Regis, a new serif font

AbrahamLee

Could be just me, but the combined upper serifs of /w really detract from the other letters. Too much blackness in one spot so it attracts my eye. I’d consider losing the center serif altogether or at least reducing the widths of each so there’s some white space between them.

Christian Thalmann

The weight and contrast look somewhat uneven to me. For instance, the /f/ is much heavier and lower in contrast than the following /g/; the /s/ and /u/ look lighter than /r/t/w/. The sharp edge in /j/ feels a bit out of character, and the /a/ is still quite inelegant IMHO.

Michel Kern

regarding lowercase /p /q I meant serifs at the bottom of the letter the descenders

Michel Kern

Hello @Daniel, I have a suggestion for the /i, more precisely for its dot and vertical spacing with the glyph body. I found that a circle for the dot is maybe "too cold" so I made a "rounded thick losange with a 90 degree rotation". Here is a sample:

Michel Kern

Also @Daniel, would you please explain how you performed your search to find if there was a digital version of the Monotype 178 "Barbour" design  ? Did you use a dedicated tool ?

Daniel Benjamin Miller

Michel Kern said:

Hello @Daniel, I have a suggestion for the /i, more precisely for its dot and vertical spacing with the glyph body. I found that a circle for the dot is maybe "too cold" so I made a "rounded thick losange with a 90 degree rotation". Here is a sample:

Sorry, but this design choice does not interest me.

Michel Kern said:

Also @Daniel, would you please explain how you performed your search to find if there was a digital version of the Monotype 178 "Barbour" design  ? Did you use a dedicated tool ?

Some Googling showed me that no such revival exists. There were no tools involved. See http://luc.devroye.org/fonts-97136.html, http://www.typophile.com/node/86027, http://johntranter.net/archives/987. (The last link mentions Corundum — https://www.dardenstudio.com/typefaces/corundum_text — as a revival of Barbou. However, this font did not offer the design I am interested in. Perhaps it is somehow based on Barbou—though I think it more likely 185 Fournier. It's not warm enough.)

Daniel Benjamin Miller

On this design: I still really love the original and want to revive it, but perhaps I will rework the entire thing and redraw everything in a more consistent way (considering that this was the first attempt I made at drawing a font, and that I've learned a bit since then).

I have some mind for drawing this as a variable font (especially since I have samples of the model typeface in multiple sizes), but I'm not sure if I'm really ready/able to do that, especially since that would probably involve METAFONT considering what's out there on *nix.

Linus Romer

METAFONT means probably much more work for a typeface like this. The uppercase letters are easy to do in METAFONT, but lowercase will be harder and italics will be super hard. I have invested much time in METAFONT for my typeface Elemaints and the uppercase letters were quite satisfying (after several years of attempts) but e.g. tear drops are quite difficult to get right for all instances of the font family. Even more difficult is keeping the number of parameters low. So I stopped development in METAFONT and used the outlines in FontForge.

FontForge can do multiple master and can interpolate between font faces (for the interpolation of the kerning tables I had to write my own python-fontforge scripts). But FontForge cannot generate variable fonts. However, the project https://github.com/adobe-fonts/adobe-variable-font-prototype shows how you could do it with AFDKO and ufo font files (FontForge can export to ufo).

Daniel Benjamin Miller

Thanks @Linus Romer — I've messed around with METAFONT before myself, and indeed it is a ton of work to get a decent looking glyph in many cases — look at Knuth's own paper on the /S. And that's besides the fact that it is annoying to convert a METAFONT into an outline font (although people such as you have written things which make it more tolerable). So while I have some desire to one day design a usable METAFONT, I don't have any hope that it will happen anytime soon.

Instead, I have done all my work in FontForge, which I find quite clunky but, I suppose, tolerable. Maybe commercial software is better conducive to drawing outlines.

Right now I am experimenting with the spiro toolset, although I have used only the traditional tools before (for example, for the OSF/Small caps for STEP/XITS). I'm not sure if I'll be able to use that in production work, though.

For this design, I like it enough that I'd rather take longer to get it right. Perhaps I'll post some new work soon, once it's in a shape to be posted.