Short f descenders: Where else and why?

R.'s picture

I recently had a closer look at Mark Jamra’s Expo Serif (which appears to be quite lovely, by the way—has anybody licensed it and is willing to share experiences?) and found myself bewildered at the following sight in the preview PDF:

That’s quite short an ‘f’ descender, isn’t it? The ‘g’ is not as short as that, but also looks distinctly on the short side. The descenders of italic lowercase letters in every old style typeface I inspected, by contrast, all had the same optical length. The same seems to be true of every sans I checked—with the exception of Expo Sans:

Do you know any similar examples? How do you feel about visible differences in descender length? I admittedly did not even notice this peculiarity when I looked at the Expo family for the first time a couple of months ago. Now it makes me shrink back from licensing. Thoughts on why the ‘f’ is so short are also welcome although, of course, only the designer can elucidate how this came about.

Thanks in advance for any helpful contributions!

R.'s picture

So I can safely assume that nobody knows and/or cares?

russellm's picture


care deeply

R.'s picture

Are you aware of other typefaces with this ‘feature’? If so, which?

eliason's picture

Tisa, Meta Serif. I think there's plenty of these.
If Expo Serif's /f/ looks unbalanced, it might be even more that the descender struggles to balance a quite generous hood at the top of the letter, rather than that it doesn't come down as far as /p/ etc.

R.'s picture

Thank you very much for the pointer! I did check several popular old style typefaces, including both older and more recent releases, without finding a short ‘f’:

From top: Arno, Cambria, Constantia (all 2007), Adobe Garamond (1989), Minion (1990), Palatino (1948), Scala (1999), Times New (1932), Warnock (2000).

Is the short ‘f’ descender a historical revival or an innovation of this century? Meta Serif, Tisa and Expo all are less than five years old. I was wondering whether the short descender serves any practical purpose I fail to see or whether it is rather a quirk that typeface designers might introduce depending on their taste.

Bert Vanderveen's picture

Bram de Does’ Lexicon family (The Enschedé Type Foundry) has differing ascenders/descenders as a feature. If I remember the sales blurb correctly the version with the short d/a’s is specifically intended for newspapers and lexicons etc., — publications where real estate is at a premium.

R.'s picture

Thanks, Bert—but isn’t this a different kettle of fish?

Offering different ascender and descender lengths appears totally reasonable to me, both on practical and aesthetic accounts. But why would you make the descender of just one letter shorter than the others? And if you decide on doing so, why the ‘f’? As far as I can see it doesn’t look nice(r) and it doesn’t solve any problems, which is why I reckon it to be a stylistic variant. Please put me right, if necessary.

quadibloc's picture

But why would you make the descender of just one letter shorter than the others? And if you decide on doing so, why the ‘f’?

I would have thought the reason for that is obvious.

The lowercase f normally has a descender in italics. It normally does not have a descender in the Roman.

Hence, if one is designing a typeface that is to have a slightly cursive feel to it, one might give the f a partial descender in the Roman.

On the other hand, if one wants to "modernize" the italic of a typeface, make it less cursive, but more like a sloped italic, then one might shorten the descender of the f without eliminating it entirely.

I don't think this is particularly common, but it's something that might occur to people when trying to design a typeface that is "just a little bit cursive". It might indeed be a horrible mistake when they do it, just as you say.

R.'s picture

That’s an important point I overlooked. Thanks, quadibloc! Of course ‘f’ is one of the very few letters that grow a descender in italics (well, not always).

Interestingly I find the shortened ‘f’ to work quite well in Tisa, especially when there are no letters with a longer descender immediately adjacent (what I do find distracting is this ligature). The absence of a sweep to the left seems to make the unusual length more tolerable. In most other typefaces you mentioned the discrepancy feels unbalanced to me or at least as if they forgot to apply optical correction.

JamesT's picture

I figured the shorter /f/ in the italics was to help differentiate it from the decender in the /j/.

kentlew's picture

> I was wondering [ . . . ] whether it is rather a quirk that typeface designers might introduce depending on their taste.

There’s your answer.

R.'s picture

Might be, JamesT. I recently read a promotion article about a typeface designed for wayfinding signage in which the differences between letters with a similar skeleton were deliberately overemphasised to avoid confusion. Maybe some do that too in text typefaces to make them look more lively, individual, whatever.

Thanks to you, too, kentlew! Probably that’s really all that is behind it.

hrant's picture

Some thoughts:

Appearance of equal length is often good to maintain, but there's
also the matter of meeting user expectations of lines not touching.

Varying extender length is a fabulous unexplored avenue in text face design.

The best Italic "f" (in a text face) is one that does not descend at all.


R.'s picture

Thank you, hrant! Why do you prefer a descenderless ‘f’ to one with a descender?

R.'s picture

I see, thanks. So for you, the best italic is actually an oblique that sticks to the Roman letterform, but has a sufficient slant to indicate its ‘italicity’?

hrant's picture

(Hey, you're good. :-)

Yes, although mechanically-oblique Italics doesn't
at all cut it; one needs much more sensitivity. For
example I think Ernestine has a Really Good Italic:


Syndicate content Syndicate content