Quentin, welcome on Typophile.

It seems you mean it :-)

Besides studying typefaces like Helvetica or Futura it might be worth exploring a Bembo or Garamond the same way, or even the engraved lettering of some old Roman inscription.

Wondering why I recommend those?
Observe e.g. the *skeleton* of the classical Roman R with your pencil, and then compare it to the skeleton of Helvetica’s or Futura’s R.
– Then you’ll probably see why.

Hey,
Thanks for your suggestion!

At first I thought I should only give a serious look at sans-serif typefaces because mine will most likely look like one of them; but I get your point and I can now see the benefit of also studying serif ones! Of course I'll be more interested in their skeleton (as you call it) than in their serifs and ornaments.

By the way, if anyone has essential resources for the absolute beginner, I'm interested!

You could also be interested in the works of Feliciano, Pacioli* and Dürer on the geometric construction of letters.

* http://new.myfonts.com/fonts/madtype/pacioli/

Riccardo, now please forgive me.

Quentin: FORGET about the „construction of letters“. F O R G E T T i t ! !

This idea is but a long-lasting misunderstanding, imposed by the constructivist’s mind upon a matter he doesn’t understand. All those pretty Pacioli- and Dürer-drawings of Latin capitals, getting their serifs tickeld by tiny little Bauhaus circles – are plain kitsch. Nice to look at, but silly.

A letter is not a construct. Let’s get to the bottom of it.

A letter is a written shape. It is a graph, and a graph in the true sense is the trail of a tool’s movement (a pen, a brush). The origin of all letters of all writing systems is the moving human hand. A letter by its very nature is the result of a hand’s physical movement, its characteristic rendering dependant of tool and material. It is not the result of any construction process.

One may compare (!) an A to a triangle and an O to a circle. But that is not to establish a kind of equation.

Why not enable comments directly on each post? That way you could get a discussion going.

Well spoken, Andreas :-)
I was referencing those studies as historical antecedents to what Quentin is trying to do, and any type designer do: trying to shape letters. Brushes, compasses, chisels, Bézier curves, all are tools which have dictated the shapes of letters, and still do.

Hey!
Frank, your suggestion is a great one! I'm new to Tumblr so I had to find a way to enable comments. That's now done, thanks to Disqus. Don't hesitate to drop your reactions!

Riccardo+Andreas: I'll probably look at Feliciano, Pacioli and Dürer's works (which I didn't know), but I'll try to focus on my first idea which is keeping close to my initial and natural draft and trying to fit it into what a major sans-serif font looks like.

For example my M's vertex didn't touch the baseline when I first drew it, but I noticed most of the sans-serif fonts have a baseline-touching vertex. So I sticked with my first draft but pushed the vertex a little further to make it look more "regular"...

I hope my approach makes sense!

I posted a comment! Here’s one in general: Century Gothic is IMO a rather poor “me-too” geometric inspired by Futura (particulary Twentieth Century) & Avant Garde. If you are going to compare sans-serifs, I would rather pick a humanist. That way you can compare differences in grotesques (Helvetica), geometrics (Futura) and humanists (x). It might also be wise to consider a more pure grotesque like Akzidenz Grotesk. Helvetica is more of a neo grotesque (not a real classification per se, but a humanistification of the grotesque genre).

The most typical humanist sans serif I can think of is Frutiger.

Hey Frank,
Just saw your comment! I'm not sure what "notan" means so I'll document myself before I can answer. Thanks for such an input!

I also noticed that Century Gothic is very close to Futura and was thinking about finding another third font to compare, I guess your Frutiger suggestion is the perfect match!

On the Helvetica/Akzidenz Grotesk thing; I think I'll stick with Helvetica as it's really well-known and widely appreciated. Studying it can only lead me to something more efficient/popular.

By the way, did you try Googling "helvetica" today?

… Helvetica as it's really well-known and widely appreciated

Helvetica is not widely appreciated. It’s widely used.
That is something else.

If you wish to learn lettering, forget Helvetica.

Andreas your statements are very reactionary. I have learned plenty from both Helvetica and the constructed alphabets mentioned above. Even it is as simple as learning from shortcomings. There is plenty to observe in Helvetica, if you bother to look, it does not matter that it does not please you. Look at Akzidenz too, because it will tell you something about Helvetica, some nuance that escapes you until you really chase it down. If you are serious about designing alphabets you will look at all type/lettering with an open mind and will be continually reevaluating designs (including why they don't appeal to you or why they are problematic), else you are just repeating tropes.

In short you will learn more by looking at Helvetica than by not looking at it, so long as it doesn't beguile you.

I'd look at the original drawings of Futura and Helvetica (or "adaptation of drawings", Nick Shinn, if you pick up this thread ;-)) and not at the digital files. The digital outlines are full of red herrings - I mean about the placement of points and handles etc. The intention of the designer is much clearer in the drawings, if not as measurable.

Quentin I look forward to seeing how this project develops online. I had the same idea, to chart a design from start to finish, including everything that I learned along the way but I did not begin collating the material in the beginning. In hindsight it would have made for a painful website as my progress learning was slow. I think I might still do the same but retrospectively, and in print. I am looking forward to how you flesh out the archetype shapes you have drawn. Very different conclusions can be drawn from the same skeleton.

For many of us it's not about pleasure. Helvetica is flawed. It only really works at very light weights set very large.

hhp

Wow, I didn't mean to have a pro/against Helvetica debate!

Andreas your statements are very reactionary.

Maybe they are. I’m sorry.

As I recall the beginning of his conversation was:
…started thinking about a simple sans-serif font

and I was only about to point at the fact that Helvetica is hardly the point to get started with when it comes to learning type.
In type history, Helvetica is rather a termination. It has not become a model, which would mean that some kind of development departs from it subsequently. And I do think, especially when advising beginners, we should make it quite clear that ”Helvetica“ is not the prototype of typefaces, although some inherited practice of computer vendors of late seem to silently suggest just that.

Of course, the study of everything can be instructive. No doubt. But those who ask for basics to be known should by no means be left with weak end-of-age-products like Helvetica to look at.

Quentin,
Buy the "Font & Logos" book by Doyald Young (The blue one).
I don't know it this is possible now, but maybe you can get an used one from Amazon.

In "Fonts & Logos" Doyald does a detailed study of letters (Serif, Sans & Script).
You will Love it, and learn a lot from it. Yo will start to see letters in an entire different way.

Of course, the other 2 are also recommended.
Buy all 3 of then if you can.

Yes, I've looked for it but it's really hard to find, especially for someone in France like me! I know it sometimes pops-up in eBay or Amazon but I never had the chance to catch one at a reasonable price...

Wow, I didn't mean to have a pro/against Helvetica debate!
No nor me! Let's not let the thread go this way.

Thanks for your reply Andreas. I agree that Helvetica is not prototypical but I'm not sure that it is the end either (after all Hrant's beloved Haas Unica followed :-)). I am a bit disturbed by this Francis Fukyama-esque ending of things!

Quentin: http://www.typotheque.com/articles/my_type_design_philosophy this article might be of interest.

1985, this article is really good. Very interesting to learn about one's philosophy and choices. And a little type history is always something I enjoy!

By the way, I just posted some (very) early hairline versions of the caps. I know there's still lots of things to improve, not happy at all with the widths for example, but don't hesitate to drop comments on the site...

Latest image looks pretty good.

You may find useful to inspect the caps by groups:
EFHIJLTU (Horizontal stems)
AVWXZ (Diagonal stems)
KMNY (Horizontal and Diagonal stems)
COQS (Curved Stems)
BDGPR (Horizontal and Curved Stems)

Just one little tip from Doyald's book

Quentin: You need to move on past the hairline weights as soon as possible.

PabloImpallari: Thanks for the suggestion, I will! By the way I ordered the blue book on Alibris (pretty cheap), still waiting for it to cross the Atlantic...

Frank: I will, but I have to work on the lowercase hairlines beforehand, right? My idea is to have correct skeleton versions for both upper and lower cases in order to work on "bolding" them all at the same time. Does this make sense? Or is there a better/recommended process?

I have to work on the lowercase hairlines beforehand, right?

This is up to you really - working from hairlines helps in a small way as you fix the archetypal shape in your head, however drawing this shape with two contours, not one line, is quite a different experience. I personally work from a central line for my sans serif drawings but a lot of the Typophilers don't believe in it :-) Also this is not the same as a hairline - imagine the central line for a bold font - the bowls won't touch the baseline as they will in your hairline drawing.

Yes, the inner and outer contour is the key. Another key is spacing. A standalone shape can have any form, but in a "rythmical" system like a typeface both sidebearings, weight and internal space must be considered.

There is no rhythm at the [text] typeface level.

hhp

There is rythm in the sense that the whites should be of equal optical "mass" internally and externally, but no picket fence rythm if that was what you implied.

That's not rhythm, because rhythm is pattern in a flow, and there is no flow; there is only pattern. The use of the term "rhythm" in [text] type is a romanticism stemming from formal music appreciation. A text font is more like the sounds of nature.

hhp

In text there is flow. Unless you usually stare at it instead of reading it ;-)

What I'm looking for is a consistent set of glyphs, including upper and lower cases. I could start adding weight to my current hairlines but I'd feel like missing a step...

I know my shapes will evolve when I add "boldness" and I'm really looking forward to it! Like Frank says, the inner and outer countour is the key. Still 26 letters to draw before that!

Take your time, Quentin. If you need to go through certain steps such as completing a hairline first just do it. Everyone has their own style of working. Sounds like you are prepared to abandon the shapes when required later down the line. In a way it is good for the purpose of your website to be very thorough in your process. However you should try and anticipate what is ahead!

There's no flow in the mechanics of reading - it's all saccades, even backwards one.

hhp

There's no flow in the mechanics of a stream of water - it's all Brownian motion.
So, what's "flow"?

Flow is a unidirectional, constant change. There is flow at
the level of lines of text, but not at the level of type.

hhp

Hey,
I haven't posted in a while but the project is still alive and feedback welcome!

I've started working on lower cases and fixed most of the caps. Still sticking with hairline versions... After a little bit of reflexion I decided to go for a double-storey a! Way harder to draw but much more interesting to me, this project is about learning after all.

Haven't received Doyald's book yet, so if there are some advices on how to work on lowercases, I'm all ears!

Drop comments on the site` if you feel like it, that'd be much appreciated.

From what I see in terms of the types of analyses you're
doing, you might also pick up Karen Cheng's "Designing Type".

hhp

Yup, that books looks exactly like what I'm doing and looking for.
Will have to order it!

That Doyald Young video you posted is wonderful.
Yesterday I felt a bit down about the creative industry - that video was quite the antidote.

The Frutiger book you have posted is also a great reference. I'm not so sure about the Kareng Cheng book though, sorry Hrant. My feeling about Designing Type is that it does little to explain all the measurements it provides - why is this stroke thinner, why is the u wider than the n etc.. I own this book and still look to it to see how different alphabets tackle various forms but I feel that it did not really help me to advance. Hrant, feel free to disagree and say why you think it is a useful book - after all, it's better that I get value out of a book I own! Reconcile me!

I've always thought it's Volume 2 in a two-volume set, where the first volume has yet to be written. :-)

hhp

Yes, that's not a bad summary.

(tracking)

Quick update to let you know I'm still working on it!

The caps are almost done, now. I just added weight to all of them and I'm in the process of fixing small issues... I also have the hairline versions of lower cases and digits!

Any feedback you'd like to share?
Thanks!

/G/ looks a bit too stretched to the right (or flat on left). The thing is it shouldn't follow the line of a round shape because it's open on the right.

Although /M/ is nice I'm not sure it it fits the family.

/S/ looks visually vertically uneven. They usually are, but with the style you're having here it should be more even to my taste.

Have you tried writing this word: EAFHBFGEPFR and see if it looks distracting.

@litera, thanks for the comments!
I think I agree about the /G/ shape, will have to work on it a bit...

About the S "unevenness": I'm not sure I get what you mean. Do you think the upper part should be wider?

I'll try that EAFHBFGEPFR word for sure!

On a sidenote, Frode Bo Helland commented on the Tumblr and suggested "Optical corrections, upstroke and downstroke widths, ink traps and joins, intersecting strokes and harmonizing both shapes and internal/external space". Not sure about the "upstroke and downstroke widths" thing, do sans-serif typefaces need that? What about ink traps, got tips?

Thanks!

ps: posting the image here too, so everybody knows where I am.

Even in a monoline sans serif you usually need some variations in upstroke and downstroke. We are apparently hard geared to expect diagonal downstrokes to be sightly heavier. I don't know. It just looks wrong :) The bowl of P and R look like stretched D's, while B's bowls are more square. B has it's own problems: where the two bowls meet you need some thinning of the strokes. The point where the curve starts also needs some polishing. It has a telltale kink in both contours.

A quick update to show my new S...

I've also started fixing the up-and-downstrokes, will post an image later!

Quentin, this might help with stroke widths etc:

http://www.kleindaten.de/kaech/

Frode, what do you think about variation in stroke width (up and down) on a capital U?