Eau de Garamond — a sans distilled from the essence of Garamond

Christian Thalmann

Not sure whether that's not a step back. The /s feels a bit alien now, as if it were stolen from Skia.

Christian Thalmann

Meanwhile, I've started to goof around with Greek. Is that /Xi too narrow?

Ori Ben-Dor

Take my comments on Greek with a grain of salt, but to me Eau de Garamond is about bringing the elegance of serif types such as Garamond to sans serif, and I've always thought /Omega looks more elegant when the bowl joins the horizontal strokes directly, becoming more diagonal and less curved as it approaches them. You know what I mean? I'd try that model.

I also feel that /Psi isn't as elegant as other letters, though I don't know what to do about that. Maybe you should play with the curvature, or make the outer stems a bit diagonal.

Also, isn't /Pi too wide? Is it usually this much wider than /Lambda or /Delta?

Craig Eliason

Might be worth mentioning that in his Grec du Roi Garamond did use a two-joint Omega and curvy Psi (but also a quite wide Pi). 

Craig Eliason

Meant to say "and curvy Upsilon."

Nathan Zimet

Bottom corners of the delta are sharp corners but the bottom corner of the Z is not

Kent Lew

That’s not uncommon. (Delta corners vs squared-off Z/Zeta)

Christian Thalmann

I based the /Delta off the /A and /Lambda, but I do see how the /Zeta might be a better reference for the topology of /Delta. I'll have to think about it.

Curvy /Upsilon: Not sure how that would work with the other letters; it would be the only «swashy» capital far and wide...

Craig Eliason

Worked for Claude!   

Christian Thalmann

I feel like the curved /Upsilon just makes it even harder to balance out with other letters.

Meanwhile, some lowercase hopefuls:

EDIT: Actually, wouldn't a lunate /epsilon fit the other letters better? Then again, it feels more clinical and less organic. What's the preference among Greeks?

Nathan Zimet

I drew this a while ago for fun idk if its of any use or any good   Just the bold weight.

Craig Eliason

/lambda needs to curve. /pi looks too stiff. Extenders look too long (and needn’t match the Latin metrics, I think). 
I usually imagine a cursive /phi like that going with cursive /kappas and /thetas. 

Linus Romer

My eye is used to Greek lowercase that is slightly calligraphic and does not have so many straight elements as yours (so: less clinical more organic). I like the variants by Nathan Zimet. I would prefer your alpha, gamma, zeta, xi, pi and chi in Nathan Zimet's style.

Leonardo Di Lena

Also the /mi is more common in the Nathan Zimet form, with the right leg curved. The /omega is too stright, it should be a little bit rounder. As for the /s, I too preferred the previous one

Christian Thalmann

I like the look of Nathan's proposal per se — it's certainly very consistent and very Greek — but I'm not convinced how well it would fit with the rest of Eau. The almost-but-not-quite-straight lines and rather strong modulations combine into a somewhat «gooey» consistency at times, which is at odds with Eau's minimalist style. Might it be possible to find a way to express the organic Greekness of Nathan's letters with the established visual language of Eau? I'm worried that a clean look and Greekness are mutually exclusive...

Linus: Your statement is ambiguous, but I take it to mean you'd like me to import Nathan's style into my font for these glyphs?

I'll do some experimentation and get back to you. I do intend to use more cursive forms for the Italics and stick to a clean look for the Roman, but I suppose I have to treat the Roman as an upright cursive, too.

I'll make a print-style /phi for the Roman and save the current one for the Italics, though.

Christian Thalmann

Is this going in the right direction?

I loathe the drooping top of /lambda, but I guess I'm not getting around it.

Ori Ben-Dor

Christian Thalmann said:

I loathe the drooping top of /lambda, but I guess I'm not getting around it.

Maybe if you slant it right a little it won't look as droopy?

Nathan Zimet

What are you using as reference for your greek?

Craig Eliason

/eta could use more of a lead-in curl. How would /kappa if its leg built off its arm instead of vice versa?

Linus Romer

For me, this goes in the right direction. The gamma looks mirrored to me (I think the stroke angle at the bottom must be mirrored horizontally). The top bar of zeta is a bit too long. The upper part of the lower bowl of epsilon ist a bit too thin. I feel the lambda does still not look right, but I cannot tell you why.

Could you add a comparison Latin/Greek?

Linus Romer

Regarding the lambda: Perhaps the stroke width should increase to the upper end? (That would then be compatible to delta and zeta.)

Christian Thalmann

Nathan Zimet said:

What are you using as reference for your greek?

My misguided and stubborn mental conception of it, for the most part.

Linus: Do you mean the gamma should have a continuous stroke to the right and a branch to the left? I have the impression that the stroke from top left to the nadir is the defining characteristic of the LC gamma and sets it apart from the /y... might be wrong, of course, but at least I see the same concept in Nathan's approach.

Christian Thalmann

Update:

Exported OTFs are available on GitHub if you want to take it for a ride.

Craig Eliason

I think just changing the angle of the shear at the bottom of the gamma will help. You have it matching all those other descenders, but unlike in those letters it represents a change in direction rather than the end of a stroke, so a \ rather than / terminal would help lead the direction back up into the right arm.

I would consider a counter with a corner for rho (like delta and sigma have).

(I'm now working on only my second Greek design, so add salt to my advice as warranted.)

Christian Thalmann

Good suggestions! The «wrong» cut on the foot of /gamma goes against my intuition, but it does make the whole thing look more natural.

I've also given the downward diagonals in /gamma/nu/chi a bit of pen angle at the top; I think that helps.

I'm a bit worried now that the fiddly onstroke of /eta is out of place, given how it's the only one of its kind. Maybe the equally fiddly outstrokes of some other letters justify it, though.

Linus Romer

Regarding gamma: Craig pointed my point out 😀

The mid of the epsilon ist now a bit too low.

I like the idea of the new lambda (especially the low curl). The curl could be even less - such that the end heads slightly to the upper left.

Consider shortening the descenders of phi and chi.

I already like your alpha, delta, iota, kappa, mu, nu, omikron, pi, tau, upsilon!

Christian Thalmann

Here's a comparison of Latin and Greek running text. They do feel quite a bit different to me, as if the Greek x-height were lower. Maybe that's because the Greek has more weight in the ascenders? Should I perhaps lower the ascender height for the tall Greek lowercase letters?

Also, that /xi is too wide. I'll fix it...

Christian Thalmann

That Greek copy looked a bit crowded in comparison to the Latin, so I widened some of the sidebearings a bit. The /lambda is now less than horizontal in the top.

Christian Thalmann

BTW, while grabbing that copy from Wikipedia, I chanced upon a nice name under which to eventually publish this family: Ysabeau, the name of Garamond's (second) wife. It's an awesomely beautiful word to begin with, and it's also full of symbolism, being Garamond's counterpart and complement.   It's late in the alphabet, but since I'm not selling the family, it should be less of an issue...  And as a tip of the hat to its working title, it even contains the string «eau».

Nathan Zimet

Yours is good already but i'm having fun.