Unofficial clone of your own "Gamaeo reglr".
You might want to lower the bar in "H" to mitigate the ambiguity with your new "M" ("W" is OK as is). "0" duplicates "O" (just unsafe for sensitive contexts). One glyph is 5-px wide (not 7-bit Ascii, though - it looks like "Đ").
4x7 is a pretty decent size, for a training experiment. You can try adding the lowercase (all basic Ascii, actually), before pushing your limits to smaller grid sizes. (You'll have to add a descender, to begin with the small letters, i.e. via a hacked baseline, or 4x6-based uppercase/numerics.)
Don't bother about the 'borrowing' of glyphs at this scale, in my opinion: except for your very 'high' style of bars (hence the unexpectedly top-cropped "4"), you did not show a lot of fanciful distortions, for instance ("ACK34" look a tad odd, though), which makes your font rather all-purpose (even common), and more readable eventually.
More comment if you need (even after PM if I seem to quit, as I cannot watch all the FS pages pertaining to interesting low-res pixel fonts.)

can you put your copyright sig in the ©  symbol...

I'll try to reply thoroughly (pardon for not being concise and for keeping some unavoidable jargon)…

• ASCII. You can take this 1960s American standard as our legacy Latin letters A-Z: it must be complete in ANY text font.
All the ASCII characters are to be available in an actual text font, at least the said 'basic' ones (#20-#7E code points in UNICODE nowadays [http://unicode.org/charts/PDF/U0000.pdf]): “ !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~”). If you count well, there are exactly 93 characters in all (numbered 33-126 in decimal, where 32 is the Space character and is generally avoidable by design from your software -spacing- settings). This set of 7-bit characters (= 94 visible ones out of 128 [2^7] of different kinds) is really all today's digital world needs to communicate safely. Even if you think 'real world', you'll need to store your data (i.e. text), otherwise it's lost once the medium get destructed (e.g. a consumed page). In order to save your message via softwares, you must rely on conventional scripts, at least to make the further exchange possible (such conventions bind e.g. one character/glyph and one numeric value/code point together). The ASCII is the most used code today, and has evolved -to me- like an onion (cf. our brain and its layers): its core is 7-bit, like a vital germ (another personal metaphor). Even if you don't need its minimum/old code points (as characters or binary cells), e.g. if you just need to store a single Cyrillic or Chinese letter (scripts that are above the ASCII part of the Latin script), you'll need ASCII as the basis, e.g. for transmission protocols (they use simple commands to start with, thus they still fit very well the scarce set of basic ASCII characters, which is even true in programming, in general). [Less trivially: the related characters are more likely to be used as binary values between softwares, instead of plain text transmitted between humans, for instance to encode images in Base 64 : 64 characters only are needed, vs the 94 available in basic ASCII.] You can just remember that this range of code points ("!" till "~") cannot be reduced safely, if you want your font or typeface to be widely usable (and properly functionaly with our modern tools). That's why I repeat here and again the annoying fact (sorry for jamming the members's shoulder), that an incomplete set of basic ASCII characters is merely a 'pseudo-font', not an actual text font, if you want it to be viable beyond artistic purposes. Because, you see, in art, we got many unproductive deadlocks, like the "U" replaced with "V" (on a lot of Parisian monuments etc. - colliding homonyms in our 26-letter world etc.), if the unaware FontStruct artist still think e.g. "0" can use the same glyphs as "O" ! Science helps in this situation, with its reproducible standards. Basic ASCII may look useless nowadays (to be supported thoroughly), but our computers (all our worldwide technology) requires it completely (for the simplicity as I mentioned, without presuming what will be the choices in the future - I already stated that the uppercase was almost useless to me, as well as our -French- diacritics).

---

• Your update:
· "0" ≠ "O" anymore (interesting choice of orientation in the new zero - of course I prefer a [new] zero, than no zero at all);
· "Ð" as 4x7 instead of 5x7 (I could not do better with your design of high bars [diacritics in small matrices are critical/bad], and this glyphs distinguishes enough from "B");
· "H" bar tweaked to fit e.g. "G" (no more confusion with "M" for me);
· "4" ditto (its bar is still high enough, and the whole looks complete now);
· "Y" ditto (though less useful than in "4" above);
· Suggestion: if it were me, to follow your slight italic, I'd remove the top-left dot from "A" (so that this character looks less square in your rather round design);
· I note the disambiguating serifs in "C" (vs e.g. "G"), which is a hack (in your sans serif rule);
· Interesting "©" (funny but readable);
· A less stylized "/" would use 2-3-2 vertical lines of dots, in my opinion (currently you have 3-1-3 in the same 3x7 size);
· "@" and "&": the most original ones I could see lately! (not quite readable for my brain, though).

• More tips/reminders:
· I need to release a couple of documentation files that explain or insist on the fact that our (pixel) Latin characters are read downwards… In the meantime: if you apply stricly this prerequisite concern of typography, you cannot draw a 'high bar design' that introduces some missing elements on the top (e.g. your former "4"), without spoiling the readability. Another side effect (of this style used in the far past in typefaces aimed at the rich, where reading slowly was not an issue): by packing too much information in just three lines of dots, you go past the limit of the available cells that can encode the differences safely (bars and individual dots). In practice, you often force the attention to the very top of the glyphs, where it is more packed, and unfortunately sensitive than in the rest of the glyphs. When a single dot counts, we need more concentration. So you have 3 lines of critical data, next 4 lines of less important one. This produces like a hiccup in the related glyphs. For instance, between "B" and "R": the reader is like compelled to parse until the last line of dots (to get the difference), while the visually disambiguation process could have been quite smoother if the bar was in the middle. If you stretch both example glyphs to e.g. 4x70, you'll get the point: you ask for too much concentration, before forcing a snooze, and a quick, final check at the bottom. This is unbalanced, but the style requires it. (Of course, all my -unreleased- small fonts occupy the expected, average lines - nothing quite fanciful at this 'micro' size, if you want a classic look and an optimal readability.)
· You asked: “I wonder how many designs are possible before the new ones look like those made before”. Short reply: it's based on your rules (which limits the selectable glyphs in binary). When you apply a mostly 'LCD segment' design, as you did here, you decrease A LOT the possibilities by yourself. E.g. you dismiss most of the fanciful binary glyphs that could be selected in your mapping (jargon, sorry: "0000 0000 0000 0000 0000 0000 1000" would map the dot, for instance). From a huge number of glyphs (2^(4×7)=268'435'456 including the unsafe horizontal duplicates), probably millions, you still have hundreds of thousands of variants/alternates, if you were not discarding the ones that don't follow your invisible layout (yes, even the tiniest raster fonts have layouts at the design steps, not made with the compass as in their vector counterpart, naturally). Since you follow strict positional rules, even a round style (as a rule) etc., you have incredibly less choices. This makes your glyphs more homogeneous (there is less 'noise' in your font), but this common approach is shared by a great deal of quite readable pixel fonts. Which introduces the risk of repeating somebody's else work/experiment. If you need your pixel font to be (more) unique (besides you interesting high bars, that is), you can try e.g. adding/deleting dots in a repeated/logical manner. This would quite lower the readability, of course.
· 4x7 (standard format, used after the display resolution in raster graphics): Width × Height (please).

Hope it helps you dig deeper into the world of the 'micro' characters (in complete basic Ascii, hopefully).

- CONTINUED -
Complete answer, too, fine! :-)
[I hope my poor English still makes sense.] Now I understand more your standpoint (it's just the opposite of 'my' project of small -programming?- fonts: I focus only on useful fonts [because as a pixel artist, I really know it's by far too easy to provide incomplete fonts {this state of mind, a tad juvenile, is behind me; I need someting more serious and constructive}, and these trials/experiments won't last if they share the same flaws as in the past: missing, duplicated or too ambiguous characters, larger matrices than declared, bad conformance between the fonts etc.]).
Fortunately, your "Gameao Halb" is correct from scratch, if you need to support more basic Ascii characters, even a smaller grid (I may archive -yours- from 4x5, not larger in 'my' project of difficult fonts). I mean: your font is strictly 4-px wide (which is still rare - I've just searched again on a competing place/repository: I think all the similar fonts there have wider "W" and "M", for instance, so they cheat in the effort of narrowing the glyphs [our -LtR- Latin script favors this direction of shrinking]). Besides, you new "A" (and "/") is ready for more crushing (at first by deleting the least important lines of dots, in this saga…). Back to you wondering, I also made a documentation file -unreleased too (it contains some final/unveiling glyphs)-: this image shows many editions from a simple 5x5 "A": plenty of shapes are readable (probably more than you'd expect). As 4x6, even 4x5, you may have fun too (the lowercase will be tougher to follow the same styles).
I cannot talk about your “nudging and kerning” direction from this tool; I need practicing beforehand (a lot!)… Say, in a few years, if my priorities are the same as today. (Pixel kerning is quite OK with me [I made my own searches for my selection of tiny glyphs], but nudging not quite, to me.)
Just for me/us to know (no poll feature on FontStruct…): how many characters are enough to be called a 'font', in your former viewpoint? People often think "0-9" and/or "A-Z" do the job. Really, most tiny pixel 'fonts' I could see (and collect) are merely batches of scattered subsets of basic Ascii (it's so disappointing, years after years). Without 93 characters, they are not fonts (one cannot convert the messages between them, except as a transitory step where many glyphs are invisible/missing). That's why, out of thousands of trials we can get on the web (I spotted a lot of one-word 'fonts, in game esp.), the actual good candidates are rare, the valid ones even scarcer. If you can provide e.g. a complete 4x5, it'd be better than 'nothing' (I'd prefer 15-dot designs or less, i.e. 3x5 or below, in complete basic Ascii of course). Why? The more compression, the fewer dots, and the easier the human remembering (I always tell that the tiniest fonts should/will be used in real life, for the greater saving on the materials and the absence of need of an artificial crutch, i.e. a software). My avatar from 2013 gives you an idea of this studied simplicity.
You'll always be welcomed to show/share a derived work of "Gameao Halb", more fanciful or shrinked. I may even post a completed version (in bitmap) if it's quickly made in the next days (for the fun of it, since it's exactly 3 times more dots than my smallest valid font ! - in terms of possibilities, it's a quite a child's play, hopefully, so I cannot really get why it's still unfinished, unless you want me to SHOUT ;-)). If you block, I may have many solutions (from my experiments, even from skilled pixel designers). The more decent small pixel(-distorted) glyphs, the more control we take back on this kind of minimalist writing. Without unity in this process/project, we repeat the same difficulties: the tiny glyphs must be understood at a large scale…
- PAUSED -

Here is a SMALL CAPS version* as a test/candidate. Review:

• "*" is an interesting choice (though a dupe of "°", later).
• "T" bar use a cheat (impossible grid): thus 2 possibilities.
• "z" can be fine tuned (it's up to your style in its center).
• "{}" pair can be made rounder (if 4-dot of width looks fine).
• "''" duplicate """ (a legacy bug in var. width designs, OK).
• "~" might be made bolder (its ends are often done darker…).
• "|" is fakely broken vs "I": if you keep them, just edit "#".
• "#" is a tad ambiguous with "H", plus the point above, now.
• "@" looks like a "ℤ" symbol with a diacritic, but it's fun!
• "&" could be more faithful, or stylized (e.g. w/o its loop).

* Providing the lowercase would have spoiled your work/choices.

جميل

Errata:

• (4×7)÷(2×4)=28÷8>3X larger than the smallest basic Ascii-compatible matrix.

• I chose the alternate possibility in "Tt": mainly intentionally (to make you react on your ambiguous "1", but the unconventional location -on the left- of the vertical bar in "Tt" fits your slightly oblique style [as seen in your "3QY", even "A" now]).

---

I may provide a quick fix for all the mentioned ambiguities/oddities, a.s.a.p. (as a suggestion in a sample image, again). Actually, this demonstrates how the small pixel glyphs are made: “edit it; check it” (against the ambiguities), in a looped -and tedious- process. (That's why, as I already confessed, I could waste two hours of continued work in not being able to change a single pixel from a puzzling 2x4 glyph…) The smallest, the toughest: not from the drawing, but from the choices (that should be understood, if need be by educating the readers via always more minimalist designs). For the moment, your 4x7 looks originally like a 5x7 dot-matrix design, which is very common (and not very copyrightable for this reason b.t.w.). Let's see what I can submit…

Rest of latin

Plz add €¥®ÀÁÂÃÄÅÆÈÉÊËÌÍÎÏÒÓÔÕÖŒÙÚÛÜÝŸàáâãäåæèéêëìíîïòóôõöœùúûüýÿĄąĆćĘęĞğİıIJijŁłŃńŚśŞşŠšŹźŻżŽžƒºª†‡¤√µ¹²³¼½¾‰•…¬♀♂ and many more...

Meant ¢ instead of €

Well done, Aeolien! (you are still quick-witted.) Almost a basic Ascii design (your double quote(s) wouldn't work in a test). Here is an older version in 4x8 too (it's 100 % valid - without any ambiguity, hopefully, to start with, but the symbols etc. can quite be changed.) If you follow the steps, 4x6 is feasible yourself, etc., but not quite with Brynda's suggestion.

---

I think I can stop commenting now (help on request / notify me on PM when 'outside'). Sorry for the didactic tone sometimes, and for my known verbosity. On ne se refait pas! ;-)

tellement impressionnéé, c'est fou ........ *bouche béé*

@nightpegasus: Merci, pour Aeolien aussi. C'est encourageant pour mes milliers de fontes 4x5 maxi (vs 4x7 ici).

"Gameao Halb" + 7-bit ASCII grid examples    
 <Aeolien's>                                 

WxH = G. Dots (Note}                         
4x8 = 32 •••••••••••••••••••••••••••••••• (1)
4x7 = 28 •••••••••••••••••••••••••••• (2)    
5x5 = 25 ••••••••••••••••••••••••• {5}       
4x6 = 24 •••••••••••••••••••••••• (3)        
3x8 = 24 ••••••••••••••••••••••••            
3x7 = 21 •••••••••••••••••••••               
4x5 = 20 •••••••••••••••••••• (4)            
3x6 = 18 ••••••••••••••••••                  
3x5 = 15 ••••••••••••••• {6}                 
3x4 = 12 •••••••••••• {7}                    
2x5 = 10 •••••••••• {8}                      
3x3 = 09 ••••••••• {9}                       
2x4 = 08 •••••••• {R}                        
2x3 = 06 •••••• {0}                          
2x2 = 04 •••• {°}                            

(1) 4x8 Stretched  (variant)          
(2) 4x7 completed (original)          
(3) 4x6 shrinked   (variant)          
(4) 4x5 feasible   (variant);         
{5} 5x5 common [with 3x5-based cheat]
{6} 3x5 common [n/a "MWmw" 5x5 hacks]
{7} 3x4 common [rather rare 7b ASCII]
{8} 2x5 scarce [I made a 7-bit ASCII]
{9} 3x3 common [but rare 7-bit ASCII]
{R} 2x4 record [mine, as 7-bit ASCII]
{0} 2x3 experimental [vs bin Braille]
{°} 2x2 etc. ditto [partial ASC sets].

The other 3 quote marks aren't consistent with each other...

3 = typo for " ?(*) This brings 2 answers, more seriously (without less ambiguity)… Anyway, thank you for letting me clarify this repeated issue for the novice in this field of minimalist creation. {I always raise an eyebrow when beginning pixel artists or users write 'impossible', 'incoherent', 'record' etc., alone or compared to 'my' project of more than several years…}

* (Of course, if you are talking about the original artwork(?): 1. my explanations are still valid, and 2. Aeolien might reply about her own font design or context of use.)

● Short. Think UNICODE: “quotation mark” ≠ double “apostrophe”.

You believe that 1 'double quote' glyph should be identical to 2 'quote' glyphs. (So, you might mix the invariable code points and their variable representation, perhaps from the non prescriptive shapes featured in this portable document, which is not a reference of the correct glyphs to be used or converted, even not at all in this context of low resolution, believe me!) I would cite the linked Pdf:

« Fonts – The shapes of the reference glyphs used in these code charts are not prescriptive. Considerable variation is to be expected in actual fonts. »

● Longer. The edition of the related glyph(s) is required, logically and visually.

Actually, "='' is a legacy flaw of US-ASCII when the related pair of old glyphs, low-res on the early displays, is ported to our modern layouts or pixel art derivations, i.e. using proportional and low-res dot-matrix fonts. The historical inconsistency pertains to the horizontal dividers, which are of two types: 'characters separation" and 'character separation' (that is 'in-glyph separation' for the latter). At that time, in the early 1960s, the deciders of the ASCII must have thought that the difference was enough for their scope. Besides, the then engineers/programmers were not limited to three vertical quotes on screen, if they wanted to, in theory. Alas, as soon as the early 1980s, when we began to need narrower layouts (more data on cheap/personal video displays) or thinner texts (via pixel art compressions), the original rule was not applicable any longer: from now, both dividers could be the same with the quotation mark (1-dot of spacing in both locations). Later in the 1980, the UNICODE copied the existing ASCII faithfully; but the mind and usage followed too. A vertically-shaped quote is not universal; it's merely 'neutral'. You may be happy to know that you can draw it the way you want, e.g. as I did in these derived fonts (with the deceptive shape of an existing Unicode point: « 2019 ’ is preferred for apostrophe »). At this scale, one can/should imagine beyond the pixels. What matters, is the absence of duplicated glyphs in ANY text font, for the sake of our data. (Else I let people draw 93 vertical quotes and declare it's a working font…)

---

* This said, again, if your question was about Aeolien's 'quote marks' (= quotation marks), do ask her once more, please, xXx XBox Gamer. :-/

@dpla ‚ „ ‛

With fewer words, in proportional 'micro' fonts:

THIS is a quote → ''' ← THAT is a double quote…

(not quite thoughtful, visually.)

@DPLa, I was talking about the quotation marks at the bottom and the reversed floating mark.

*phew!* then sorry, it'd be up to Aeolien to answer. (Since my derived work is based on a single version made a little earlier [on 2017- 08-13 12:46], and she may have edited her font once or twice afterwards - my 'fork' is not related to her changes anymore…)

OK, thanks to your sample text, I can see you are right… but to me, it's another case of priority of creativity in her great catalogue and know-how. I can imagine she wanted this subset to be distinct… (thus inconsistent if the range is the whole punctuations.) Eventually, I cannot say it's wrong. You know, our characters change with time. This is even more true at this scale, where tricks and hack are to be found (else the resolution would not be an issue).

{ On my French keyboard, 3 = " (hence the possibility of a typo)… :/ }

PM'd Aeolien: pending the permission to fontstruct (or not).

Meanwhile…              

4x8 = 18 fonts [#01-#18]
4x7 = 14 fonts [#09-#32]
4x6 = 10 fonts [#33-#42]

Overall = 42 font names:
(compact / less compact)

AeolienGH01dpla48c7b0 / Aeolien_GH01 dpla_4x8 C7_B0
AeolienGH02dpla48c6b0 / Aeolien_GH02 dpla_4x8 C6_B0
AeolienGH03dpla48c5b0 / Aeolien_GH03 dpla_4x8 C5_B0
AeolienGH04dpla48c4b0 / Aeolien_GH04 dpla_4x8 C4_B0
AeolienGH05dpla48c3b0 / Aeolien_GH05 dpla_4x8 C3_B0
AeolienGH06dpla48l7b0 / Aeolien_GH06 dpla_4x8 L7_B0
AeolienGH07dpla48l6b0 / Aeolien_GH07 dpla_4x8 L6_B0
AeolienGH08dpla48l5b0 / Aeolien_GH08 dpla_4x8 L5_B0
AeolienGH09dpla48l4b0 / Aeolien_GH09 dpla_4x8 L4_B0
AeolienGH10dpla48l3b0 / Aeolien_GH10 dpla_4x8 L3_B0
AeolienGH11dpla48c6b1 / Aeolien_GH11 dpla_4x8 C6_B1
AeolienGH12dpla48c5b1 / Aeolien_GH12 dpla_4x8 C5_B1
AeolienGH13dpla48c4b1 / Aeolien_GH13 dpla_4x8 C4_B1
AeolienGH14dpla48c3b1 / Aeolien_GH14 dpla_4x8 C3_B1
AeolienGH15dpla48l6b1 / Aeolien_GH15 dpla_4x8 L6_B1
AeolienGH16dpla48l5b1 / Aeolien_GH16 dpla_4x8 L5_B1
AeolienGH17dpla48l4b1 / Aeolien_GH17 dpla_4x8 L4_B1
AeolienGH18dpla48l3b1 / Aeolien_GH18 dpla_4x8 L3_B1
AeolienGH19dpla47c6b0 / Aeolien_GH19 dpla_4x7 C6_B0
AeolienGH20dpla47c5b0 / Aeolien_GH20 dpla_4x7 C5_B0
AeolienGH21dpla47c4b0 / Aeolien_GH21 dpla_4x7 C4_B0
AeolienGH22dpla47c3b0 / Aeolien_GH22 dpla_4x7 C3_B0
AeolienGH23dpla47l6b0 / Aeolien_GH23 dpla_4x7 L6_B0
AeolienGH24dpla47l5b0 / Aeolien_GH24 dpla_4x7 L5_B0
AeolienGH25dpla47l4b0 / Aeolien_GH25 dpla_4x7 L4_B0
AeolienGH26dpla47l3b0 / Aeolien_GH26 dpla_4x7 L3_B0
AeolienGH27dpla47c5b1 / Aeolien_GH27 dpla_4x7 C5_B1
AeolienGH28dpla47c4b1 / Aeolien_GH28 dpla_4x7 C4_B1
AeolienGH29dpla47c3b1 / Aeolien_GH29 dpla_4x7 C3_B1
AeolienGH30dpla47l5b1 / Aeolien_GH30 dpla_4x7 L5_B1
AeolienGH31dpla47l4b1 / Aeolien_GH31 dpla_4x7 L4_B1
AeolienGH32dpla47l3b1 / Aeolien_GH32 dpla_4x7 L3_B1
AeolienGH33dpla46c5b0 / Aeolien_GH33 dpla_4x6 C5_B0
AeolienGH34dpla46c4b0 / Aeolien_GH34 dpla_4x6 C4_B0
AeolienGH35dpla46c3b0 / Aeolien_GH35 dpla_4x6 C3_B0
AeolienGH36dpla46l5b0 / Aeolien_GH36 dpla_4x6 L5_B0
AeolienGH37dpla46l4b0 / Aeolien_GH37 dpla_4x6 L4_B0
AeolienGH38dpla46l3b0 / Aeolien_GH38 dpla_4x6 L3_B0
AeolienGH39dpla46c4b1 / Aeolien_GH39 dpla_4x6 C4_B1
AeolienGH40dpla46c3b1 / Aeolien_GH40 dpla_4x6 C3_B1
AeolienGH41dpla46l4b1 / Aeolien_GH41 dpla_4x6 L4_B1
AeolienGH42dpla46l3b1 / Aeolien_GH42 dpla_4x6 L3_B1

Should be free for non-commercial use too, Aeolien.

Currently discussing the mixed licenses (in private)… for a common ground. 8-)

Back to the question of Aeolien's extra quotation marks…
When checking side by side, the original font(s) began curving those punctuations, namely in "Gameao reglr" and "Gameao Itlc"… I think it's the limited grid that forced the current pixel artist to keep the extra horizontal bit of pixel, for a consistent design in the typeface (i.e. not limited to the current font: "Gameao Halb"); while you are still right, xXx XBox Gamer: the neutral way would have been to keep'em all quite straight. If you just need this font in this manner, you can edit it via the Clone button (made available on purpose by Aeolien), and share your work (more consistent than 3 glyphs, hopefully).

If somebody needs to CLONE Aeolien's "Gameao Halb" font (v2017-08-23) to match the valid basic ASCII without further trouble, here is an exemple of mine (under the same CC NC SA license as hers):

Aeolien's agreed today:

My 42 derived fonts, all free for Non Commercial use, will be applied a Non Derivative license on the extra work I could provide, in order to maintain the integrity of the whole derivation. A.s.a.p.

Anyone ready for a wave of dpla-esque Gamaeo variations?

Aeo, I dare you to play Minecraft with this font on...

Hi Aeolien/Brynda/members, and FontStruct/FontShop staff!

This Sunday, I'm on the presentation of this derived project

(1 big collage of v16; the log before v7; Wikipedia perhaps…).

Instead of 42 fonts, you'll get a complete typeface (w/o I/B)…

Most of the plain text files are ready; their logics rock-solid.

You get 394 proportional fonts + 3*394 monospaced ones!

That is 1576 fonts in this typeface. All US-ASCII, w/o dupe.

• For the moment, FontStruct cannot display the big sample…

• For the moment, FontStruct cannot help building as scripts.

• For the moment, then, all is going to be released as PNGs…

on my hosting space mainly, else this would flood FontStruct.

---

See you very soon!

Yes. But we have to wait a little more, since a new full day was not enough… (this short project takes dozens of hours all the same, perhaps between 50 and 100, I didn't count!) This Sunday, I could only finish the pixel art + ASCII art (using Aeolien's avatar in 12 grays, with the font numbered 'M2393' - where M2= 'wide mono 4x4'…) + aniGIFs. To do: collage v16, log <v7, making-of images, license text… probably at least a full day. I need a short break. Stay tuned! ;-)

1,576?!

That is a lot for me to create in one day, dpla…

NEWS:
• All the presentation pictures are finished
 (this includes the Wikipedian sample lately).

TO-DO:
1. Append/complete the LOG* text (v16+ & v7-);
2. Selection/(re)creation of MAKING-OF images;
3. Add the METADATA in all Pngs & (Ani)Gifs;
4. Merge the LICENSE in the Txt documentation;
5. Final UPLOADS (dpla, Wikimedia, FontStruct);
6. (No fontstruction before a tough selection…)
   [@xXx XBox Gamer: you may script the whole…]
*
The log might tell us how many hours it took me
{2 months on the paper, including long breaks…}

So: still a few hours of work, and it's over :-)

Thanks for waiting (a couple of days, perhaps)…

NEWS:

Documentation .Txt almost finished… It features some ASCII-art of mine b.t.w., but more seriously, it's probably the first and real technical introduction to what are small pixel fonts from a designer's (our) viewpoint (before my more advanced explanations for much smaller fonts). So, just for this task (updates & artwork), it took me several extra days, unexpectedly, but it was worth the effort, I think, you'll see it very soon (and understand the time-consuming info & format).

---

@Aeolien: yes, on Wikipedia too, since people still need to know what are 'low-res pixel fonts', actually (compression & scope), before writing a dedicated page (cf. e.g. Type Rider's pixel stage in 2013, next nothing new, except always more 3x5-based but incomplete designs, in hi-tech sometimes, in mobile 'retro' apps more often). « Les wikipédien(ne)s apprécieront votre style au passage », though the doc is entirely in (my) English… :-)

Released tomorrow hopefully (I can't be sure at this point, if I wanna sleep…)

FINAL PREVIEW:

(I was expectedly busy all this Sunday, alas! thus this 'small' overview…)

@dpla, kind of hard to see...

can you send a copy to my email???

@xXx XBox Gamer : I just cannot for the -short- moment, since legally, I have to post the related and unreleased files on my hosting space before 'here and there'… (because of the non-commercial licenses.)

So just please wait, I don't know, maybe another 24 or 48h… (If I were not working elsewhere non-stop during 24+ hours, the downloadable files (PNGs) would already be on dpla.fr/fonts, again… sorry. Now, as you'll read in the doc (itself incompatible with FontStruct, too), the next challenge (for the others) is a '1 million fonts' type from Aeolien's seminal font (still in 4x8 max)…

Worse, xXx XBox Gamer, all the 1:1 size samples, fonts, demos need this legal prerequisite. I'm sure you'll be able to wait a little more. It's my case for a couple of weeks (because without solid explanations, people repeat the same ambiguities forever, and our US-ASCII gets forgotten). I hope you understand my/our viewpoint.

“kind of hard to see...”

Yes, 2560x3200 (beautiful on a wide screen, as well as the related demos, esp. the Matrix effect)… ^^

@Aeolien (only): Before I can release it, I leave you all my last private messages as a .Zip archive, temporarily here. I guaranty it's only one safe .Txt file inside. Password protected because of the public post : just use the string of characters used in my last subject (= last PM message, before the ones that got stopped, you know, it's precisely 35-character long). Anyway, I'll delete this file and post the missing parts of this interesting discussion, not before tomorrow, for the understandable reason we have to do with.

---

'News': I am/was on the doc file, but let's say its finished (licenses included, more or less). :/

And yes, my little experiments might help us build betters low-res pixel fonts in the future.

OFF-TOPIC PM @Aeolien: Yes, indeed, a funny film from the past. I must have seen it (in French), perhaps just once, and not at all in this decade. (Some kind of remembrance of the sequence with the crushed car between the both trucks.) I read it was released seven years after the great, yet creepy Duel. Good luck & coffee with my PM'd reply (10/13pp. only)^^

AGH PROJECT COMPLETED :-)

= 1576 US-ASCII pixel fonts

(modular 4-dot-wide typeface).

WEB PAGE (directory):

• Dpla.fr > fonts > micro > derived > agh

ARCHIVE (.7z):

• All-in-one archive /!\ 6 MB

DOC & LICENSES (.txt/.png):

• Documentation and agreement
• Documentation screenshot
• Documentation screenshot small
• Documentation screenshot small tn

FONTS (.png):

• Layout colored
• Layout colored small

• Layout matrix
• Layout matrix small
• Layout matrix small tn

• Layout mono black on white
• Layout mono white on black

• Layout mono who did what

SAMPLES (.png/.gif):

• Sample ascii art 4x4 black on white flickering ani
• Sample ascii art 4x4 black on white flickering still simulation
• Sample ascii art 4x4 black on white flickering still simulation small

• Sample ascii art 4x4 with black grid black on white
• Sample ascii art 4x4 with black grid black on white small

• Sample ascii art 4x4 with white grid black on white
• Sample ascii art 4x4 with white grid black on white small

• Sample ascii art original and decoloring ani
• Sample ascii art original and decoloring magnified ani

• Samples wikipedia(*)
• Samples wikipedia magnified(*)

• Temporary files
• Temporary files small
• Temporary files small tn

* I could not upload to the Wikimedia Commons. They still limit their choice by discarding the “Non Commercial” (NC) licenses from the Creative Commons (CC), besides my AGH project is “Non Derivative” (ND), for the moment at least, alas! (See the documentation.)

I shall not add the support to e.g. extended (8-bit) ASCII characters;

I have no plan to fork even more, to e.g. 1 million of US-ASCII fonts.

3 images should follow…

dpla

Documentation screenshot small.

Layout matrix small.

Sample ascii art 4x4 with black grid black on white.

@dpla - Congratulations on the success of this massive project!   :^)

@Aeolien: the following image already features actual/living micro characters *pun*…
@Goatmeal: thank you for your private message too; you are now bound together *amen*…
---
I'll be back to (3d) isometric minimalism (pixel and voxel art), a.s.a.p.…
before I can resume my main project [of very low-res US-ASCII characters (started in 2013, even in 1988 or before)].

Wikimedia commons 5-font sample pending Aeolien's decision :-)

OK, thank you! (we've just discussed it via e-mail)! A.s.a.p.

Now sampled on the French Wikipedia > Pixel art:

@dpla - What a wonderful sample picture.  The characters appear to be "jumping for joy"!  :^)

Wonderful sample, dpla.

Faster gif of that sample...

@Goatmeal + Aeolien + xXx XBox Gamer.

• Goatmeal: you always understand my/our designs! (we'll check this again, when I can post elsewhere my forthcoming/interesting samples of very minimalist voxel, I'm sure.) My goal was to make the reading the documentation secondary (sample = on topic summary & teaser).

• Aeolien: well, at least my funny experiments could help here, from the early 1990s (demoscene) and 1980s (8-bit computers). The developers cannot be, and never got serious AFAIK (Easter eggs, ripped code, obfuscated bugs, big pixel fonts, and video game culture in our digital generations…) B.t.w., I never credited the online dictionaries in public, so : thank you e.g. Reverso and Wiktionary!

• xXx XBox Gamer: I thought about this maximal speed too (no delay), but it is known to be still largely incompatible nowadays (that is, depending on the host viewer/renderer/system, the GIF animation can hiccup a lot - yours look deceptively fine, but it freezes at times in my dektop browser, besides you lost the 'bouncing smiley' effect). Moreover, the reading on Wikipedia needs less 'disturbance' IMO (it's not quite a place for an actual demo). I understand always more your will of performance, from your comments/posts (a fresh mind that fits FontStruct pretty well, since competition is a good incentive in general).