Still on E, tried bringing in the curved part a bit to align with the straight stem above a bit more.Do you think C could have its top ball terminal further down to make a more curly top?
You mean like it was before Sindre advised me to raise the terminal? :-P
Gosh, not sure. It just seemed a little restrained in comparison to the new R and E which have gotten quite extravagant now.
I think a little balance is a good thing.
I like that idea. Ignore my tendency to make this all flowery ;)
That's been an interesting part of designing this font: to fit the concept, some letters require all sorts of bending and twisting, and others none at all. I do think I'd want the latter more to counter the craziness rather than match it. But that said, I think the drawing in the C is not yet satisfactory...
I like that the C seems normal in comparison, but if you look at it alone you see that it is a little crazy with the different terminals.
Just a crazy thought on the numerals - how about combining them with the character you get by using the shift key with on your keyboard? i.e, combine "1234567890" with "!@#$%^&*()" respectively. Not sure if this would work, but it makes sense with the combining uc/lc theme...
Interesting idea but after sketching I've concluded that that lies beyond my melding powers!
Besides, it would make sense for English keyboard users only. My Norwegian shift-number row reads !"#¤%&/()=
I've been messing with something like this for a couple of years, but I'm glad to see somebody making a hard effort on the ground! I'll try to read the whole thread and hopefully have some useful input. Keep it up!
Great, I look forward to hearing your feedback, Hrant.
Considering a descending J.
This typeface is shaping very pretty.
My feeling is that it is slowly drifting away from a uppercase/lowercase mix towards a more decorative direction.
Not that it’s a bad thing, mind you.
It looks like the ball of the descending J needs to go under the baseline. Perhaps that glyph is a good candidate for alternates, as the semi-descending J is very nice, but highly problematic. That AJ hole makes too much whitespace, I think.
This typeface has had a marvelous evolution, by the way, you really must finish this.
My feeling is that it is slowly drifting away from a uppercase/lowercase mix towards a more decorative direction.
Yes, it's interesting how some of the more difficult combinations of late have led it in that direction (BDR for example). Sometimes I feel like I'm just along for the ride to see where it goes!
It looks like the ball of the descending J needs to go under the baseline.
Yeah, my reaction was similar. I like the shallow descender space of PGQY but J creates "highly problematic" spacing as you said. Initial attempts to leave the terminal lower than the baseline looked bad, but maybe I should keep working on options.
Thank you very much. I certainly have every intention to see it through!
New glyph: german double s. I had some fun with it. Can I get away with this? :-)
Aw. I read some German, and I must admit that I have a hard time recognising this as a double s. (But it is a cool glyph! Perhaps the second s and the right part of the joining stroke needs to be softer.) The first half is based on the long s in all typefaces I've ever seen, even in upper case versions. I think the obvious solution for a double s is to make the long s descending.
Maybe you could start with an Esh joined into a lowercase s? The curve of the cap part makes the whole thing a bit wibbly to me, though I did recognise it as an Eszett...but I don't read German so Nina might be in a better informed position about this one.
When I said recognise, I meant in the context of reading, not deciphering. The main problem with this glyph, is that it is an ss-ligature, not an ſs-ligature (or ſz-ligature).
Some ß source of inspiration:http://typefoundry.blogspot.com/2008/01/esszett-or.htmlhttp://www.typography.com/ask/showBlog.php?blogID=82http://en.wikipedia.org/wiki/ß
Now THIS is what I call a can of worms! :-)
"The main problem with this glyph, is that it is an ss-ligature, not an ſs-ligature (or ſz-ligature)."
Actually, there are severe limits (IMO) to how much the eszett should be seen/interpreted as a ligature at all. It essentially is an autonomous character; and it's not even 100% clear that it was a ligature to begin with (much less which one). So I'm really not sure connecting two "S"s would fully do justice to the character. (And so far yours didn't parse very well for me.)
I can't really offer any ingenious ideas right now, but as to inspiration: http://typophile.com/node/33647
(But I see you're on that thread, Craig.)
It essentially is an autonomous character
Yes, I know, the same is the case with Norwegian æ, ø and å. But I was thinking of the physical construction of the eszsett glyph rather than its evolutionary history. Isn't it the case that it always starts with a long s-shape, and never with a short s?
Hmm. The concept here, in keeping with the concept of the font, was to make a glyph that melds "ß" (the lowercase double-s) and "SS" (the form it often takes when uppercased).
Oh. Well but since there finally IS an uppercase form, why not try fusing "ß" with u+1E9E? Scary, I know. :-)
Sindre, yes it starts with something like a "longs". Historically the second component is unclear (IIRC it might originally have been a "longs" with a modifying squiggle of sorts, not necessarily a ligature – but I'm no expert).
:-) If that's not coming through I'll have to rethink it.
Well but since there finally IS an uppercase form, why not try fusing “ß” with u+1E9E?
I suppose that's my next step. In the meantime I've pushed my first attempt around a bit and wound up with this.
Back to B - is #2 too swashy?
Even though I always see #1 as a upside down я, I still prefer it to the other two. (Yes, #2 appears a little too “swashy” and #3 appears somewhat disharmonic.)
I think I agree.
Here's another go at a descending J, this time with the descender reduced dramatically so that it can tuck under the preceding letter.
And here's an attempt at a german double s that incorporates some of the controversial capital form (new design is at the bottom). Is this way too wide? Boy I find it hard to judge this character.
I think the last eszett is by far the best one. Yes, it is wide, but I think you could make it a little narrower on the right side. Perhaps that vertical stroke could be a little thinner?
Another solution would be using a foot serif and a straight descender, like the P, with the right side of this eszett.
New J descender terminal looks a little too small to me, why not use the same descender as your new eszett?
Slightly narrowed ß, and a version with a P-like terminal:
Actually, I did build J's descender by enlarging (mostly widening) ß's! (Different scales of samples above may make it seem otherwise.) At any rate, here are some options for the J descender:
1 is the version posted previously
2 is the form taken verbatim from the ß.
3 is #2 with the ball enlarged a bit.
4 is #1 with the ball enlarged a bit.
I think #3 may be best.
Actually, I did build J’s descender by enlarging (mostly widening) ß’s!
Aw, I'm embarrassed.
I agree that J no. 3 looks best, the ball mustn't stray too long from the stem, as I think is the case in nos. 1 and 4.
I think both new ßs could work, but perhaps there is a slight conflict between the two serifs in no. 2? Could the bottom of the right part have a ball terminal, perhaps?
I too prefer the old B to your new ones.
Another crazy thought - I'm guessing this is an all uppercase/unicase, so how about inverting what you currently have for lc or alternates? Like on your 'T', you have uc on the top half of the character and lc on the bottom half. So on the alternate or lc character, it would be lc on the top and uc on the bottom hallf.
This is beautiful! I reminds me of some kind of modernized ancient writing, then upon a close look. Wow - Awesome!
Thanks Kermit. I've had the same thought about generating a whole "case" of alternates, but I'm finding that for most letters the satisfying ways of combining forms number either one or zero...
Here's a new treatment of the E, particularly the middle.
Top is as before. Middle and bottom show new middle, which takes more from the lowercase form. The only difference between middle and bottom is the lower terminal. And here's those two with the serif-like bit in the middle removed:
And here's some more tries at ß, trying different terminals on the P-like ß. (first two are as before)
I really like the second E on this page and prefer the two Eszetts with the fishhook terminal.
I like your tries to adding back some “lowercaseness” to the E. The risk here is that it could easily become too “busy”.
As for ß, it’s a tough match between #1 and #3, but I think #3 flows better with the rest of the letters.
Old E for me. Yes, I'm a boring guy. Eszett no. 3 is my favorite. (Lye-eating is actually very common where I live, à propos.)
(I knew I should have numbered these!) For clarification and consideration, here are the E's.
Bendy, you prefer "number 4" and Sindre "one," is that right?
"Eight" is new but is probably too c-ish. Any merit to "nine" or "ten," also new?
THREE (even if it’s a little THRFF).
I’m no big fan of the bottom left “bump”.
Four's cool because that chopped hairline in the centre echoes the chopped diagonal on A. This is really going in a nice direction.
I find the bottom right corner of N unconvincing. Did you try making the diagonal join to the vertical in any different ways?
Some more bumpless options for riccard0:
Eleven is nice! :-)
That came from trying to force the cap diagonal into a semblance of a lowercase hump at the top left, which automatically pushes that stroke away from the left vertical. Consequentially, I felt I had to make the lower right connection as broad as possible so that the counters would be roughly even (i.e. so the right vertical wouldn't wind up way too close to the diagonal, relative to the left vertical).
But maybe the unconventional counter balance is less troubling than the unconventional lower intersection...
This is a persistent dilemma. I don't think knocking the verticals off vertical will work. I've thought of dropping the beginning of the hump or the entire left side of the letter down to "x-height," but that also reads very strange. Hmmm...
Eleven is nice!
Yeah, I kind of like that one too!
I dig eleven too. It is undoubtedly an E (unlike that swashy F-lookalike, and is truly ingenious in its case-ambivalence. (By the way, would "Ambica" be an even better name than "Ambicase"?)
Man, you're what I'd call prolific…! :)
I still like the old "E" (ONE), but the one in ELEVEN is pretty smart (and I think works) as well.
Re your Eszetts, you're pretty close to a (IMO very good) capital shape, so I'd probably prefer the one with the descending uh, hook (first one in your 25.Aug.2009 9.25am), which gives it back some lowercaseness.
Thought I had the answer to N here, but I was more convinced by my sketches than by what I could do in Fontlab. Here they are for your consideration, along with some thoughts on M.
For N, 0=current design, 1 & 2=variants on my new idea.
For M, 0=current design, 1=similar to what Ben proposed ages ago, 2=similar to what I had ages ago but with a thick left vertical.
(Note that these N's and M's are not paired in any way - one of the issues of these variants is that none offer an analogous solution for the other letter.)
0 = best from before
1 = same with unidirectional serif
2 = nondescending, spurred version
3 = hooked version similar to the one Nina preferred, but narrower and with two ball terminals.
N, M: #0 remain the best options.
I don't think it's necessary for the M and N to follow the same line of thought, and actually I really like the zero version of M. Would a pi-shaped N be out of the question?
Would a pi-shaped N be out of the question?
I tried it before - see my 27.Jul.2009 8.13am post on the first page of this thread, the lower part of the .gif that reads MEN AND WOMEN ADMINISTRATION. But here's another go that's probably an improvement on that:
And here's another try at capturing that tailed N:
And here's the same with an angle at the bottom.