Flair Headline: Wide Didone

Mark Davis

Hello, TypeDrawers! Looking on feedback on a new typeface I’m working on; it’s based on my current logotype.
It’s a wide-set display didone. My design uses hairline serifs, and then flared shapes that flow from the hairline to construct letterforms that usually are defined by serifs, where a hairline serif would be too ill-defined. I’ve almost completed the uppercase, but am looking for critiques from this point. After completing this font, I’m going to draw a new master for a smaller print size and then interpolate for intermediate optical sizes.

Here is my newly modified logotype that I am expanding into a typeface, I added a curl to the /R/ leg, as well as tracked out my first name.

My original logo is at the top of my website.


Does this hairline/flare construction make letters that are too disparate from each other?
I’m looking for solutions for /CEF/ using this language, and my /X/ is struggling.

Also what do you think of the name MM Flair for the family name? (I’m putting MM at the beginning of my typeface names for consistency in branding for my “foundry.”) Looking for suggestions there too. I know there are typefaces that have Flair as a name, but it seems like they are older faces or free typefaces that are poorly made.

PDF specimen attached, no kerning yet.

Thank you for your time!

Scott-Martin Kosofsky

It looks like railroad lettering, those exceedingly extended forms that were intended to tighten up when when you saw the train speed by at close range.

Mark Davis

Thanks for your input Scott! I didn’t know that that was my influence, but I think it was!

Here’s a proof of some edits I made to MM Flair, I think it’s much improved! Shout outs to Alex Kaczun and Eimantas Paškonis for giving me email critiques.

I love critiques, so let them rip!

PabloImpallari

Hi Mark,
/E /F /L /T /Z serif treatment is very different from /C /G /S, etc...
If you look at the word "SESSIONS" the /E seems out of place
But if you look at the word "ITEMS" instead, The /E seems ok, and now is the /S the one that looks out of place. It's like you have 2 different concepts going on. You will have to decide which one you like the most.

Mark Davis

Hello, Pablo

I’ve been trying to get the two concepts to play nice together. What I’d like to happen is to define all serifs by hairlines only and no brackets.

The construction of the /S with hairline serifs works because the main stroke can define the letter almost with or without terminals. However applying the same style to other letters where the main stroke can’t carry the character, I don’t think it works as well.



Which direction works better?

Jasper de Waard

Both work. It's a matter of taste. What do you want it to express?

LeMo aka PatternMan aka Frank E Blokland

Jasper: ‘Both work. It's a matter of taste.’



What do you mean with ‘both work’? That you still can read it at the present size or at all possible sizes? When I reduce the image, then the version with weighted serifs (I have added some weight to the serifs of the N too) seem to work better. One could state that this display font is not designed for this or even smaller sizes, but there is no physical restriction in digital fonts, so this could happen, of course.



I don’t think this is a matter of taste either. Serifs are structural elements which:
– emphasize the ending of a stroke,
– represent the contrast,
– are an indication of the contrast-sort and contrast-flow.


In this design the weighted serifs are needed, because otherwise there is no contrast at the end of the thin strokes. I made a quick sketch to indicate that the reduction of the weight in the thin strokes ‘automatically’ results in an increasing weight of the serifs to keep things in balance.

LeMo aka PatternMan aka Frank E Blokland

Another issue is the spacing of the original example. In this the capitals are irregularly and tightly spaced as if they are fitted for a (non-existing) lower case. IMHO normally one should space (design) the capitals on a rhythm (stem-interval) based on the H, but these letters are so wide that this is hardly possible, even if the serifs were made larger.



More space between the letters shows even more the necessity of contrast at the end of the thin strokes:

Mark Davis

Good point James, designing type can be very subjective. Thus far, I’m worried that my subjective ideas aren't educated enough yet to be good ones. Most of my designs tweak existing type paradigms to create new ideas, but those tweaks end up being weak parts…like removing serifs from a Didot…

Frank E. Blokland—I’m grateful for your critique because I’m working up the courage to apply to KaBK, so having your thoughtful input (and sketches??) here is incredibly massively encouraging to me. Please note your thoughts have not fallen on deaf ears. I've been abroad in the UK and left my computer in the US, so I'll post changes next week.

Mark Davis

Hello, guys.

I’ve been reworking MM Flair a lot…I hope you can tell, compared to my original designs.
Looking for feedback, if you have any.

Martin Silvertant

I agree it's not a matter of taste but of application, however it's a bit odd to scale the typeface down to a size that is inappropriate for its hairlines and then argue how the typeface needs the thick terminals. No, it doesn't. The size range is restricted by the thickness of the hairlines and thus in its current size, both versions indeed do work as long as the use of terminals is consistent (so also in N and the left food of M).

One could state that this display font is not designed for this or even smaller sizes, but there is no physical restriction in digital fonts, so this could happen, of course.

Of course it could happen, but do you feel a type designer needs to take this into account? There may not be a physical restriction, but there are certainly practical and aesthetic restrictions. If a display typeface with hairlines is used at such a small size, that's bad practice from the designer. The type designer might make an optical range with thicker hairlines to extend the range of application size, but as the typeface is now I think it's obvious that it shouldn't be used at the size you're demonstrating.

More space between the letters shows even more the necessity of contrast at the end of the thin strokes

But why add more space to a display typeface like this?

Mark Davis

Hey, Martin. Thanks for the thoughts.

I decided to add full serifs to streamline the concept of the design and improve legibility at more sizes.

As for the added space, Blokland was suggesting to do so to improve legibility of the uc in the absence of lc. I tried this and Tobias Frere-Jones took a look. He recommended that I design it with very tight spacing, as that was the most challenging part of designing the face—if someone wanted to track it out for legibility they could do so themselves, but if they wanted it to be tight, it would be ready.

Do you have feedback for my newest revisions?

Martin Silvertant

The full serifs do offer a more stable design so you could go a bit smaller in pt size, but you're still restricted by the thickness of the hairlines. As for spacing, my comment was in regard to the notion that display typefaces like this should be spaced tightly, so it may not be necessary to design it as if it's a text face if it's going to sacrifice the design.

I think Frank's comments might have given you the idea that in this case you ought to go for the full serifs, but it really was a matter of taste up to a point. First and foremost its application should be a consideration. The full serifs to provide a better color and texture though, so that also slightly increases how small you can go with this typeface, because even if you set the text to be too small for the hairlines to be visible, you can still see the serifs, so the letter shape is not lost as it would have been with the hairline serifs. But this is a display typeface with very thin hairlines, so you wouldn't set this at a small pt size anyway. The lighter the hairlines and the higher the tracking, the greater the inconsistency in color and texture, which is generally not what you want. So I think what you have now works well, but I would probably keep the old design with the hairline terminals as a display cut intended for the larger range of application.

I like the full terminals, though I would also add some weight to the left foot of /A and right top of /U. There also seems to be a kink in the top curve of /C.