Upper/lower case strokes relationship

Graphirus's picture

Hello everybody,

I am working on a new font, this time a serif one. The upper case characters are more or less defined, but I'm having trouble designing the lower case ones. I wish to know if there is such a thing as a rule of thumb for determining the widths of small case characters in relation to the stroke widths of caps. Any other info pertaining to the design of small case characters based on caps is very welcomed too!

Regards

hrant's picture

Is this derived from an existing font - maybe Times?

hhp

Graphirus's picture

Are you asking for the inspiration for this font?

hrant's picture

Sure. But mostly, if you used actual outlines from an existing font:
- Disclose that fact.
- Make sure the results don't leave your hard drive (unless it was a "libre" font).

hhp

Graphirus's picture

I see. The initial inspiration came from antique soviet lettering used on some books that are laying around my apartment. After that I have used any serif font as a source of ideas, from the shape of the serifs to the modulation of the O's and so on. As a matter of fact I have not even decided yet the kind of serifs I will be using in my final design, nor the final shape of some key glyphs like a, g, e, Q

And no, no actual outlines were used, just painful drawing from scratch, something that can be better seen in my lowercase g:

Dunno if there's such a g... lots of refining to do yet.

So, let's return to the initial question of the thread please (strokes widths)

PabloImpallari's picture

I wish to know if there is such a thing as a rule of thumb for determining the widths of small case characters in relation to the stroke widths of caps

Usually, the are just a bit thinner. For example, if the stem of the caps is 90 units, the stem of the lowercase will be something arround 82 units (more or less).

Lot of great tips here:
http://66.147.242.192/~operinan/2/2.3.1a/2.3.1.01.notes.htm

I you want to buy a book to learn more about this stuff:
Doyald Young's blue book: Fonts and Logos, is the probably the best one
http://www.doyaldyoung.com/FL01.html

Graphirus's picture

Thanks Pablo, that's what I noticed after measuring some random samples (the lowercase width is around 92% of the upper case's). Nevertheless, I also noticed that in some fonts this difference is minimal, specially in diagonal strokes and horizontals. I don't know, maybe I'm doing something wrong but sometimes, to my eye, after applying the theoretical widths to a lowercase character it looks a)darker or b) lighter.

I bought Karen Cheng's Designing Type book, I hope it will arrive soon!

William Berkson's picture

There are optical factors, but there are also design decisions to be made, so there isn't one solution. The caps are a different alphabet originally, and they are now used to signal a difference, such as the beginning of a sentence. Some old style serif caps were quite heavy in relation to the lower case, and now some sans have caps with low cap heights and very similar 'color' to the lower case.

Because the caps take up more space, to have the same 'color' (grey) as the lower case they need to be somewhat bolder. But whether you want the caps and lower case the same even grey is a question for you to decide. Do you want them to have more signaling function, or do you want them to blend in more? The answer will tell you how bold you want to make the stems.

hrant's picture

the design of small case characters based on caps

Uh... no. :-)
1) If this is for text, lc comes first, so if anything it should go the other way.
2) As William hints, you can't force the two cases to be too close* without ruining one, or both.

* Unless your plans are "experimental", which isn't a bad thing actually.

hhp

Nick Shinn's picture

An important distinction can be made between old style (Garalde) and modern (Didone).

In the old style genre, the broad nibbed pen informs terminals, producing a consistent wedge-shape. So majuscule /C looks like miniscule /c.

However, in the modern style, the pointed nib informs terminal shape, producing a variety of balls, hooks, triangles and hairlines. So that /C has a triangular top terminal, whereas /c has a ball.

You don’t have to slavishly follow convention, but this distinction provides a useful benchmark for conceptualizing and organizing your terminal strategy between the cases.

Graphirus's picture

Thanks for all the comments guys, very useful.

Sometimes I wish there was an easier way to generate/compare different variations of a same glyph. I wish there was a multiple layers function in FL or something to keep variations and activate/deactivate them for comparison. I see myself over and over modifying a character, generating the font, previewing in ID and assessing the changes... one, 5 times just to think that the ideal solution is between option 1 and the last one. Do you have any tips about assessing corrections you make to a glyph?

And Pablo, Simplepolator is not working for me, I get an "There's nothing to wrap?" error.

eliason's picture

I asked a similar question a few years ago: http://typophile.com/node/56755
Nowadays I use Glyphs app rather than FontLab, and that makes it super-easy to try out variations.

Graphirus's picture

I'm also using a trial version of Glyphs, but since I've got FL, and don't wanna spend more on new software, I wanted to know if there's an alternative.

eliason's picture

The first reply in the thread I linked should help then.

PabloImpallari's picture

Graphirus, do you have robofab installed?

Graphirus's picture

Yes Pablo, I do.

Syndicate content Syndicate content