My take on Garamond

Daniel Benjamin Miller

Here's a sample of my new font (unnamed as of yet) — Note that /ffi etc are not present yet. It's my own personal take on Garamond, meant to align with twentieth-century Garamonds. It is particularly modeled after Tschichold's Sabon (specifically, the phototypesetting version from the 1970s), though my typeface is not striving for 100% authenticity to any model. I'm less interested in being accurate to the original tradition of the renaissance, and more seeking to mimic the 20th-century versions of Garamond. But I have also made other choices in my interpretation. For instance, I have left the terminals flat for all the characters. I also seek to keep my letters as consistent as possible with one another, using the same widths and components to form much of the alphabet. I have not yet designed an italic/bold version, but these will come in the future. (I am considering following the principle of the original Sabon — though not all the digital adaptations — and designing an equal-width italic and bold. Maybe yes, maybe no: it's an interesting principle to me.)

Why another take on Garamond, you may ask? Well, for one, I'm a fan of Garamonds. But I also wanted a better free and open-source Garamond (and I will release my font under the OFL when it is complete). EB Garamond is a nice font, and based on the same model as Sabon (for the Roman face), but it is slightly too distressed for me, as I am more interested in a very clean design. I also have a strong belief we need more open source fonts which are appropriate for text.

George Tulloch

Is it just me or is the pdf corrupt?

Paul Miller

The PDF is corrupt.

Daniel Benjamin Miller

I will try again with the pdf, sorry. Hopefully this works for you. It works for me in Okular and the Google PDF viewer on Android as well. And while we're at it, here's the Peano sample as well.

Also, not reflected in this sample: I may want to make the /u a little bit shorter.

André G. Isaak

Acrobat reports the (original) .pdf is corrupt, but oddly Apple Preview and QuickLook both display it without problems.

Daniel Benjamin Miller

André G. Isaak said:

Acrobat reports the (original) .pdf is corrupt, but oddly Apple Preview and QuickLook both display it without problems.

Interesting. I wonder if it's because of something with the font itself, maybe because it is named Untitled1. The PDFs are normal, set in LuaTeX, except one which was XeTeX (alphabet-again). Anyway, if anyone has thoughts on the font, and not the PDF, I'm happy to hear them!

Hrant Հրանդ Փափազեան Papazian

My 2¢: if you'd like to encourage non-Latin extensions, go for a license more libre than OFL.

Daniel Benjamin Miller

Hrant H. Papazian said:

My 2¢: if you'd like to encourage non-Latin extensions, go for a license more libre than OFL.

I am myself working on a Greek extension. It will definitely be available under the OFL since many people want something compatible with the OFL. But I would also like to make it available under other licenses as well. I am considering at least GPL+FE/LPPL, and I could also go for X11, even.

I also want to do a math font, however, and this will likely only be available under the OFL, since I would use another OFL math font as my basis. But we're not near that yet.

Also, I've decided to name this font Salieri (nice name! not taken!) Source is here: https://github.com/dbenjaminmiller/salieri

Christian Thalmann

There are not that many libre Garamonds yet, and since EB Garamond regressed with its change of leadership, I'm happy to see another attempt.

If you're serious about making this a Garamond, though, I would suggest:

• A long-flagged /f/. Your current /f/ will do nicely as a contextual alternate to avoid collisions. Though for the true Garamond look, you'll need ligatures as the first choice for such situations.
• A narrower /a/. The /a/ is perhaps the most iconic letter of Garamond, and the one that led me down the rabbit hole of making a Garamond sans five years ago. This one doesn't feel like a Garamond /a/ to me.
• Better spacing. For instance, there is a rather big gap in Arith-meticae. I think the /i/ also has too much space around it.
• Another look at the rounds of /d/, /q/, etc. It's currently a very bumpy stroke that deforms the counter.
• I'd also like to see a long /Q/ in combinations like «Qua», but that might be a minority opinion...

Nice name indeed!

Daniel Benjamin Miller

Christian Thalmann said:

There are not that many libre Garamonds yet, and since EB Garamond regressed with its change of leadership, I'm happy to see another attempt.

If you're serious about making this a Garamond, though, I would suggest:

• A long-flagged /f/. Your current /f/ will do nicely as a contextual alternate to avoid collisions. Though for the true Garamond look, you'll need ligatures as the first choice for such situations.
• A narrower /a/. The /a/ is perhaps the most iconic letter of Garamond, and the one that led me down the rabbit hole of making a Garamond sans five years ago. This one doesn't feel like a Garamond /a/ to me.
• Better spacing. For instance, there is a rather big gap in Arith-meticae. I think the /i/ also has too much space around it.
• Another look at the rounds of /d/, /q/, etc. It's currently a very bumpy stroke that deforms the counter.
• I'd also like to see a long /Q/ in combinations like «Qua», but that might be a minority opinion...

Nice name indeed!

• The /f is reflective of the /f in Sabon. But I will include full ligature sets for f, including with characters like /b and /k. I just haven't gotten around to drawing that yet. I may include a longer-flagged /f as an alternate. Remember, I said I didn't want a fully historical Garamond.
  
• Never thought of a narrow /a as iconic. I will look into it.
• The spacing is provisional.
• I've redone them, per a comment not received on here. Here are the new versions.
• Maybe later as an alternate, just not yet.

Also, tangential, but why do you feel EB Garamond regressed? I never felt the project got worse under Pardo, and I rather appreciated the new weights. But the core problem is that it focuses more on a historic distressed look, rather than cleanliness. I far prefer a clean font. And I've tried to do that here!

Hrant Հրանդ Փափազեան Papazian

Daniel Benjamin Miller said:
The /f is reflective of the /f in Sabon.

OK but please don't do its Italic "f"...  :-&

Christian Thalmann

Sabon doesn't strike me as very Garamond-y. BTW, it looks like Sabon Next's /f/ is significantly wider than the original... I guess they admitted their mistake. (EDIT: Apparently there were two forms of Sabon, one for machine composition and one for hand composition, with the latter suffering from fewer design constraints and thus sticking closer to the Garamond genome. I guess Next preferentially uses those forms.)

Those new bowl shapes are much better!

As for EB Garamond: As I said, Pardo's version offers much fewer ligatures and contextual alternates. That kills the typeface as far as I'm concerned. Ligatures and alts are the life of a Garamond. I agree that Pardo's shapes are generally cleaner, which is a good thing.

Daniel Benjamin Miller

Christian Thalmann said:

Sabon doesn't strike me as very Garamond-y. BTW, it looks like Sabon Next's /f/ is significantly wider than the original... I guess they admitted their mistake.

Those new bowl shapes are much better!

As for EB Garamond: As I said, Pardo's version offers much fewer ligatures and contextual alternates. That kills the typeface as far as I'm concerned. Ligatures and alts are the life of a Garamond. I agree that Pardo's shapes are generally cleaner, which is a good thing.

Sabon is as true a Garamond as most Garamonds. Sabon Next is a liberal reinterpretation; the reason the /f is longer in it is because Sabon Next not was, unlike the original, designed to work with all major typesetting technologies of the 50s/60s (monotype, linotype machines, etc.). This devotion to compatibility is why Sabon's italic /f (which @Hrant H. Papazian mentioned) looks the way it does. Speaking of which, OpenType stylistic sets may be someone's friend here … I have open decisions to make on the italic still.

Hrant Հրանդ Փափազեան Papazian

Christian Thalmann said:

Apparently there were two forms of Sabon, one for machine composition and one for hand composition

As @Daniel Benjamin Miller says, wasn't the whole point of Sabon to set the same on Monotype, Linotype and in coldmetal?

Daniel Benjamin Miller

My present draft for the regular, in a sample: font source on GH.