Karole's just a working name

rcapeto's picture

Dear typophiles,

what follows is something I've done a few years ago
as a kind of warm-up exercise before a very different
job (which didn't turn out as expected BTW). I only
did then part of the lowercase set (the easier glyphs ;),
and five or six UC letters. Recently, I've come across
these files again and wondered if it would be worth the
trouble to revive and complete them.

What do you think?


karolehard.gif

There were three versions, which I'm calling hard,
soft and sharp. The hard version was the less
incomplete, so it's the only one I'm displaying here.
The soft version would naturally be the one more
appropriate for text (well, would it?). I compare
the three below.


karoletrio.gif


The soft version is of course the most difficult and
demanding. I had a vision of how it should be, but
haven't been competent enough to do it back then
(or rather I didn't invest the necessary time on it).
But I guess sharp was the sexiest baby to me. ;)



karolesharp.gif


General or specific comments are welcome.

Thanks.

eomine's picture

e a

eomine's picture

wondered if it would be worth the
trouble to revive and complete them.
What do you think?


sure!
i don

hrant's picture

Rodolfo, this is very nice - although I do have a weakness for semi-serifs. There's a bit of Unger, a bit of Granjon, and a bit of Times in there. But the result is totally original.

I won't go into specific glyphs (yet). What I'd point out is that I think you need to make some text versus display decisions here (as you're already prepared to do, it seems). For example, the Sharp and the Hard are very well suited to display work (at large and moderate sizes, respectively), so the narrowness is right on. I agree that the color and contrast of the Soft make it the most suitable for text (while maintaining that rigidity I for one find priceless), but there the degree of narrowness will act against readability too much.

The other thing I'd point out is that you're mostly avoiding the classic problem with semi-serifs: backslant. The delicacy of your serifs helps. But do consider putting slight swells* on the other sides of the semi-serifs, especially in the display cuts.

* But definitely not mushy ones like in Avance. Look at FF Profile, or maybe Proforma.

hhp

hrant's picture

BTW Rodolfo, sorry, yesterday I'd resolved to break up my lines
from now on, but I forgot... Next time. Well actually, this time!

hhp

rcapeto's picture

Eduardo,

- in your "k", seems like the lower arm is too much bigger
than the upper one;


Do you find it disturbing? That was a very conscious design decision.

- i think "t" doesn

hrant's picture

> But the drill work involved in making a complete face bores me big time.

Some font houses are willing to apply their in-house designers towards the completion of a promising font, although your cut of any profits would go down.

> You know I like Times, don't you?

I like it too, but only for display.

> Smeijers's theory

It doesn't make sense, because we read boumas, and that relies on tight spacing. Loose spacing is in fact the main weakness in Eureka and Rotis. Smeijers might be talking about very small sizes? There greater letterspacing is indeed more important, but so is large apparent size: if you save horizontal space through narrowness, you'll need to make it up through a larger point size.

The general view concerning narrowness is that it causes a drop in legibility through decreased decipherability of individual letters, but that's also misguided. The real problem is two-fold:
1) Bouma decipherability relies heavily on the overall width of the silhouette, and narrowness causes a decrease in the absolute difference of width in the lexicon of boumas. In the parafovea, the best way to tell apart "dearly" and "dastardly" for example is through their width primarily (and the second "d"'s ascender secondarily).
2) Too much narrowness also puts a given bouma too far outside familiarity.

BTW, do you think Patria is wide or narrow?
You might be surprised that it's more economical than Times... The trick is the massive counters. On the other hand, that has a certain bad effect in the subvisible realm.

> mostly with flat-foot half-serifs

Wouldn't serifs rotated counter-clockwise actually cause more of a problem?

hhp

rcapeto's picture

Some font houses are willing to apply their in-house designers towards
the completion of a promising font, although your cut of any profits would
go down.


In that case maybe that's what an assistant is for, right? Only, I'm
not prepared to pay an assistant for work that's not already backed
by a committed client. On the other hand, it seems that "carolingian"
half-serifs are rather uncommercial, aren't they? Has anyone ever seen
a real-world usage of Syndor? I have not.

> Smeijers's theory

It doesn't make sense, because we read boumas, and that relies on tight
spacing. Loose spacing is in fact the main weakness in Eureka and Rotis.
Smeijers might be talking about very small sizes? There greater
letterspacing is indeed more important, but so is large apparent size:
if you save horizontal space through narrowness, you'll need to make it
up through a larger point size.


The "theory" boils down to one caption (on page 139): "we tend to assume
that a typeface for a small body needs to be wider than one for a larger
body. Haultin doesn't follow this rule. Instead he justified small characters
rather widely, and this gives the eye the scanning time it needs for reading
text at small sizes."

BTW, do you think Patria is wide or narrow?

I'll have to check again.

> mostly with flat-foot half-serifs

Wouldn't serifs rotated counter-clockwise actually cause more of a problem?


No, I believe, because of our cultural frame of references and some
important differences in the way roman and italics work (I'm talking
about "conventional" serif faces here). Roman relies in large part
on the baseline for stability and balance. You can call that static
equilibrium. Italics, on the other hand, has a sort of "dynamic
equilibrium" (isn't this perhaps the most important difference between
them?), even though it's of course also sitting on a virtual baseline.
When you have a "flat-foot" (roman) half-serif it seems as if the glyph's
center of gravity isn't quite supported by the base and it tends to fall.
The same effect doesn't happen in a "carolingian" half-serif as the base
is not what supports it to begin with. Also the serifs angle provide a
sort of southwest-northeast thrust to the stems that helps to keep
them afloat.

BTW, this kind of investigation was one of my interests here.

hrant's picture

> I'm not prepared to pay an assistant for work that's not already backed by a committed client.

Which is why a [committed] font house is a great substitute.

> it seems that "carolingian" half-serifs are rather uncommercial, aren't they?

For text work, yes - there's too much lethargy.
But for display work it might be the other way around. I know a [relatively famous] French type designer who recently made a semi-serif even though he didn't want to, because there was high commercial demand for it.

Syndor is too affected. Plus it leans backwards.

BTW, I wouldn't call Karole carolingian - it's better than that.

>> "... this gives the eye the scanning time it needs for reading text at small sizes."

Which is completely contrary to how we really read... I guess Smiejers has the talent/instinct to make a good text face (Quadraat) without really understanding reading. The danger is that -as he progresses- he might find such faulty theories subduing his original good direction. Which is exactly what happened with Dwiggins: he thought extenders contributed nothing to readability (which he was apparently confusing with legibilty), so he never made anything better than his early Electra. "Things are at their best in their beginning" - Pascal.

> .... Roman relies in large part on the baseline for stability and balance ....

This is some interesting theorizing!
But to me it might exhibit a somewhat dangerous reliance on the display realm.

hhp

rcapeto's picture

BTW, I wouldn't call Karole carolingian - it's better than that.

I don't know if it's "better", but certainly it's not really a
"carolingian face". This was just a way of subclassifying
the larger field of half-serifs, based on how the serifs work.

This is some interesting theorizing! But to me it might
exhibit a somewhat dangerous reliance on the display realm.


Most certainly. In small text sizes things may be the opposite.
As we've talked about once, in small text sizes we're in a
quantum-mechanical world: all our macroscopic intuition may
break down.

ricardo's picture

Hello Rodolfo, very nice forms that you have on this samples. I like the idea of the same skaleton applyed on this differents applications (serif to calligraphy). The only thing that I would like to ask you is that your l/c (k) don't need a cursive terminal on the leg of the letter? like you use on the stroke of the same letter. What size you are thinking to bild (for text / display) for all version?. One more time very interesting your Karole work. See you soon. Regards for you and for Hrant.

eomine's picture

>in your "k", seems like the lower arm is too
>much bigger than the upper one;
Do you find it disturbing? That was a very
conscious design decision.


i looked at it again, and now i realize that
hard k seems very different (the arms'
proportions) from sharp k. i think sharp k
is ok. but hard k still bothers me: doesn't it
look too wide in "ikko tanaka"? what about
other "angled" lowercase (vwxyz)?


The name Karole had a reason: these are
basically carolingian->italic "serifs"


i think the name is fine, but actually i thought
it was related to some obscure writer from
eastern europe... :-)


hrant:
>it seems that "carolingian" half-serifs are
>rather uncommercial, aren't they?
For text work, yes - there's too much lethargy.


what kind of lethargy?

hrant's picture

> what kind of lethargy?

In this case, the kind that says that only the conventional serif structure works for text, and/or anything between a full-serif and a sans can never be serious. In reality though, serifs are not "appendages", they are very much structural elements (scary but true), so the presence/absense of a give serif (or half-serif) needs to derive from readability, not how dead people used the pen, or how your teacher taught you to draw pretty fonts.

In another case, the kind that says that the bowls of the "bdpq" must have the same structure - while the truth (as far as text faces are concerned) is exactly the opposite.

hhp

rcapeto's picture

The only thing that I would like to ask you is that your l/c (k)
don't need a cursive terminal on the leg of the letter? like you
use on the stroke of the same letter


Ricardo: well, that would be a different design. I like the contrast
when the shapes, instead of acknowledging the pen, suddenly seem
to acknowledge the brush (thus annoying Hrant even more ;).

i looked at it again, and now i realize that hard k seems very
different (the arms' proportions) from sharp k. i think sharp k
is ok. but hard k still bothers me: doesn't it look too wide in
"ikko tanaka"?


Maybe, Eduardo - isn't that an image artifact? I think it's more or
less the same. The "sharp" version is lighter, in any case.

It seems that this k is somewhat polemic. Just to give the due
credit: its design is partly derived from the famous gala portrait
of Louis XIV by Hyacinthe Rigaud. Look it up and you'll see how. ;)

eomine's picture

>needs to derive from readability, not how dead people
>used the pen, or how your teacher taught you to
>draw pretty fonts


thx for explaining hrant. i agree with you.


>isn't that an image artifact?

maybe. i'd like to see how "k" works in others
"contexts" (different words and letters combinations)...


>the famous gala portrait of Louis XIV by
>Hyacinthe Rigaud


interesting... i don't wanna be picky, but
the "k" i see there is very different from
yours. :-)

Syndicate content Syndicate content