Most of the CBM 8-bit computers come with two character sets.  One of them has been referred to as the uppercase/graphics character set, and it is usally the default set.  It does not have any lower-case letters!  The letters (that it does have), the numbers, and almost all of the punctuation characters match those of ASCII, the American Standard Code for Information Interchange. 

Like many 8-bit computers, CBM also extends the 7-bit ASCII definition to 8-bits.  There are many extensions to ASCII, generally incompatible with one another.  Anyway, Commodore has their own, and I call it ASCII-X (ASCII extended) to distinguish it from the other character set, PETSCII.  [The currently most popular character code (UTF-8) is a superset of ASCII.]

 

An extension from 7-bits to 8-bits should add 128 codes;  well it does, but half of them are duplicates in Commodore's extended ASCII!  So ASCII-X only adds 64 new "characters"; 32 of them are graphical symbols and 32 are control codes.  ASCII-X has 64 graphical symbols, however, because there are no lower-case characters.  Anyway, you'll see a lot of duplicates if you examine the tables below.  To make matters worse, when the C64 came out, there was a problem with color "smearing" with the thin vertical characters.  So all vertical characters were made two pixels wide (instead of one).  This made a few codes into duplicates... but now the horizontal lines wouldn't mesh well with the new vertical ones... so the horizontal lines were changed too... resulting in even more duplicates!  This was carried over to the TED series (C16 and Plus/4) and the C128 (which has a VIC-II like the C64). 

 

Technical note (okay to skip this paragraph).  The problem with color smearing isn't strictly the video chips, but also the fact these were designed as home computers to be connected to a TV, and broadcast standards limit the color information relative to luminance.  So this is more-or-less expected with the TED and VIC-II.  However the VDC of the C128 is perfectly capable of showing the "proper" font, but the US / International version has the same font as the Plus/4.  Nationalized versions (those with ASCII/DIN key) do have a "thin" character set, but I don't own one to test the fonts, relative to the original in the PET and VIC-20 (they do add special characters for the language/region they were designed for, so they would differ anyway).

 

Note the computer (at least BASIC and the KERNAL) use the same codes regardless of which character set the user chooses for display.  This can make things confusing.  Many official publications and magazine articles from back in the day just use the term ASCII without referring to the character set in use; in this case you should be skeptical.  I would assume they refer to ASCII-X if they did not specifically mention the other (lowercase/uppercase) character set, in which case I would infer the PETSCII set.

 

So here is a table listing the decimal and hexidecimal values of ASCII-X, the character itself (Glyph), and the ASCII and Unicode equivalants, if they exist.  Some of the codes do not normally print a character, but peform an action.  These are shown under Glyph with a description and the reverse-font character that may appear in some circumstances (so-called quote mode and insert mode).  Otherwise, the glyph is only present if it has a Unicode value, although some browsers may not render them correctly (may appear as a blank, or square box, or ?).  There are also notes that describe the action of some codes, a keyboard combination to generate it, and/or a description of some characters.  There are some difference in the layouts of the CBM machines, but they are mostly equivalant.  The descriptions for keyboard shortcuts are based on the VIC-20/C64/C128.  I've also broken the table up into blocks to hopefully highlight the structure of this character set, although some things just seem to defy logic.