Satyr: No straight lines

Sindre's picture

This started as an attempt to draw type with curves only, and soon became an exercise in tension. I drew this several months ago, and had almost forgotten about it, until [Bad link] brought it back to mind. I don't really know if this experiment has identity enough to be considered a typeface in its own right. There's a companion italic in the works too, but I want to wait with that.

First specimen41.24 KB
Text setting31.99 KB
Revision A–Z, a–z41.14 KB
Pangrams (thanks, Craig!)39.04 KB
Testing25.33 KB
Italics32.87 KB
Trevor Baum's picture

You sir, are quite a talent.

The stroke contrast is a bit inconsistent - especially in the 'a' and 'e', but it looks like a solid, sturdy text face. How does it print?

Sindre's picture

Thank you for your compliment, sir.

That contrast weirdness with the 'a' (and several other glyphs) is deliberate. Perhaps it's overdone ... I'll look into that.

This is indeed supposed to be a text font, 8 points its intended size. The typeface, if developed, will have additional optical sizes of 10, 12, 18 and 60 points.

I haven't printed it yet, and I'm ashamed to admit that I don't have access to a decent printer.

eliason's picture

Glad you're doing this, and glad you're sharing it here. I love looking carefully at your designs - I think they're all top notch and this is no exception.

Some observations to take or leave:

I would consider dropping the /A/ crossbar a little.
I think the leg of /K/ and /R/ and the tail of /Q/ might be too dark. Maybe in addition /K/'s leg could use more of the straight vs curve tension that the other two have.

/W/'s middle vertex comes too high to my eye.

I'd like to see more variations for /e/. It's not the contrast weirdness that bugs me, but it just looks too soft. Is it worth trying a spur (or maybe just the hint of projection like the beard of /G/, which I love)?

I'm not sold on /y/'s tail terminal. Would a flag terminal like /f/ has work there?

Diacritics look big--particularly tall, especially circumflex and caron.

I like the monoline zero, but think it could be a hint darker.

That four is seriously weird. Not necessarily a bad thing!

Tittles' rightward placement may be overdone a touch.

Favorite bits: /E/ is great; /b/c/m/ are all terrific; I like the "upside down" /s/; /2/ is very tightly wound but wonderfully drawn.

Looking forward to text sizes and settings. I'm eager to see whether that cusp on the right side of /U/ can work without being an eye-catcher.

Sindre's picture

Thank you very much, Craig. It's great to be back. As usual, I think you've found the most important flaws in my design. That tall crossbar in 'A' is just a weird hangup I have (caused by the 'A' crossbar of Zapf's [Bad link]. Will be lowered. I'll make some 'e' variations. I really like its wide open mouth, but I agree the eye lacks character. That 'y' tail is a remnant from an earlier incarnation, where 'f', 'r' and even 'y' had similar terminals. The 'y' must go too, I agree. Are the 'r' and 'f' terminals slightly too light, by the way?

It's well past midnight here, so redrawing will have to wait till tomorrow. Thanks again for invaluable feedback.

Bendy's picture

Wow, Sindre, your talent has really blossomed. My goodness, this is beautiful! Good work my friend. Let me do justice to this by coming back after a full inspection! ;)

eliason's picture

Are the 'r' and 'f' terminals slightly too light, by the way?

/f/ I don't think so, /r/ maybe: text settings will tell.

hrant's picture

This reminds me of two things: my own Paphos*, which I started over a decade ago but soon stopped since I realized I wasn't ready to do it right (I'm still not ready); and Bremer Antiqua**, which is of course a much higher compliment. I like this a lot, and it doesn't have to try any harder to be a "typeface in its own right".

* You can see the "j" at the far right here:


That said, you do have a decision to make: how conventional or not you want this to be. You could end up with a solid, charming, somewhat old-fashioned design; or you could end up with something that will make people (including you) think a little bit (even though it might still seem old-fashioned). The strangeness in the "a" is probably the best example of what might or might not need to be happening.

This -very personal- decision has to come before any specific glyph critiques can really be worthwhile I think.

> 8 points its intended size

Although the ample width might point towards such a smallish size, the vertical proportions are more significant, and there (especially in the longish descenders) I'm seeing 10-11 point.


Sindre's picture

What a pleasure waking up to such praise! Thank you, thank you, Hrant and Ben. This really is the greatest encouragement.

Hrant, this typeface originally had a 'j' very similar to yours, like this:

I abandoned that idea because it made the very usual (in the Nordic languages and in Albanian) digraph 'gj' close to impossible.

I've never even heard of Bremer Antiqua, thanks for that link. I'm looking forward to studying that now.

I want this to be an original typeface rather than a conventional one. And I agree that my 'a' is the ideologically most important glyph. I'll start redrawing things as soon as possible, that will be in seven hours or so.

Ben, I'm really looking forward to your critique.

Bendy's picture

Well well! I can see you've had a lot of fun drawing this! I like the way you've balanced the variables: smooth/cuspy, sharp/soft, lively/restrained. It's very artful, almost every glyph has some unexpected delight! I especially love the Pagan-cross shapes of F and the 4, which is just incredible.

I'm having to go into the realm of vague suggestions since this is already so polished, so be wary of taking my comments too seriously.

A seems somehow a bit plain. I like the spur on the apex of your AE, wonder whether that would work on the A. Also the foot serifs feel a little flat/shallow.

B is perhaps heavy on the left, meaning add a fraction more weight to the arches. Compare with D.

Tail of C looks unresolved. And the tail of J may be an area to play with.

Q: This puzzles me. The overall shape fits well with the rest of the alphabet, but I'm curious whether there is a better solution for the tail. Crossing the bowl?

T arm-serifs look lighter than on E; I'd expect them to be heavier.

On U, I'd be tempted to try bringing the weight down the left slope more.

W: I think the apex needs a slightly wider top, but that would lift some of the weight there upwards, which would not look so great, so also lowering it a fraction might be the solution. In fact that right hand downstroke could do with a couple of units extra weight I think.


a, f, m, q, r, t and u are super. I'd personally agree that the flags of f and r could take more weight, but Craig's also right, see how it looks in text.

c is wide. Just an observation, I don't know what to think of it :)

Counter of d looks somehow oblong compared to the more oval ones in p and q.

e: yes. I'm a real fan of the inky eye, but I'd want an alternate for larger sizes, somewhere closer to the e part of /ae/ and /oe/. (I think these two may be too wide, and the vertical side to the counter of /a/ in the /ae/ bugs me. But I'd happily defer to your familiarity with these glyphs.)

g: I'm not accustomed to the narrow tail, and to my eye the terminal seems a bit far to the right, but I'd want to see how it works in text before recommending any change.

h may be a bit wide, or not.

I'm curious about the placement of the tittles.

I don't know how, but the w seems very much to bear your signature, it reminds me of your previous typefaces, which are actually rather different. I like the way you draw w.

Loving the shape of thorn! Diacritics may be slightly too large (but only slightly, I like the weight of them). Placement of acute and grave may be too central.

Comma may be out of character?

Section sign may be falling left. It's an absolute beast to draw.

I absolutely love the numerals, but some of them feel too sharp/crisp/stripped/modern. Like comparing the 567 with the glyphs in the line above, they sort of seem like a different tone. This is probably connected with what Hrant said about the strangeness. Is the contract too vertical on 5 and 6? I fear the 4 may be too wide, but don't change the shape please! Also I wondered whether you could curve the diagonal of the 7 into a swashy terminal like the 2 and 3. 3 and 5 may need more open lower bowls.

Well I hope this hasn't put you off by seeming like there's too much to's been instructive to look so carefully at what works and doesn't work for me. Thanks for sharing! :)

Sindre's picture

Thank you for yet another invaluable piece of criticism, Ben. Not offputting at all, rather the opposite.

Here's my very first partial revision. I've only concentrated on some of the most important lower case letters. More to follow.

'e' is now more related to 'a'. Have I've done the opposite of what you both suggested? Feel free to slaughter it.

's' is a litter wider, slightly more "upside down" and slightly rotated and fattened, inspired by Bremer Antiqua.

'd' was faulty, thanks for spotting that, Ben.

'g' is somewhat tweaked. I didn't quite understand your advice, Ben. Is this any better?

'f' and 'r' slightly altered.

'c' a tiny bit narrower now.

Except for the 'e' (and perhaps the 's'), these are all rather minor tweaks. Maybe I should follow Hrant's advice and think a lot more on this typeface before I continue drawing. Any way, all your input will be taken into account. Thanks again.

Sindre's picture

Er ... that 'c' really is too wide, isn't it?

1985's picture

My humble input would be that the diagonal stroke of the /4/ is curved in the wrong direction. I'd suggest that it be concave not convex. But only a fraction. Not that there is a right or wrong, IMHO more harmonious.

eliason's picture

I still think it's worth trying to harden that outer contour of e at the right end of the crossbar to a point - make it akin to /G/'s beard (or even, in a way, the famous /U/ cusps). Right now it just seems too soft to me (softer now that you've sloped the crossbar--which, this problem aside, I think works well).

Bendy's picture

Well, yes, that's not quite what I was expecting with the e! Interesting, but for me not quite there yet.

I guess what I meant about g was that the bottom half of the tail wasn't quite long enough...I think it makes the whole shape look a little falling to the left, but that could also be because the top of the tail slopes quite a lot. The latest one feels more comfortable to me, but best seek others' opinions on this.

c: it did strike me as somewhat wide, but that's because I draw rather narrow ones usually. It's perfectly acceptable wide, if you like it.

eliason's picture

Working out the /e/ will help with deciding on /c/'s width.

Sindre's picture

'e' alternatives, one sharpened, one faux-Venetian. Any of these going in any direction? This is darn difficult!

(B is tweaked as per Ben's suggestion, by the way.)

Sindre's picture

Andrew, those numerals are all very first draft. I'm almost certain that my '4' is overdone, but I think such a structure could possibly work. I'll sort those guys out after lower and upper case.

eliason's picture

I would say the top one doesn't go quite far enough, and the bottom one goes quite a bit too far.
Maybe something yet in between, where that outer contour coming down just loses its convexity, but doesn't bump out into a spur.

Sindre's picture

You make it sound so easy. :-)

This any better? (I feel typeblind already.)

eliason's picture

I'm picturing something (roughly) like this.

Sorry if I'm overstepping boundaries drawing on your letters - it's just hard for me to describe in words what I'm picturing.

Sindre's picture

No problem at all, obviously I was too dim to understand your description. I've wound up with approximately that shape several times myself, and must say I'm unsure if it fits my vision for this typeface. I'll try some more.

1985's picture

No problems, Sindre!

Frode Bo Helland's picture

That a bowl is crazy similar to one of my sketches. (Off course) your type is balancing a lot more concepts simultaneously, which IMO often seems to be a cue for quality designs. I'm looking forward to some close inspection tomorrow.

Frode Bo Helland's picture

One thought off the shelf: have you tried a more straight sided cusped e?

Bendy's picture

I'm unable to propose a solution to e, I'm afraid. But I know somehow it's the key to the rest of the font. Once you've nailed it, everything else will fall into place.

hrant's picture

I think adding a beak to the "e" is part of the solution; but a total solution -in terms of the whole font- still seems distant - assuming the current "e" and "a" are in fact being taken to heart. Alternatively you could tame the "a" and "e"* and soon end up with something easy to sell (and "sell").

* In fact the original "e" seems tame enough for me.


cerulean's picture

I really like the 4. You could tone it down just a little bit, but it's a great idea. The diagonal of the Æ seems uncharacteristically straight, and I think it could take inspiration from 4.

Sindre's picture

'e' tweaking is slowly driving me mad ... this beak-shape feels alien to me, though I think it might be right. This typeface it not about doing the obvious.

The 'a' stays the way it is.

Frode, not sure I understand your suggestion, remember to show me when we meet.

Thanks for liking the '4', Kevin. I think it'll be to the numerals what my 'a' is to the lower case. But some polishing is necessary, yes.

Please be candid, Hrant. What would you have done?

Here are three 'e' variations.

Sindre's picture

No. That beak isn't right. This is what feels right to me right now.

litera's picture

/s/ is a bit top large isn't it?

Sindre's picture

Yes, it is. But that's a feature, not a flaw. :-)

eliason's picture

I like the third 'cares' /e/ best, FWIW.
(All the /e/s look a little wide to me but maybe that's just that these words put it in the midst of naturally narrow letters...)

Bendy's picture

FWIW I like the last one you said feels right. If you do narrow it, you might also move the crossbar a little higher to maintain the proportions of that counter.

hrant's picture

I'm not holding back. :-)
I just think any ideological decision is personal so it has to come from you; all I can do is try to help reveal the issue. Until you settle on the spirit of the design any specific glyph critiques risk becoming moot.

That last "e" fits with the rest. The troublemaker -in the best possible sense- is the "a". You might actually leave it that way - for a text font it's not going to jump out; it might even become a conversation piece (like the [in]famous "g" in Meta). Or you could tame it, get over it*, and polish the font away. The hardest thing -again, in the best possible sense- would be to let the "a" trigger a full re-evaluation of the font - perhaps even hibernating the design again...

* It's a maxim of sorts that -for a text face-
one must discard the most beautiful glyph.


Sindre's picture

A different approach: The white of the eye of 'e' now is a morphed version of 'a's internal whitespace. I think this a minor breakthrough. The glyph is a little narrower too, and slightly tweaked in every thinkable way.

Hrant, thank you for encouraging words. That 'a' is the raison d'être of this typeface. So I have to make an even better glyph for discarding.

eliason's picture

Thumbs up!

Sindre's picture

Thank you, Craig! For the idea, for your sketch, and for thumbing up the result. Now, thirty-five minutes later, I'm still satisfied. I will dream about it tonight, for sure.

Trevor Baum's picture

I've always seen the relationship between the counters of the 'a' and the 'e.' I think more type designers should focus on this synergy. Well done, Sindre!

Sindre's picture

Thanks, Trevor. In this case, there's a strong relationship between the other halves of those two glyphs as well. The 'e' needs that huge aperture because 'a' has an exceedingly large one, without any kind of serif. I'm afraid this will have further consequences, though ...

litera's picture

Unfortunately I don't think that /g/ shares the same style with other letters. It seems odd and out of place. The descender simply doesn't seem right. Maybe the upper curve of the descender should be a bit more horizontal as it is now (and maybe a bit less heavy as well).

eliason's picture

Now, thirty-five minutes later, I'm still satisfied.

How can you not smile at it, when it's smiling so big at you! :-)

I'm afraid this will have further consequences, though

Yup, time to reexamine your apertures. Hey, maybe your nice /s/ will be the beautiful letter that Hrant needs you to throw away! ;-)

Sindre's picture

Apertures reexamined. I've tried to inject more whitespace everywhere. The 'g' and the related 's' now has a faint echo of the 'a' and 'e' trick. I've chopped off most of the left-facing serifs, but not to the radical extent of Bremer Antiqua. I think this is going places, perhaps even the right ones. And no, I haven't forgotten all your helpful critique, I'm just following Hrant's advice. What do you think of this new direction, dear critics?

Sindre's picture

I followed your advice on the 'g', Robert, or rather it wound up the way you suggested as an attempt to make it more in style with other glyphs. The main reason for its slightly unorthodox shape is establishing some relation to the 's'. All letters need friends.

litera's picture

It did yes. I think it looks much better now although descender should go a bit more to the left to stabilize it even further (or move the right part a bit more to the left). I very much like the top part descender stroke contrast. Very playful and a bit feminine.

P.S. I never wanted you to follow my advice. I way only trying to point out things that I don't like. It's on you whether you see them the same way and agree or not. My observation about /s/ being a bit vertically unstable still stands but you said it's a feature. So it will stay. :)

Bendy's picture

It looks to me as though the c is falling right and the s falling left in the latest sample.

Trevor Baum's picture

The 'i' is definitely falling left too.

Sindre's picture

I can't understand how thickening 'c's stem by two units can have such an effect, and I can't see it myself. I've however made the 's' more stable by moving its upper part five units to the right (back to where it was, but the new structure is kept). I guess one could say that the 'i' (and the 'l') is less balanced now, but the structure itself is now almost completely different too. What about the Bremer Antiqua if my glyphs are falling? Perhaps I've made a big mistake by doing this. When I've worked my way through the lower case, I'll upload a pdf with text settings. I think that will answer that question.

Sindre's picture

Pdf with text setting attached. No kerning at all, very basic spacing.

Sindre's picture

Er ... I realise my 12.15 pm posting (server time) sounds a little angry, irritated or agitated. That was not my intention at all, and I apologise for its harsh tone. I'm truly very grateful for all criticism. Keep it coming!

Bendy's picture

This is looking grand! I like that you've decided to go with the 'strangeness' or 'inkiness' and would like so see how it translates onto some other glyphs (for example could the right arm inner serif do with a looser connection to the arm to mark another inky change in dynamic?)

Lowercase s looks slightly too light now. You're right it is amazing how much difference a few units can make...quite a puzzle really!

I like the way the square brackets don't even have straight lines!

Syndicate content Syndicate content