Ascender & descenders in Latin type design

Ray Larabie

The technique of determining ascenders & descenders has been discussed recently in font critiques. I think it deserves it's own discussion.

@Hrant H. Papazian mentioned his technique here.

Here's how I determine my ascender/descender levels.

I work from the heaviest weight and design my f. I decide if I want to keep the f crossbar aligned with the x-height. I need to decide where it's more important to keep the ascender low or to keep the top of the f from looking crushed. Then I use the f to determine the height of the lowercase L and i. The ascender height doesn't have to align perfectly with the f , it just has to flow nicely. You can come up with all the theories you want but that fat f has plans of its own.

Next, I work out the g and y. The relationship between these letters is crucial so I design them simultaneously. I go back and forth, making adjustments. When I'm testing, I'm seeing how it looks in words with ascenders. Making sure it feels balanced. Or unbalanced if that's what I'm going for. Then I based the other descenders on what looks best with the g and y. I'm not too concerned about using up valuable descender space because the descenders usually aren't the lowest glyphs. The comma accent and lower circumflex almost always lower than the ascender unless I'm deliberately crushing them. In the lightest weight, there's less pressure on the top of the f and the x-height is usually different. I re-evaluate the ascender and descenders try to maintain the balance established in the heavy weights.

I'd like to hear how other people determine their ascender/descenders.

Craig Eliason

Thanks for posting this. Your method is a bit surprising to me in two ways. First your approach starts from the micro level out instead of the macro level in (i.e., shapes dictating the space used instead of vice versa). So /l/ is first related to the space demanded by bold /f/ rather than to the mass of weight in the x-height of more text set in multiple lines. And second, it's the outlier letters (/f/g/y/) that set the terms. I'm more likely to force my /g/ into the depth I want rather than letting it steer the other descenders. (Of course every sound method will include some give and take.)

It might be useful to fork this discussion into extenders for display types and extenders for text types.

John Hudson

For traditional text faces, I use a simple rule: x-height + descender = cap height. Ascender height is basically stylistic, i.e. it can vary pretty freely depending on the style of type and the effect one wants to achieve in text. But the relationship of x-height and descender height to cap height is foundational. Interestingly, I didn't start from this idea, but realised it after noting that my inverted exclamation mark always fit perfectly into the x-height and ascender: I was doing it intuitively.

PabloImpallari

John Hudson said:

...inverted exclamation mark...

Opening exclamation mark

Nick Shinn

For news types, always extremely short descenders.

Other than that: on an ad hoc basis, whatever looks right for the face in question.

I do try and consider the Extra Bold (i.e. boldest weight) before the face is too set in stone, so that it can fit the same bounding box as the rest of the weights.

Kent Lew

John — That is an interesting observation, which I’d never made before. But I just went and checked a handful of my own faces in progress and noted that I too seem to be intuitively falling in a similar range.

I don’t know that I would go so far as to codify the formula for myself as you have, but I will now be more likely to consider earlier on how my xht+descender relates to capHt.

Call me old-fashioned, but my initial core letters for designing are H h i l n o p and I just work out my key vertical proportions intuitively at that point, keeping in mind considerations like accents and room for heavier weights.

More and more frequently now, I find myself sketching at least a circumflex and cedilla to factor the accent space more intentionally earlier in the design process.