤ ▎ ▍ 🮈 🮂 🮃 ▃ ✓ ▖ ▝ ┘ ▘ ▚
Cx ─ A B C D E F G H I J K L M N O
Dx P Q R S T U V W X Y Z ┼ 🮌 │ 🮕 🮘
Ex NBSP ▌ ▄ ▔ ▁ ▏ ▒ ▕ 🮏 🮙 🮇 ├ ▗ └ ┐ ▂
Fx ┌ ┴ ┬ ┤ ▎ ▍ 🮈 🮂 🮃 ▃ ✓ ▖ ▝ ┘ ▘ ▚
  Displayed case matches the Commodore PET 8032. The opposite case is used on the Commodore PET 2001.
  Differs from standard PETSCII.

Control characters

While the graphic characters were mostly shared between Commodore systems, the control characters frequently varied. The follow table describes what the control characters represent on the Commodore PET 2001, Commodore PET 8032, VIC-20, Commodore 64, Commodore 16, Commodore 128 (40 and 80 column modes).

PETSCII control characters[6][17]
Hex Decimal PET 2001 PET 8032 VIC-20 C64 C16 C128 (40 col) C128 (80 col)
00 0 —
01 1 —
02 2 — UNDERLINE ON
03 3 STOP
04 4 —
05 5 — WHITE
06 6 —
07 7 — BELL — BELL
08 8 — LOCK CASE —
09 9 — TAB UNLOCK CASE TAB
0A 10 — LINE FEED
0B 11 — UNLOCK CASE
0C 12 — LOCK CASE
0D 13 RETURN
0E 14 — LOWER CASE
0F 15 — SET WINDOW TOP — FLASH ON
10 16 —
11 17 CURSOR DOWN
12 18 REVERSE ON
13 19 HOME
14 20 DEL
15 21 — KILL LINE —
16 22 — ERASE TO RIGHT —
17 23 —
18 24 — TAB SET/CLEAR
19 25 — SCROLL UP —
1A 26 —
1B 27 — ESC — ESC
1C 28 — RED
1D 29 CURSOR RIGHT
1E 30 — GREEN
1F 31 — BLUE
80 128 —
81 129 — ORANGE DARK PURPLE
82 130 — FLASH ON — UNDERLINE OFF
83 131 — RUN
84 132 — FLASH OFF —
85 133 — F1
86 134 — F3
87 135 — DOUBLE BELL F5
88 136 — F7
89 137 — TAB SET/CLEAR F2
8A 138 — F4
8B 139 — F6
8C 140 — F8 HELP F8
8D 141 SHIFT + RETURN
8E 142 — UPPER CASE
8F 143 — SET WINDOW END — FLASH OFF
90 144 — BLACK
91 145 CURSOR UP
92 146 REVERSE OFF
93 147 CLEAR
94 148 INST
95 149 — INSERT LINE ABOVE — BROWN DARK YELLOW
96 150 — ERASE TO LEFT — PINK YELLOW-GREEN PINK
97 151 — DARK GRAY PINK DARK GRAY DARK CYAN
98 152 — MEDIUM GRAY BLUE-GREEN MEDIUM GRAY
99 153 — SCROLL DOWN — LIGHT GREEN LIGHT BLUE LIGHT GREEN
9A 154 — LIGHT BLUE DARK BLUE LIGHT BLUE
9B 155 — LIGHT GRAY LIGHT GREEN LIGHT GRAY
9C 156 — PURPLE
9D 157 CURSOR LEFT
9E 158 — YELLOW
9F 159 — CYAN

The colors of the VIC-20 and C64/128 are listed in the VIC-II article.

Base 128

Out of PETSCII's first 192 codes, there are 128 graphic characters: 32–127 and 160–192. This permits "base128"-style encodings in DATA statements, or perhaps between PETSCII-speaking machines. This can also include control characters, which are visible when quoted, although which control characters are defined varies between systems.

The primary application for "Base 128" is in DATA statements in Commodore BASIC. Binary data can be stored with relatively low overhead, allowing one character of data to encode seven bits of data. On a standard 80-character line, typically four characters are used for the line number, and two characters for the tokenized DATA statement, leaving 74 characters for data. At seven bits per character, one DATA line could store 518 bits of binary data, for 80% efficiency. If three-digit line numbers are used, efficiency increases to 82%. If two-digit line numbers are used, efficiency is 83%. 

  Line Numbers    Data chars per Line    Bits per Line    Efficiency   Max. Lines    Max. Total Bytes
  1-9*            77                     539              0.8421875    9             606
  10-99           76                     532              0.83125      90            5,985
  100-999         75                     525              0.8203125    900           59,062
  1000-9999       74                     518              0.809375     9,000         582,750
  10000-65535**   73                     511              0.7984375    55,536**      3.5mb (appx)

For storing binary data in Commodore BASIC, it appears that three-digit line numbers are typically the best choice.

(*1) Assume line 0 is a GOTO.

(*2) Maximum line number is probably off-by-one.

See also

References

  1. ^ Reunanen, Markku; Heikkinen, Tero; Carlsson, Anders (22 November 2018). "PETSCII – A Character Set and a Creative Platform" (PDF). Replay. The Polish Journal of Game Studies. 5 (1): 27–47. doi:10.18778/2391-8551.05.02.
  2. ^ a b Bagnall, Brian (2007). On the Edge: The Spectacular Rise and Fall of Commodore. Winnipeg: Variant Press. pp. 43, 54–55. ISBN 0-9738649-0-7.
  3. ^ Tramiel, Leonard (27 December 2021). "Creating PETSCII". Vintage Computer Stories. Blogspot.
  4. ^ "A Conversation with Chuck Peddle, Bil Herd, Jeri Ellsworth - part 3 - BIOS - blip.tv". blip.tv. 5 September 2010 [2009]. 6:30. Archived from the original on 9 January 2011. (mirror)
  5. ^ Andersson, Larry (25 November 2000). "THE COMMODORE PET COMPUTER FREQUENTLY ASKED QUESTIONS FILE". Zimmers.net. 1.7.
  6. ^ a b c d e f g Oy, Aivosto (2014), "Commodore PETSCII character sets" (PDF), Aivosto
  7. ^ a b c Ewell, Doug; Bettencourt, Rebecca; Bánffy, Ricardo; Everson, Michael; Marín Silva, Eduardo; Mårtenson, Elias; Shoulson, Mark; Steele, Shawn; Turner, Rebecca (4 January 2019), "ReadMe.txt", L2/19-025: Proposal to add characters from legacy computers and teletext to the UCS (PDF), The Unicode Consortium
  8. ^ Brain, Jim (16 March 1996). "Commodore Trivia Edition #26 Answers for February 1996". Zimmers.net. Q $195) On CBM machines prior to the VIC-20, what chr$ code outputs the same character as chr$(44), the comma.
    A $195) 108.
    Q $196) Is the character described in $195 of any use?
    A $196) To put commas in strings read via INPUT. Remember, INPUT treats a comma (chr$(44)) as a delimiter between input fields, but chr$(108) does not produce the same effect, so you could replace 44 with 108 in data written to disk, and read it in with INPUT.
  9. ^ Ewell, Doug; Bettencourt, Rebecca; Bánffy, Ricardo; Everson, Michael; Marín Silva, Eduardo; Mårtenson, Elias; Shoulson, Mark; Steele, Shawn; Turner, Rebecca (4 January 2019), L2/19-025: Proposal to add characters from legacy computers and teletext to the UCS (PDF), The Unicode Consortium
  10. ^ a b Bettencourt, Rebecca G. "PETSCII to Unicode Mapping". KreativeKorp.
  11. ^ Bettencourt, Rebecca (20 April 2018), "CVICIPRI.TXT", L2/19-025: Proposal to add characters from legacy computers and teletext to the UCS (PDF)
  12. ^ Bettencourt, Rebecca (20 April 2018), "C64IPRI.TXT", L2/19-025: Proposal to add characters from legacy computers and teletext to the UCS (PDF)
  13. ^ Bettencourt, Rebecca (11 October 2018), "CVICIALT.TXT", L2/19-025: Proposal to add characters from legacy computers and teletext to the UCS (PDF)
  14. ^ Bettencourt, Rebecca (11 October 2018), "C64IALT.TXT", L2/19-025: Proposal to add characters from legacy computers and teletext to the UCS (PDF)
  15. ^ Bettencourt, Rebecca (20 April 2018), "CPETIPRI.TXT", L2/19-025: Proposal to add characters from legacy computers and teletext to the UCS (PDF)
  16. ^ Bettencourt, Rebecca (11 October 2018), "CPETIALT.TXT", L2/19-025: Proposal to add characters from legacy computers and teletext to the UCS (PDF)
  17. ^ Commodore 128 Programmer's Reference Guide (PDF). Commodore Business Machines, Inc. February 1986. pp. 666–668. ISBN 0-553-34292-4.
Retrieved from "https://en.wikipedia.org/enwiki/w/index.php?title=PETSCII&oldid=1215505468"