Current 5 bit values:
(is there any way to get rid of the space between each item in the list below?:)
00000 space
00001 a
00010 c
00011 d
00100 e
00101 g
00110 h
00111 i
01000 l
01001 m
01010 n
01011 o
01100 p
01101 r
01110 s
01111 t
10000 u
10001 v
10010 ,
10011 .
10100 ~
10101 `
10110 '
10111 ?
11000 !
11001 (
11010 )
11011 :
11100 |
11101 TBD
11110 TBD
11111 TBD
3 Comments
How does this work?
The goal is to have an 'alphabet' of no more than 32 glyphs, so that it can be stored in 5 bit numbers while still being generally readable.
So the font translates each letter either into its own glyph or a combination of glyphs. For example c is 'c' and x is 'cs' in the font - appearing as 2 five bit glyphs.
I have been able to crunch the core letters down to 17, therefore 9 letters are combinations of others: b=pv, f=ph, j=dg, k=cc, q=cccc, w=uu, x=cs, y=ei, z=ds.
I do the same thing with symbols.
The core symbols are .,~` and ' - these are 'dots'. Other core symbols are ?!(): and |. Core symbols can be used to build other symbols, for example the { is ~(
Capitalization is done by prefixing a letter with a `. Numbers and some symbols are done by spelling them out after a tilde so 1=~one
The compression is lossy but is still pretty good. for example kk=cccc which then resolves back to q. It's not perfect but it's pretty good.
Thank you for the detailed explanation of your experiment, jongrover. I tried myself a 5-bit design, showcased in 'Ascii Art' (Unicode), titled sample G. “5-bit confusion”, where the lowercase and the numerics are mixed in a l337 way, but in a difficult 2x3 pixel matrix… (this pseudo font is part of my -unreleased- main project of 'micro typeface.) I'll need more time to train myself for using your (apparently quite) smart design. Thanks for sharing/saving ;-)
Please sign in to comment.