The first version of Tiphares dates back to late 2018. At the time, I’d been in the PhD program at the Royal Danish Academy for about eight months and had already worked my way through a decent pile of research papers. My supervisor, Sofie Beier, made sure I understood the foundations of modern legibility research. I quickly became interested in the structural adjustments that sit at the core of type design. Inevitably, I wanted to translate all that new knowledge into a font. And since I was doing a PhD on legibility, it felt natural to look toward typefaces for wayfinding, user interfaces, and other contexts where text is often read at small visual angles.

The studies that shaped the project were already discussed in the previous post, but after finishing the PhD, I realized Tiphares had potential. Not just as a font, but as an example of legibility research applied to design.

Designing a semi-serif (ish)
Tiphares belongs to the semi-serif category, an underpopulated genre sitting right on the crossroads where borders and classification get blurry. Its serifs are carefully distributed and thoughtfully structured, but not exactly canonical if you’re comparing it to something like Otl Aicher’s Rotis. I went over the rationale behind these choices in the previous post. And since I’ve yet to come across a single study focused on uppercase letterforms, the caps in Tiphares are essentially sans-serif in spirit. In that sense, it’s a bit closer to another well-known semi-serif: Luc(as) de Groot’s TheMix. That said, Tiphares leans more heavily on rational contrast and ductus logic.

Strip away the serifs, and what’s left is a neo-grotesque: an adaptation of 20th-century grotesques for modern use. If you’ve got a sharp eye, you’ll spot the influence of late-90s German design: Ole Schäfer, Erik Spiekermann, and typefaces like Fago, Meta and Info. All of them with compact proportions, tense curves, squared counters, and generous apertures. They’ve always excelled in branding and packaging, but structurally they shine in information-rich environments—signage, interfaces, anywhere hierarchy matters.

Maybe it should be said that Tiphares is a neo semi-serif?

Width is a luxury
One question I get a lot: if legibility studies show that wide letterforms and generous spacing improve performance at a distance1Mangas, H., Biľak, P., Arista, R., & Beier, S. (2024). Enhancing distance reading for low vision: A reading acuity experiment on letter width. Information Design Journal, 29(3), 223-239., why aren’t more wide fonts for wayfinding or interfaces?

Because horizontal space is an expensive commodity. Our visual system gets lazy with long horizontal formats. It prefers frequent line breaks over having to do full-on neck gymnastics while reading. So the reality we’ve built around this limitation is vertical: we’re okay with taller layouts, but horizontal space is always at a premium.

That’s why Tiphares was drawn with proportions that are carefully studied. We’ve already talked about how serifs might help from a theoretical perspective—but they also add precious surface area to recurrent glyphs like a, n, i, or u. Narrow characters like f, t, and I have been made wider too, following research recommendations2Beier, S., & Larson, K. (2010). Design improvements for frequently misrecognized letters 1. Information Design Journal, 18(2), 118-137.. And yet, Tiphares still holds a compact, contained visual footprint.

On duplexing
We touched on Tiphares’ numerals in the previous post, but there’s one detail that deserves a bit more airtime: Tiphares is partially duplexed.

Now, “duplexed” is one of those terms with a wobbly definition, it shifts slightly depending on who you ask. At Pathfinders, when we say a font is duplexed, we mean that specific glyphs (or the whole font) share the same width across different weights in the family. Not to be confused with monospaced fonts, where every glyph has the same width in every weight. In a duplexed system, only some glyphs are harmonized across weights.

And there’s more: those widths often follow fixed ratios—1:1, 1:1.5, 1:2, and so on. The space character defines the base unit. Punctuation like the period and comma share that unit. From there, other characters scale up: tabular numerals and math symbols occupy two units (space × 2), the percent sign and fractions occupy three. 

Easy to explain, not so easy to pull off. All of this needs to look harmonious across an entire family. But if you get it right, the payoff is huge: laying out complex tables full of alphanumeric data becomes easy. You can make layout tweaks with simple space adjustments. No code hacks, no table styles, just clean, flexible typesetting.

Soft, Negative, and Grade
The Soft and Negative styles came later. They were part of a second wave of development, influenced by various readings, and especially by the work of David Berlow. Berlow isn’t just a generational type designer; he’s a wildly creative thinker and a true champion of taking type projects beyond the usual boundaries. I’ve had the pleasure of working with him over the past few years, and it’s been a real intellectual spark.

Together with Berlow and teams from Google Fonts and Glyphs, we’ve done a lot of work around variable fonts. And among the many design axes we explored, one of them is grade.

Grade is a typographic concept where the font appears heavier or lighter without changing glyph widths. In physical design, grade helps deal with inconsistencies across print processes. In digital environments, it addresses halation: the optical effect where white text on a black background looks heavier than black text on white.

With the tools and knowledge we’ve developed, implementing grade in Tiphares felt not just possible, but inevitable. The process? Surprisingly smooth.

The decision to design the Soft version came after a close study of FF Info. It was part of a broader effort to expand the font’s utility and communication range, without losing its backbone.

Is Tiphares a pinnacle of legibility?
Tiphares is a demonstration of how independent, contemporary legibility research can directly inform and improve type design. This is common practice in other fields, like furniture design, fashion, UI, or packaging, where independent research doesn’t replace internal development, but guides it.

Typography is getting there. The research is mature enough to be useful, but it still needs more qualified designers, people with the right experience and judgment, to bring it into practice. We also need a bigger, deeper body of literature to explore other aspects of type beyond just legibility.

So… is Tiphares a legible font?

It’d be fantastic if, after these five blog posts defending scientific rigour, I showed up and said, “Yep, Tiphares is extremely legible. Use it everywhere.” Its foundations are solid. The execution is sharp. But it hasn’t been tested in scientific studies—so there’s no data, for or against, to claim that it performs better than Fago, Helvetica, or Frutiger.

But that shouldn’t make it a suspicious font. I hope it makes it an honest one.

In recent years, we’ve seen a wave of fonts (Sans Forgetica, Atkinson Hyperlegible, Bionic Reading) proclaiming breakthroughs in legibility, with little to no scientific backing and, in some cases, questionable performance.

I firmly believe that every design decision I’ve outlined here has made small, incremental contributions to reading performance in low-visual angle and possibly glance reading. But the overall performance compared to other fonts? Still unknown.

What I can say is: I’ve stood on the shoulders of researchers, collaborators, and mentors to make something better than I could’ve built on my own. So now I’ll close the circle and hand the baton to you, dear sharp-eyed designer. You get to decide if what I’ve laid out here is compelling enough; if Tiphares is the right fit for your next project.

What does "Tiphares" mean? (And how do you pronounce it?)
I love comics: European, American, Japanese… Many of the fonts in the Pathfinders pipeline borrow names from that universe. Our studio, Ashler, is named after a villain in Mazinger Z, half man, half woman. Tiphares (pronounced Tee-Far-ess) comes from Gunnm: Battle Angel Alita by Yukito Kishiro. It’s the name of the suspended city (called Zalem in English) and a place I’ve always felt a deep, reverent awe for. Exactly what Kishiro probably intended.

When I started the project, I imagined what the signage of that city might look like. A futuristic, high-functioning city, without the sci-fi clichés.

For the record, Robert Rodriguez’s film adaptation is… well :(