Walleye (added: plea for Cyrillic critique)

Number3Pencils's picture

[Updated]

Hello everyone!

I made a new font. I would call it a text font somewhat in the vein of Electra. It has heaps of OpenType features and it supports lots of languages, including Greek and various Cyrillic-alphabet languages (as exotic as Mongolian).

How does it look? I could especially use critique on the Cyrillic and Greek alphabets, but all critique will be useful for me.

AttachmentSize
WalleyeStarts.pdf174.26 KB
WalleyeStarts2.pdf176.74 KB
WalleyeBig.pdf692.75 KB
RomanArchLeadin.pdf38.4 KB
WalleyeBig2.pdf696.48 KB
Stannum.pdf36.18 KB
WalleyeBig3.pdf696.77 KB
WalleyeBig4.pdf705.39 KB
WalleyeBig5.pdf707.96 KB
Upsi.pdf68.04 KB
WalleyeBig7.pdf718.45 KB
JustCyr.pdf82.46 KB
JustCyrBold.pdf79.67 KB
JustCyrAll.pdf133.07 KB
JanekZ's picture

Hi,
my (dilettante) impression: cyrilic roman: UC "Pe" a bit too wide, lc "yeru" (61) a bit too loose
cyrilic italic UC "Pe" again too wide, "ze" too narrow, "zhe" too dark and it doesn't suit your design (vide "Lazurski" type)
Very nice design!
Best
J

Number3Pencils's picture

It took me a while to get around to it, but I addressed what you said, Janek. I didn't change the form of the zhe, because I'm not convinced it can't work, but I did tweak it a bit, mainly flattening the top and bottom of the middle stroke.

Besides those little things, I also designed some old-style figures. There's a new pdf up top. Anyone have any thoughts? Are any of my letters too distracting? I feel like I might be pushing it with some of them, like perhaps /a/, so it'd be good to know if someone else thinks they all work okay. I'm open to opinions about the disconnected bowls on B, P, and their Russian and Greek kin.

eliason's picture

This is looking good. /a/ works for me. I think roman /W/ is splayed too wide (I'd make the bottom vertices closer together), and /y/'s tail is a bit too long for my taste (maybe /j/'s too, but maybe not).
I wonder if the bracketing on the serifs on the thick strokes should be trimmed back a bit, particularly on the capitals - to my eye, you're getting some dark spots there.
I think the disconnected bowls in the Latin work fine. /R/ (like /K/) has some sassiness to it which it would be neat to see work. Right now, it (/R/) might be a little dark overall.

Number3Pencils's picture

I think you're right about /W/--that's been bugging me a little bit for a while. You may also have something with the bracketing of the serifs, and I don't know if I would ever have thought of changing that. I'll definitely try out these changes.

Number3Pencils's picture

Alright! I'm back.

Since the last post I've done lots and lots of work on Walleye. I've created a bold and bold italic, and I've made what's needed for supporting polytonic Greek, and I've coded many OpenType features, and I've done some rethinking of various characters. It's all shown in the new specimen that I'm attaching at the top (WalleyeBig.pdf).

What do you think of this? I have to admit that when I started this two years ago, I didn't think it would get to be nearly this big of a project. I also didn't do any serious research into readability, Cyrillic design, or the history of the Greek alphabet. All my learning was done on the fly. So it's a miracle if there are no glaring mistakes in this font.

One of my big fears is that the design itself is too weak and I should've waited until I had a stronger idea before sinking two years of my font designing energy into this project. Yet I'm this deep already, so I might as well polish it into something usable. Don't be afraid to offer rather sweeping criticisms.

As before, I'm hoping especially for someone to let me know whether my Cyrillic and Greek alphabets are just totally way off, since I don't actually fluently read those alphabets. But I would also really appreciate any and all comments.

Thanks a bundle for looking!

hrant's picture

I don't know what the problem is - this seems quite complete and mature to me! And it has charm. The biggest problem I'm seeing is the Italic - it seems really unhappy, but I can't put my finger on why.

I'm no expert in either Greek or Cyrillic, but the "ksi" stands out to me as being malformed. I think the head is way too large.

hhp

eliason's picture

Looking good!

The challenge you've set for yourself particularly in the italic is summed up in the cap /S/. How do you work together the more "typographic" details with the "pennish" qualities like those thick blunt unseriffed terminals that you want? In the italic /S/ it starts with a "typographic" terminal serif at the top right, then switches gears around the first bend and becomes looser and more "written" looking. Maybe this ambivalent character is what Hrant is seeing as "unhappy." That /S/ is a little Frankensteiny.
Perhaps the cap italic serifs should try on a less formal form?

Italic /Q/ (full-size and small caps) tail may be a touch too dark.

I love the italic /Y/!

Is italic /H/ a touch too wide?

I would try revising all your cap /M/s to give more space to the lower side counters and less to the upper middle counter. This looks especially strange in italic small cap /M/ but might need addressing in the other forms too.

Superscript/subscript figures could be bigger to make them easier to read.

I'm not confident enough in the alphabets to offer critique of the Cyrillic and Greek (though I will say that I see dark spots caused by your italic /ж/).

HVB's picture

Very nice looking and cross-language consistent! I don't read or speak any languages that use the Cyrillic alphabet, so in practice this may not be a problem, but I find that the very thin horizontal strokes combined with the tight letter spacing makes combinations of two or more instances of Tse, Dzhe, Sha, and Shcha confusing. They all seem to run together as a series of IIIII's.

Number3Pencils's picture

Thanks for the comments! It's good to know it's an okay design. I think familiarity has been breeding contempt for me lately.

Hrant, when you said the italic was unhappy, my first thought was this: Maybe you're picking up on the detailing of the lead-in serifs for lowercase letters. In the roman those strokes are all convex, but in the italic I've somewhat inexplicably made them concave. I think convex looks more optimistic. But I've been thinking concave might actually fit better with the design, so I've started considering changing the whole roman to concave serifs. I haven't mocked that up yet, but when I do I'll post it here.

But it could be you're seeing what Eliason is seeing. Eliason, I hadn't thought of changing the italic serifs, but it's an interesting idea. I'm just having trouble picturing it in a way that would leave the italic looking like it's still related to the roman. I kind of have this thing where all the italics I design are very different from the romans, so I always cling to every remnant of similarity they have so they won't look like I just made two different fonts. The capitals are where those similarities abound most. If I can't get a different serif to work, maybe I can fix the S by making the banner less prominent (shorter, a bit thinner on the right).

I'll work on the Q and M, and take a look at the H (I actually thought it might be too narrow). And I'll check out the italic ж and the general spacing of the roman Cyrillic. I should study up on italic ж's.

JamesT's picture

It looks really good! My suggestions would be to look at the lowercase /zhe (particularly the italic) as they seem a bit too dark. Regarding the italic, maybe try lowering the joins in the /h/n/m/r? I think having a bit more white space there would help.

Number3Pencils's picture

Looking a bit better? (Before and after.)

1996type's picture

This looks like a really well crafted text typeface! I only wonder if the somewhat exentric k, s, and some others might scare off buyers...

JamesT's picture

That seems better. It's hard to tell when it's not in running text like it was in the pdf. What helps me is to look at a block of text and see if any letters consistently stand out.

Number3Pencils's picture

Yeah, I was thinking as I posted it that four letters really isn't enough context to judge. I'll post a new pdf when I've done a few more revisions, enough to warrant making a whole new pdf.

Number3Pencils's picture

1996type - I would think the /a/ would scare them off before those, but it's hard for me to look at this through a buyer's totally fresh eyes by now. Anyone else have thoughts on whether these stand out as odd?

Number3Pencils's picture

I put a new pdf at the top (RomanArchLeadin.pdf) with some ideas I've been kicking around for a bit of an overhaul on the roman. It's two possible sweeping changes—a concave lead-in serif for bdijklmnpru, and an ironed-out arch for hmnu. The serif change turned out to be more of a quick mock-up than what the actual change will look like if I do it. Anyhow, anyone have any opinions on which one's best?

1996type's picture

I prefer old, but it's just a matter of taste, I guess.

1996type's picture

edit: double post

eliason's picture

I think I too would lean to the older shape. Those cool but tricky blunt snipped terminals (on /k/ arm etc.) fit better with the convex shape IMO.

Number3Pencils's picture

I'm coming around to the same conclusion. I might actually stay with the old arch, too. I made it early on before this project taught me a lot about beziers, so its curves are hacked together ungracefully, but I think even so it might still be better than my new try.

Catharsis's picture

Very pretty, with a satisfying chunky texture.

The only think that bumps my eye is the |y|. It looks good in small print, but close-up I find it rather ugly. But maybe that's just the bit of eccentricity that lends personality to a font.

Number3Pencils's picture

Is the tail too long for you, or is it the join between the tail and the body that's catching your eye?

Number3Pencils's picture

New pdfs at the top—WalleyeBig2.pdf and Stannum.pdf. In Stannum, I try out a slightly different S (probably not changed in the right direction, I'll keep thinking about this) and /hmn/ with deeper arches.

WalleyeBig2 shows everything. Things I've changed since last time:
—Evened out the white spaces inside all the /M/s.
—Added contextual alternate for f before things that come above x-height and shouldn't be collided with.
—Fixed old-style figures' heights so they're all on a level with small caps. For some reason I had them all miscellaneous.
—Lightened tail of the /Q/s.
—Enlarged sups, infs, numr, dnom.
—Shrank all the serifs on the vertical-stem-based Cyrillic letters, after discovering that they were alarmingly big and flappy. Relatedly, spaced these letters out a little looser.
—Changed xi and ж. Looking better?
—Changed the methods of the gj, jj, gf, and jf ligatures. Now the second letter is replaced with a special alternate that's made to fit with the preceding one. So now it's handled through calt.
—Various tweaks here and there.

eliason's picture

Ooh, I find that new Stannum /S/ much more convincing. Much better!

praitsidis's picture

Hi,

I had a glance at the Greek and it looks very well done! I think your uppercase letters seem absolutely fine.

For the lowercase, the only things that stood out for me are:

  • I found the "bar" (?) on the "ε" (epsilon)a bit too sharp and too contrasted to the weight of the rest of the glyph.
  • The "κ" (kapa) is too open at the top. Would consider closing it up a bit.
  • The left descender of the "χ" (chi) seems a bit heavy and too low. The whole glyph feels biased to the left.

Hope this helps!

Number3Pencils's picture

Thanks! Your advice, and a bit of other advice I got in an email (thanks Giorgos), have helped.

What seems to catch everyone is the kappa. I went back to it and tried making it in some different ways. Am I on the right track with any of these? The first one is the original.


I've done some work on the chi too. For the epsilon, did you mean the top, middle, or bottom bar?

praitsidis's picture

The upward stroke in the 3rd and 4th samples seems much more natural for me. I prefer the third one. I'd even close it up a tad more, personally.

For the epsilon I was talking about the middle bar. Just seemed too thin...

Number3Pencils's picture

Alright, I'm back (see WalleyeBig3.pdf up top), and I've kerned. I think I've also finally solved the kappa problem. I improved a few other Greek characters too, including but not limited to the ones Giorgos and Praitsidis mentioned. I think I'm juuuust about done with this font, but I still would like critiques of what I might have left to do, and of the kerning if you notice anything wonky.

The other two fonts I've made are being sold through Veer, but Veer isn't taking on new fonts anymore, and anyway they don't really target a diverse-language market. Does anyone have an idea of what company I could best sell Walleye through? By default I thought of MyFonts, but I don't really know a lot about what else is out there. (I kind of just hunch over my computer and move pixels around without thinking a lot about the market.)

eliason's picture

Looking great. Some random observations:
Where acute accents are followed by ascending stems it gets a little crowded.
Italic /p/ is too heavy along the mean line.
In the roman, is /hy/ kerning too loose? Is /Wa/ kerning too tight? Have you done caps to caps kerning? In the string of italic capitals it seems rather uneven (e.g. /R/ looks closer to /S/ than /Q/, /G/ looks closer to /F/ than /H/).
Any way to open up the lower counter of bold roman /k/? Maybe trimming back that stem serif?
I still think italic /Q/ tail is too dark, especially in the bold. And/Or its bottom bowl could be another place where you take a little weight off.
You might raise that lovely upper right part of italic /v/w/y/ a hair. Consider raising the top of roman /t/ a touch too.
Bold roman /g/ looks too tall, like it is popping up above the mean line.

Number3Pencils's picture

Thanks for the comments! Sorry I didn't reply—I was a bit burnt out on font stuff. But now (WalleyeBig4.pdf) I've put in most of the changes you suggested, Eliason. (But I forgot to play with raising the top of roman /t/ until I'd already made this pdf, and didn't want to generate the font and pdf again.) Good catches.

I might need to lighten the bold italic Q tail still more. And I might have a little work left to do on all-caps kerning. I didn't really put much all-caps text in the specimen, but if anyone sees any dodginess in the caps stuff that does exist, feel free to call it out.

I couldn't figure out any good solution for the í-ascender collision, so I moved the acute a tiny bit and decided it might have to just be a little claustrophobic.

This pdf has a little bit more detail on the faux ligatures, on the character set pages.

eliason's picture

Hmm, I didn't see this before, but I think your bold and bold italic letterspacing is too loose.

Number3Pencils's picture

Are you seeing that in the Kafka on the Shore bit, or in the waterfalls at the end? I just noticed that InDesign seems to have messed with the letterspacing in the waterfalls in order to justify the text. (I'll fix that problem for the next pdf I put up.)

eliason's picture

Ah good, I think it was just the screwy justification I was seeing.

Number3Pencils's picture

Sweet, I was worried for a little while there.

Number3Pencils's picture

I guess I'm getting just about close enough that I can call the project finished. But before I make that judgement call, I really would like to hear from someone who reads in the Cyrillic alphabet. Is there anyone out there who can give me some critique?

(There are few enough Greek type designers that I was able to find just a few people to ask personally, but the Cyrillic-reading countries seem to be full of type designers, and I would feel weird just picking random people.)

Thanks!

hrant's picture

Definitely do not release non-Latin fonts without at least some basic native help.

hhp

Number3Pencils's picture

My thinking exactly.

HVB's picture

The first glyph on the first page of the pdf (Walleye4) exhibits a problemwith the W. It appears to not be solid, but a combination of overlapped outlines, which should be combined. Under some circumstances the W doesn't display as intended. Note the white overlap in this screen capture:

On my system, looking at the pdf in Acrobat V7, it exhibits this at magnifications of 100% and less, so it's not a problem with contour directions.

Number3Pencils's picture

Yeah, that's because I still haven't merged all the outlines and flattened my contours. I found a macro that's supposed to do all that when I generate the font, but I can't figure out how to make it work. I'll figure it out somehow or other before the final product though.

Number3Pencils's picture

I figured out how to use BetterGenerateFont, and I've uploaded a pdf (WalleyeBig5) with all the outlines flattened, plus a few small fixes, mainly to kerning.

At this point I'm not really sure where to turn for advice on the Cyrillics, since my hobbyist's budget wouldn't really fit a professional consultant in it, and I feel odd about contacting a Cyrillic-reading Typophile out of the blue to bug them to take a look at some text font. All I can think to ask is, if no one looking at this reads Cyrillic, does anyone know (of) someone who might welcome such an out-of-the-blue email?

Other than that, if you see anything in the new pdf that I haven't caught yet, please do tell me. Thanks!

hrant's picture

I know that Alexei Vanyashin runs free workshops here and there, and I've suggested that he latch on to type conferences. You could add your voice, and if you really are in South Korea trying to make ATypI HongKong click might be just the ticket.

hhp

LexLuengas's picture

After a brief look there are a few things that caught my eye:
—The leg of the small-cap «R» seems a bit dark to me.
—The top diagonal of the small-cap «K» is hanging a bit too far to the right.
—The bottom opening of small-cap «W» is more open than the lowercase and uppercase versions.
—Crossbar of «H», again the small-cap instance, is somewhat high.
—The shapes of «upsilon» and «psi» can definitely look better, comparing them with what you did with the other greek characters! What you did with the italics looks ad hoc.

—What I really like is the italic-roman relationship and their dissonance in width.
—«X» and «Y» are original, as well as «R».

Overall I think you really achieved a consistent and distinctive design, which are perhaps two of the hardest tasks a typographic designer can strive for. Enviable ;-)

Number3Pencils's picture

Hrant—unfortunately, by the time ATypI happens, I'll be very far away from both South Korea and Hong Kong. I would like to go to a type conference someday, but I'm just barely getting out of the "poor college student" phase of my life, and into the "vagabond hitchhiker" one, neither of which are very well suited to attending things that are in other continents and/or have an admission fee greater than zero, so I might have to wait a while for the right opportunity.

Lex—Thanks for all the critiques! I think you've caught a lot of holdovers from when I was just starting this project and hadn't learned as much, for instance the w.sc and the upsilon and phi. I'll get on these post haste. As for the upsilon and phi, are the roman ones looking okay while the italic ones are wonky, or do you think the roman ones need to be rethought too?

It's really good to hear that I'm hitting my target of consistency and distinctiveness. I also think it's interesting that you like the interaction between the roman and the italic, since earlier people were saying the italic was too unhappy for the roman. I'm happy with it too.

hrant's picture

As an aside concerning conferences, some of them take on volunteers and let them get in for free/cheap.

hhp

Number3Pencils's picture

Cool—next time there's one that I can get to at the right time, I'll check that option out. Sounds like my kind of way to do it.

Number3Pencils's picture

Revisions. I forgot to save the old versions of h/k/r/w .sc and of the italic upsilon/phi. But I did save both versions of the roman upsilon/phi; the top version is the old one. I think they're looking better... what do you think?

LexLuengas's picture

The small caps look splendid!
—The joining at 7:30 of «upsilon»/«psi» is still too abrupt. You have a pen rotation of almost 90° that happens precisely at the bottom left, where both strokes are thin. In your latest PDF-Sample "WalleyeBig5", regarding the italics, that problem is inexistent.
—Also, I noticed the shapes of «upsilon»/«psi» in the roman diverge from the bold! Settle on that ;-)
—Bold «epsilon» may be to tight.

LexLuengas's picture

The small caps look splendid!
—The joining at 7:30 of «upsilon»/«psi» is still too abrupt. You have a pen rotation of almost 90° that happens precisely at the bottom left, where both strokes are thin. In your latest PDF-Sample "WalleyeBig5", regarding the italics, that problem is inexistent.
—Also, I noticed the shapes of «upsilon»/«psi» in the roman diverge from the bold! Settle on that ;-)
—Bold «epsilon» may be to tight.

Number3Pencils's picture

So, you're thinking that something more like this upsilon would work better?

hrant's picture

http://www.typecon.com/archives/1374
Although I don't think there'll be any significant Cyrillic action there. Even so it's worth the trip - TypeCon is very cool.

hhp

Syndicate content Syndicate content