Italics workflow
Jani Paavola
Posts: 48
What is your workflow on italics?
I've noticed that you can get quite far by finishing a font and setting kerning, and then tilting/slanting glyphs + of course modifying the glyphs that needs different shape. But what about (negative) side bearings? What else needs to be considered?
I've noticed that you can get quite far by finishing a font and setting kerning, and then tilting/slanting glyphs + of course modifying the glyphs that needs different shape. But what about (negative) side bearings? What else needs to be considered?
Tagged:
0
Comments
-
I expect that to be rather dependent on the application you're using. In Glyphs, I like to use «Cursify» with respect to half the x-height, which makes the glyphs stay in place rather than tilt over to the right side. I find that curved letters generally require manual re-tuning, though, especially if they're supposed to look monolinear.
0 -
oh that would be Fontlab. And I would slant by "center of selection", but this makes many of the glyphs have negative side bearings. Which doesn't matter, or does it?0
-
Don't slant by center of selection. You will get different relative horizontal shifts depending on the height of the glyph. For instance, a tall glyph will effectively be shifted more to the left than a short glyph.
Pick a point on the y-axis to slant from and use it for all glyphs. I like to slant from 1/2 of the x-height. Some designers like to split the difference between 1/2 the cap height and 1/2 the x-height.
The way to think about is this:- If you mix roman and italic, and you slant from the baseline, the italic will be shifted to the right relative to the roman. This will be especially noticeable with the caps.
- If you slant from the center of the x-height, the italic lowercase will be aligned with the roman lowercase, but the italic caps will be shifted somewhat to the right relative to the roman caps.
- If you slant from the center of the cap-height, the italic caps will be aligned with the roman caps, but the lowercase will be shifted somewhat to the left relative to the roman lowercase.
- If you split the difference, the italic lowercase will be shifted slightly to the left relative the roman lowercase and the italic caps will be shifted slightly to the right relative to the roman caps.
And, it's okay to have negative sidebearings with italics. Sidebearings are perpendicular to the baseline, so it's unavoidable.4 -
Thank you Mark! Excellent advice! Picking a y-axis point to slant is a next thing for me to try and study more. I'd appreciate if someone else would like to contribute their ideas as well.0
-
I usually also create an extra glyph which is nothing but a vertical line[similar to vertical bar] sloped to your italic angle and with very narrow centered sidebearings. I position it at the centerline of the xheight and use the measurement line (red) on so that the sidebearings are equal when this line is on. You can make another one for Cap center or smcp center or just shift the line up the those positions. It makes it easy to match sidebearings this way.1
-
I've done that, too, except I don't think you want to change it's position for characters of different heights. Otherwise you'll run into the same problem as slanting from the center of the selection I was talking about above.0
-
What's the difference between slanting from the middle and slanting from the bottom and adjusting sidebearings?0
-
There is no difference as long as you slant on the same height.0
-
The user and all related content has been deleted.2
-
What's the difference between slanting from the middle and slanting from the bottom and adjusting sidebearings?
Possibly less work, since, if you slant from the bottom, every glyph will need to be shifted to the left in order to look right when mixed with the roman style in text. If you slant from, say, the middle of the x-height, the slant and the shift happen in one step. I realize I'm not counting setting the anchor point for the slant as a step, but you have to set it or at least check it before you slant anyway.
0 -
Here's my workflow for obliques, not proper italics:
Often with obliques, I'll reduce the width before slanting to add extra emphasis. I find the rhythm changes when I squeeze & slant. I add sidebearings until the rhythm feels right.
To calculate the offset, I copy the H and the H to a new, blank font. Then I type HHHHH in the metrics window. I offset the H until is looks right and use the difference as my offset. It doesn't have to balance perfectly between the H's but the top of the H shouldn't collide with the following H. I tried using a formula and the results were similar but exaggerated. Even when I used a formula, I needed to check and adjust it visually so I don't waste my time with that anymore.
I feel like it should be a bit tighter on the right than the left... it feels correct and the lowercase italics end up exactly in the right spot. Since heavier weights usually have tighter sidebearings, the offset needs to be calculated for each weight/axis.
When I'm shifting italics to the left in FontLab, I use the adjust metrics command to reduce the left sidebearing and once more to increase the right sidebearing. That way it won't mess up my composites like the shift command does. I manually realign any 180° composites such as ¿¡>)}]ə. *
* Don't worry, they get decomposed before export.0 -
Thank you for all of your answers. Chris's extra glyph -tip is beyond my comprehension, I just don't get it...
Got more tips on italics (or obliques) FontLab workflow? These are puzzling me:
- Do you have to set reference point in every glyph one by one?
- Do you decompose the glyphs with components before slanting? Or set components on place manually after slanting the components?
0 -
Do you decompose the glyphs with components before slanting?
Nope. If the accents look like the need to manually offset, shift the original accents and all the component accents will shift the same amount. Don't use the shift command to move them. Go to each original accent, select all and nudge by tapping the arrow keys.Do you have to set reference point in every glyph one by one?
I've never used a reference points in FontLab...I never figured out a use for them that didn't waste time.
1 -
...extra glyph -tip...
I think I know what Chris was referring to. I've used the same (or similar) technique sometimes. It easier to show than to explain. Here's an image I posted on Typophile.com about ten years ago when there was a similar discussion:
Basically, you create a glyph that is a slanted rectangle. The angle should be the same as your italic angle. It should be full-width (e.g., if the set width of the glyph is 300, the rectangle should be 300 wide). It should also cross the side bearings on either side at the point on the y-axis at which your italics slant. The idea is that when you set this glyph on either side of the glyph you are working on, you will be able to see if the space around it is balanced.2
Categories
- All Categories
- 43 Introductions
- 3.7K Typeface Design
- 803 Font Technology
- 1K Technique and Theory
- 622 Type Business
- 444 Type Design Critiques
- 542 Type Design Software
- 30 Punchcutting
- 136 Lettering and Calligraphy
- 83 Technique and Theory
- 53 Lettering Critiques
- 485 Typography
- 303 History of Typography
- 114 Education
- 68 Resources
- 499 Announcements
- 80 Events
- 105 Job Postings
- 148 Type Releases
- 165 Miscellaneous News
- 270 About TypeDrawers
- 53 TypeDrawers Announcements
- 116 Suggestions and Bug Reports