Techniques & Strategies for creating Italic glyphs

Alex Kaczun

I would be interested to hear the techniques other type designers use in developing an italic font from the roman style.
Depending on the italic 'slant' or the degree of slope the shapes can get very distorted.
The greater the slant, the greater the distortion.
Many years back 'Ikarus' software had a great module for adjusting these type of problems on the fly. Playing around with the parameters, it did a fairly good job, as I remember.
It's pretty obvious visually what needs to be adjusted in order to make the overall shapes pleasing to the eye again.
I don't mind doing this manually, character by character. The round shapes require the most work.
I was just wondering what techniques others might be using to adjust these distortions.
One particular technique I like to employ is copying the regular roman glyph, in the mask layer, and rotating the glyph to the same degree of the italic slant. I find this a great visual aid. You can easily see where to adjust the curves and points.
Sometimes, depending on the shapes themselves, you can use the rotated glyph almost as is. Just a few visual adjustments and you are done.
Anyone else care to comment?
I find sharing technical aspects of type design very interesting. Something and is not discussed that often.
Looking forward to hearing from others on this topic.

Mark Simonson

I've used Briem's technique for a years and it works pretty well. Unfortunately, the site where the page about it is hosted is frame based, making it difficult to link to.

Here is a link to the page where the technique is described:

http://66.147.242.192/~operinan/2/2.3.4a/2.3.4.34.curves.htm

However, the exact address for this has changed in the past (maybe the IP address changed), so here are driving directions in case it changes again:

Go to http://briem.net
Click on "Notes on type design"
Click on "Italic"
Click on "Curves"

Alex Kaczun

Great link, Mark. Thanks.
Yes, I've used the 'Briem's' technique all my days in type design as well.
Although, I never knew the name. I learned this from a 35-year veteran letter drawer at Mergenthaler, John Quaranta. Good stuff... keep it coming guys.

Alex Kaczun

Oh, another good point to make.
Don't forget to condense (scale body widths), 85-95%, and on the top and bottom overshoots, increase by few points (to make glyphs taller). Because when you rotate (particularly) the round glyphs, you will not need to adjust the heights afterwards for the italics.
It goes without saying, but I'll say it anyway... Italics should always appear more narrow and lighter in stem weight and overall color (for contrast).
Also, spacing gets tighter on italics as well.

Georg Seifert

I have a "curify" function in Glyphs. It tries to compensate for the distortion from the slanting. I you try it, you should set up your stems in font info.

Göran Söderström

There is also RMX Slanter which works on some kind of curves, sometimes.

But I think no technique can be better than our eyes

Michael Clark

Maybe a stupid question, but, don't people "draw" italics any more... as a subset, based on the Roman, as opposed to relying on a "program" to do/skew it for you? I find the hand far more capable of the subtle nuances needed to alter/reconfigure a Roman form.
And I mean draw as opposed to calligraph. :-)

Alex Kaczun

It's not a question of hand drawing versus using a computer to manipulate letter forms and shapes. You can use many of the techniques mentioned here already to adjust shapes on paper as well. I still keep a sketch pad handy and do hand drawings when the mood strikes me. When I started in the NYC Mergenthaler Letter Drawing department in 1976, we only did hand drawings. Later on we sketched on velum and inked the shapes in by hand. Using french curves and a blade to clean up the shapes. Around 1979 we started to digitize our drawings and used Ikarus software to interpolate the weights in between.
I think it's fair to say, that today, most of us are extremely comfortable working directly on a computer manipulating points and shapes as we go.
I sometimes still find it easier to draw the letters by hand, scan in and or digitize by hand in order to get the glyph into FontLab. Whatever works.
Sometimes I even use my handy 'Wacom' tablet to draw the strokes directly into the program.
The topic here is what various techniques type designers employ to obtain enhanced shapes quickly and effectively.
For example, I often turn a drawing upside down, or even reverse the drawing on a light box in order to refine my shapes further. This is an old technique that works particularly well. Distortions and shape irregularities are much easier to see this way.
I'm sure others here can elaborate further. Other interesting examples of how they like to work.

Nick Shinn

I do start by slanting a few Roman glyphs.
But generally I build italics the same way I do the roman, Frankenstein mode.
That’s to say, once I have a few basic glyph parts, I assemble new glyphs from these; this usually requires numerous adjustments as I splice paths together, and I may do this by dragging points, paths and handles, selecting several points and transforming them, and making new sub-paths by drawing them with the various pen tools—whatever it takes.

As for assessing the shapes, I work in outline mode, and hit the ASCII grave/tilde button (in FLab) to see a solid shape. The vertical flip button in the Metrics window is another handy way to get an objective view.

I generally have a previous version of the glyph in the mask layer, for reference, and this I update as necessary.

I find the measurement line in the Metrics window invaluable for italics, especially for determining the sidebearings of round glyphs.

Michael Clark

"The topic here is what various techniques type designers employ to obtain enhanced shapes quickly and effectively."

Sorry to take you off topic... I was just wondering about the current usage of an old tried and trued method. I use the Frankenstein method also Nick but I have to know I am chopping and assembling the best "pieces-parts."

Jan Schmoeger

Alex Kaczun

Haha, Jan. Looks like my neighbors coming over for a visit.

Nick, I work in a very similar method. Okay, let's call it going forward the 'Frankenstein' method. Very funny.

Seriously, I find working techniques and working styles very interesting topic(s) for discussion. I knew a few individuals in type design over the years, I will not name names, that will only work behind closed doors (or curtains) so that others will not discover their secrets. Kind of silly really.
We all pick up tools or techniques that enhance our working styles. Every carpenter has his 'bag-of-tools (or tricks)' for getting the job done. I experiment constantly, and I can sometimes get some surprising things to happen visually on screen, quickly and efficiently. I was hoping that we all could share our different working methods. Italics can be particularly difficult so I focused the discussion on adjusting italic outlines, but we can talk about anything.
I still like to draw, but working directly on screen manipulating points, splines and curves is much faster. I like to come up with what I call the 'skeleton' for the particular typeface. The overall heights, body proportions, specific design characteristics and spacing for the font. As I work, I'm constantly envisioning how else I can change the design. My mind is always working. Sometimes, it's very difficult to finish a font because I'm off in another direction. But, that's the way I like to work. I'll try different types of serifs on my original 'skeleton' concept, or change a few specific design elements, and a new font emerges. An example would be my recent 'Axion' family of fonts and style variations. I'm still working on different alternate designs and it all started from a single font concept. What I find interesting, is that some of my alternates become more successful then the original concept design. I usually try to exhaust my ideas on a particular look before I move on to something else. But, usually, I just jump around from font-to-font going in 10 directions at the same time. Then, it's hard to go back and finish something that was uncompleted. You have to rethink the whole thing all over again. It can get very messy, especially if you are working on several weights. You scratch your head trying to figure out what was worked on and what still has to be done. But, that's my style!

Nick Shinn

…have to know I am chopping and assembling the best "pieces-parts."

Indeed; it was a great impediment to Dr. F that he had to work with such dubious materials—and that forms the dramatic basis of Shelley’s novel.

(However, this dramatic device was a clever fiction that mixed contemporary practice with that of an earlier time, when the number of bodies of executed criminals sufficed for dissection classes, and were delivered fresh, not dug up. By the time of “Frankenstein”, there were fewer executions and more medical schools, so the majority of corpses supplied were robbed from graves, and weren’t criminals.)

It’s not like we’re constrained to recycle only bits of Arial, Papyrus and Comic Sans.

Ray Larabie

Copy the upright to the mask layer, slant all, swap with mask layer. Use the slanted letters as a guide for drawing italics.

[Deleted User]

The user and all related content has been deleted.

James Puckett

but it never was the cure all that Ed claimed it would be.

There is no panacea to slanting type. I think the trick is to just design new characters for the italic so it’s a least enjoyable as opposed to the mechanical slant/rotate/tweak grind.

Mark Simonson

Benguiat's technique sounds more or less the same as Briem's.

Jackson Showalter-Cavanaugh

I don't believe there is an exact workflow you can apply. They all give you different results, suitable for different design styles. Briem's method is good for very round Os. Mechanical slant and adjust gives better results for square Os. I start by doing both and then finding something in between that looks right.

Also, I've given up on drawing Italic Os with only with nodes at the extreme. I've been getting better results having an extra set of points on the long curves.

Alex Kaczun

Hey, James P. Right you are. Nothing beats drawing. If I were still makings drawings on all my typefaces, like in the good-old-Linotype days, that would be the way to go.
But, realistically, most everyone today works directly on the computer, in some application. As such, we all need to start someplace. In order to get the heights, body proportions and spacing consistent—best way is to take the romans and slant/rotate until you get a desirable look. I think we all work pretty much that way. And besides, there is nothing "mechanical slant/rotate/tweak grind” as you call it—not many can draw a shape on paper perfectly every time. Everyone refines it on paper in a similar way that we adjust and manipulate on screen. The process doesn't really matter as much as the final result.
And as James M. pointed out it's not a 'cure all' that Ed Benguiat taught. No matter how good a technique, there is no substitute for the human eye. Even after I make a drawing and scan in, digitize or wrap curves around, I still have to fine tune the overall shapes. There is no way to create great shapes without some manipulation until satisfied.
Mark, you are right. All these variations on Briem's technique are the same. Thank you again for giving us the name and links to this method.
I actually use this overall technique for a slant up to 8 degrees (works great), but for 11 or 12 degree slant, I first adjust the extrema points and their handles a bit in order to get a more pleasing overall result. I do this now automatically, but it took years of experimentation to get this technique refined. Again, it depends on all kinds of factors. I always make adjustments depending the the complexity of the design. It goes without saying, that If you were trying to use this same technique on a fancy calligraphic script design, you could use it somewhat but it would require a lot of cleanup nevertheless.
Sometimes there are no substitutes to just making a drawing. And then there are these wonderfully talented individuals like calligrapher Denis Brown (http://www.quillskill.com/) whose work I much admire. No mechanical technique (or algorithm) will ever be able to create such beautiful typography without the human hand and spirit involved. I guess I shouldn't say 'never'—it's conceivable that someday someone will write a program to do just that.
But, until then, we will continue to labor and fine tune our work.

Alex Kaczun

Jackson, I do the same. Having an extra point on long curve segments does help in the overall shaping process. But, at the end, if I can eliminate some of these additional points and still maintain the shapes, I will. Thanks.

Max Phillips

Sometimes there are no substitutes to just making a drawing.

Amen.

I've been afraid to jump in and say so for fear of being branded a Luddite, but I really agree. If you're making an obliqued roman, as with most geometrics and grots, these mechanical routines work fine. (And I'm looking forward to trying Slanter on a geometric sans I'm working on.) But If you're doing a proper italic, a design in its own right, as with most garaldes, Italians, transitionals, and even modern faces, I think mechanical routines often produce mechanical results, or at best, forms that are correct and not-wrong instead of positively, affirmatively right. FWIW, I use a pencil, sometimes working on tracing vellum over an obliqued printout of the roman as a rough guide to proportion. Some people can create wonders direct on screen—Veronika Burian says she drew every bit of Maiola with a mouse—but for me to make curves with any kind of rhythm and life, I have to draw on paper. I also do this whenever a glyph's refusing to come to life onscreen; print it out, draw over it until it perks up, scan the drawing back in and redigitize. I find that the muscles of my hand and arm know things my head doesn't know.

Nick Shinn

Alex and Max, old media is a mechanical techique too.

By drawing I take it you mean drawing with a pencil/pen on paper.
That’s how I learned, but nowadays I draw with a stylus on a tablet, manipulating Bezier paths.
The motion of playing with a handle to bend a curve, or massaging a line into shape by grabbing it between two points and wiggling, or tweaking the angle of a glyph by the live slant tool—are these not all “making a drawing”?
Isn’t the work of type design drawing glyphs?

Alex Kaczun

Right you are, Max. Very insightful indeed. I thank you for discussing further.
Everyone has different preferences and skills. Although I think with time many can get comfortable manipulating strokes, splines and curves directly on screen.
Drawing on vellum is a great way to do it. It's transparent material, easy to sketch, erase and make correction, on either side. Using a small light box is a perfect combination. I work this way many times.
As an aside, Matthew Carter, at Linotype and Bitstream worked exactly in this manner. I have never meet anyone with better drawing ability and skills then him. Matthew would ink his drawing, towards the end, and used an exacto blade and french curves to slightly clean up edges and shapes. But, his drawings are incredible. He is a true master type designer. I think we all aspire to be like him.

Alex Kaczun

Yes Nick, I work with a tablet as well. I'm starting to prefer this method over drawing. Much quicker sometimes. I cannot believe I waited so long to get my hands on a tablet. Wonderful tool. Thanks.

Max Phillips

Alex and Max, old media is a mechanical techique too.

Well, there's mechanical and then there's mechanical. Any sort of tool-using activity can be called mechanical, but sketching letters with a pen or pencil's not quite as mechanical as using subroutines to perform preset mathematical transformations of existing bezier paths. I'm not calling mechanical techniques inferior—I hope that's clear. Often they're superior.

By drawing I take it you mean drawing with a pencil/pen on paper. […] The motion of playing with a handle to bend a curve, or massaging a line into shape by grabbing it between two points and wiggling, or tweaking the angle of a glyph by the live slant tool—are these not all “making a drawing”?

Sure. There's an infinite number of ways to make a drawing. I just used "drawing" as a shorthand term for old-fashioned graphite on paper. And it's a method that's much more organic than creating directly onscreen (which I also do). Drawing w/ BCPs is like pushing strands of al dente spaghetti around with chopsticks. You have to work hard to make it flow. With a pencil drawing, even a bad one, the gestural flow is there from the first mark.

Of course, if I had one of those nice Wacom tablets, I might use my pencil less. Anybody want to recommend a model?

Alex Kaczun

The one I bought is
http://www.wacom.com/en/products/pen-tablets/intuos
The medium size is adequate.
You will not be disappointed. Amazing tool.

Nick Shinn

I’ve been using a Wacom since 1989, never touch a mouse.
At the time, there was a lot of media concern over carpal tunnel syndrome, which was one reason I got one. Never had a problem with carpal tunnel.
I've had the Intuous 2 for several years now, and it’s quite big/sensitive enough for an iMac screen.
However, I had to get a different stylus than the one that came in the package, which is one of those stupid chubby rubbery thingies—I recommend the Classic stylus.

Max Phillips

Thanks, Alex and Nick.

Mark Simonson

I've never gotten totally comfortable with a tablet and stylus. I've always had one, and I keep trying to make it part of my habits, but I always go back to the mouse. I use it most often when I'm working in Photoshop. The mouse just feels more natural to me when working with Bézier curves. With one hand on the keyboard and the other on the mouse, I can work very efficiently. Also, sadly, I get a pain in my fingers if I use the Wacom stylus too long. Maybe the Classic stylus would be better?

James Puckett

I have an old Wacom Graphire I pull out for Photoshop work about once a year. I had a big intuos 3 for a few months but never could get use to the inaccuracy of the cursor—likely due to my coffee intake—when working with vectors. My input device of choice is a big optical trackball. It’s incredibly accurate and if my finger stops moving, so does the cursor. Very nice for just pointing at something and having the pointer sit.

Alex Kaczun

I have heard that if you buy one of the older Wacom 2 or 3 tablets you can buy an accessory lens cursor for precise point placement over your drawings for manual digitizing.
http://www.neuroscript.net/tablets/wacom_intuos3_lens_cursors.php
DTL IkarusMaster supports Wacom Intuos 2 and 3 tablets with lens cursors for manual digitizing. This way offering a unique and versatile alternative for scanning and auto tracing, especially when a 1:1 conversion is requested. In this case 1.000 IK units IK equal 1 centimeter, offering detailed control up to 1/100th of a millimeter.
http://www.youtube.com/watch?v=kcLYDKr6wSs
I still have this old type of digitizing puck tablet that I use for DTL IkarusMaster.
http://www.flickr.com/photos/nicksherman/6811821562/