@Umbreon126: if I seemed to have left (over 4 years without a new comment here, sorry), I never stopped this exhausting project for more than a few weeks, I think. Collecting and sourcing/comparing/debugging incomplete 'micro' fonts is quite tedious (these creations/rippings are actually often quickly made/copied by novice designers that unawarely boast about having unleashed the -world- tiniest Font, repeatedly since the 2000s); this process is also very time-consuming for a single-man project, believe me! hundreds of pages, and probably thousands of examples (video games included - these files will help document my 'typeface', its unique value by the way). I sometimes talk/design pixel art on FR.Wikipedia.org to have a break - which is a long task itself, with bindings to this place, e.g. the isometric fontstructions that I can convert, even shrink in voxel art too, with appealing minimalist stuff to come). I almost completed my 'micro' fonts (as asserted on my FS profile : thousands of them in the meantime, usable in real life - more or less by human memory), but since the current typeface technologies still cannot manage this big amount of fonts (with their specific settings in a single/mere family, where e.g. x-height must be greatly completed, a lot of dedicated variables added, the kerning and several concepts rethought/improved), our 'good' old .ttf format on FontStruct is no longer a solution (all the 'modern' type extensions cannot fit either, I'm afraid). That's why I needed a custom script (lightweight, 8-bit computer compatible, e.g. with a standard Speccy 48K - even 16K?). It shall look (internally) like a rather short piece of code, albeit powerful (with multicriteria selection besides a rustic renderer). This software is based on a limited number of shared glyphs (as I already explained - because I made my choices for a long time now, after looking at a few competing ones, beginning with a 4x5 -created basically to make things comparable- and shrinking the matrix to what you'll think is too much, still logical in general, hopefully). The programmed interface has to be always optional (e.g. I want all the fonts to be accessible from the more conventional way too: a series of conventionally well-sorted individual files - a manual selection that will turn out to lack of efficiency, every time you need to change/adapt the font, named in a non trivial way because of their complicated/precise nature). An ideal solution will be some kind of AR on a mobile (with realtime 3D parametric textures, so that your system can track the matrix your decide in real life, if I can omit the necessary contextual fine tuning in my schematic explanation). I can already afford unveiling this possible and further part of the project, because I know it's unlikely to be imitated in a valuable/successful way, or spoiled too much either: the idea is one thing, the know-how is another one. Without the right 'typeface' (with all its -we say- 'graceful' degradations, thought from scratch and not per font), people are still repeating decades of trials and errors. (In 2017, for instance, I'm already pleased to see more and more examples of 3x5 matrix designs, but the beginners of the genre still cannot get how their accidental 5x5 glyphs slow dow the popularization of this process of minimalism/simplicity/saving/ecology/etc. A lot of applications remind me the same concern: one cannot be always satisfied with deceptive measures (unless you enjoy drinking an empty glass of water in the land of the blind, you see): for example, again, 3x4 is not 5x4, else the indelicate designer might have to pay the difference by commitment. In this situation, I would order my name in huge, gold-plated voxels, and have me offer implicitly more weight than expected by the seller if he only counted the full letters. This amateurish nonsense cannot work everywhere. Badly named/described fonts are defects by design (the author might not understand his creation). Lazy (i.e. larger/wider) glyphs are to be avoided (or reminded as such). Let me try to stop here before I talk verbosely about the pseudo fonts too (duplicates are lame 'hacks' or omissions; not supporting basic Ascii too, else " " would represent a multipurpose set of characters, full of usable glyphs -according to the sleeping designer-, forever and everywhere, kiddies… the typographical minimalism to its extreme representation: a null font). So, come on, people: please take your pseudo fonts back, and make them usable at the stated size (nxn grid) and completeness (e.g. letters set vs real font etc.). Our effort is always appreciated in this thankless field of creation; and it deserves a longer reusability than our playtime on FontStruct, in my opinion.

Hi, Ohsin! “3x3 Enhanced“ looks like an unofficial clone of your “3x3B” font. :-)

Your description looks partly erroneous b.t.w.: “Full 'Basic Latin'” would stand for “basic ASCII”, in my opinion… (thus with lowercase support.) 72 % of the expected font is done (if my arithmetic is correct, because I'd like to see the missing 26 glyphs from yours too)… So: do you think it's worth trying (with your own solutions)? Tip #1: you may want to change the numerics radically. Tip #2: your x-height is of 3 dots (the max as well)…

---

We share a long, long way. But nowadays, since we know that it is possible to draw (and quite readably in my own designs, even in smaller grids), time works in our favour… [I mean, it is unlikely -long-fingered-people of the future go back to -deceptively- easier designs, e.g. 8x8.] Bye!