I’ll go on record as saying I like it. Not much I don’t like about it.

What’re you gonna do with the other weights? I like it a lot.

Thanks. As for weights, well, I might do at least a lighter weight. I don’t know if I can use intrapolating... I would probably have to draw it more or less from scratch?

An italic would be nice too, although I have never drawn a serious italic to fit a roman before. I will give it a try. And of course caps.

Hi Jelmar

I can’t make up my mind if I like it the way it is or if the eyes/bowls are bothering me in some of the glyphs (especially e and o). The curve (little break) just before the angle shift inside the u (left side) I think is worth looking at again (it looks like it’s a mistake rather than intentional - either emphasize it or make a smoother curve). The problems with the v and w the way I see it is that you have shaved the corners. Apart from that they work fine the way they are.

I’m looking forward to see this finished though. It’s looking really nice which is a great starting point;o)

Hey Endre, thanks for your reply.

The eyes/bowls are more or less as I wanted them. Although the corners are the most emphazised in the ’e’ and ’o’. Maybe I can make those more subtle (although I actually like the ’o’ very much).
The little curve inside the ’u’ was intentional. It’s also in the ’n’ (obviously, because the one is an edited version of the other). I will check what I can do with it. When it was completely straight before the angle shift the letter looked too much like two vertical lines with a horizontal line attached to it.
With ’shaved’ corners you mean the little cut-offs I did on the endings on the ’v’ and ’w’? I wasn’t sure about that too, but the really sharp corners didn’t look right either, it looked too ’ridgid’, so to speak. But if you think it looks wrong, I will look into it. :)

I’m trying to add an updated PDF, but it is not really working.
It’s uploading, says it’s finished, it’s in the little list, listingbox is checked, but when I submit it’s not there. Any tips? :)

Yes just select the file and press submit without actually adding the file before submitting.

Hi Jelmar,

Looks nice. I love that ’k’, but the ’s’ looks a bit ’italic’ :-/ Oh, and I think those curves inside the ’u’, ’n’, ... could be a bit softer... By the way, I like a lot all those details you put in the glyphs. A small cut here, a curve there... Very nice job!

Robert: Thanks, that worked! New file is up (rough-cut2.pdf)

Anderson: Thanks. The ’s’ indeed looks a bit italic. It just sucks in my opinion. ;) I havent changed it yet in the new PDF, neither the ’u’ and ’n’. But I’ve added some first drafts of numbers and diacritics. I have never drawn numbers for a typeface before, let alone oldstyle figures. Lots of fun to do, but there are probably a lot of problems with them. I do agree with the ’u’ and ’n’, although I think the problem is more apparent in the ’u’. And about the details, I tried to add a lot of those small details, without them taking over completely. I think that worked out pretty well. :)

The lack of crits is astonishing! ;)
Anyway, for the people that might be interested, I’ve uploaded a new PDF (rough-cut3.pdf). It has some very first drafts of a couple of caps, and edited numbers.

Looks nice.

I like the ink traps.

The lower lobe of the B looks a little dodgy to me. The corner seems to strong.

Nice work. Some places to look at:

k : leg
R : leg
g : descender a little wobbly, too light?
y : the notch isn’t working

Mark: Thanks. I agree on the B part, it is very dodgy. I think I might just remove the corner.

test (whoever you are ;)): k leg and R leg, could you elaborate please? :) Because a lot of people are very positive about the legs of those characters.

’g’ descender, I think the negative space might get too small if I thicken it, but I will look at it. About the ’y’, I’m not sure about that whole glyph to start with. I might redraw it completely.

overall quite nice. just a few things that might just be personal preferences to me. you decide.

mainly i find the tails on some of the characters a bit distracting: a, d, u and a bit too much swoosh on the leg of the k. i think the tails work for i and l.
likewise, i find the analogous entry stroke on m, n, p distracting.
i like the kinks in your strokes, but they seem a bit wonky in places, particularly the eye of the e, in the u, the bottom bowl of the B, and the bottom portions of 3 and 5.
the negative space in the top hook of the f feels a bit tight. not sure what you can do to remedy that.
the r makes my stomach churn a bit.
i haven’t seen your spacing yet, but the ligs seem a bit wide.

Looks very nice. I’ve always wanted to combine these edgy and fluent shapes like you did:)
A few little remarks in addition to other comments (e.g. what I would do)

1) Where ’b’ and ’h’ the bowls connect the stem it seems just a little bit to thick for me. (Compare with ’n’)
2) I would make these point(s) a bit more smooth
3) Think this should be a bit heavier
4) Should be higher

It’s really nice.
I would like to see the lighter versions as well.

Hi Jelmar. What you’ve got here is a nice appealing lineale (Italian term for “sans serif”) that needs refinement and conceptual definition to make it more specific, and, by extension, more useful.

Conceptual definition:

* what kind of uses do you have in mind for the type? Be as specific as you can.

* will it be a display type or a text type, or capable of straddling both display and text work?

* are the curls (tails) functional or stylistic, or a combination of both?

Refinement:

It could benefit from more rigor / tension / stressing. This can be achieved by increasing the difference between inside and outside profiles. Results will be more apparent if you draw the characters at a lighter weight. What rigor it has now is hard to appreciate because the weight is fairly bold. It looks soft, lacks hardness—-surface tension.

If its main use will be a display type the inside kinks in the counters will be distracting, drawing attention to themselves; in that case I would make them smoother.

If its main use will be for text the inside kinks won’t stand out so much and it may be possible to retain them much as they are now.

One way to find out is to build a thinner weight suitable for text setting, complete the rest of the capitals, then make a PDF with sustained text setting—-around 300 words at 10 - 12 points size. Read it. Don’t look at the character design(s). Read it at the speed you normally would read. It reads okay? Then the kinks aren’t a problem, they can stay. It doesn’t read smoothly? It may be those kinks holding back the flow of readability.

* tighten up the branching points, where strokes branch from stems make them thinner.

* r needs more weight at end of stroke.

* curl on lower leg of k is too much, rein him back in, make it more subtle.

* consider replacing the diagonal roman style w with a cursive structure—-two u’s joined together. IMHO roman style w’s stand out in every font they are a part of, compromising integration of forms. Only a suggestion. If you really like w as is by all means keep it.

* keep up the battle for the s. Keep working on it. Walk away from it and work on something else for a while, then come back to the s and try again. Keep at it, work on that little mother and it will eventually submit to your will.

Otherwise it’s excellent.

j a m e s

I should add that increasing the rigor is optional, up to you if you want to do it, if you think it’s appropriate. My advice on rigor, contrast and stressing are based on the experience of having built several text-worthy sans serif types. Most people have never seen them because they are proprietry commission jobs.

To show you the kind of thing I mean:

The s is a bit overweight but that’s because it isn’t finished yet.

j a m e s

@ paul:
mainly i find the tails on some of the characters a bit distracting: a, d, u and a bit too much swoosh on the leg of the k. i think the tails work for i and l.
likewise, i find the analogous entry stroke on m, n, p distracting.

I agree that they may be a bit distracting, although mostly the entry strokes you mention. The tails to a lesser extend, maybe I could make them more subtle (e.g. shorter/lighter).

i like the kinks in your strokes, but they seem a bit wonky in places, particularly the eye of the e, in the u, the bottom bowl of the B, and the bottom portions of 3 and 5.

Good to hear some bottlenecks about the kinks. I’ve never made such kinks before, so any tips on that are welcome. I will try to improve at least the ones you mentioned.

the negative space in the top hook of the f feels a bit tight. not sure what you can do to remedy that.

I thought that too, the only thing I came up with at the moment is that the right side of the f tapers a bit, so that the stem is less thick at the top hook.

the r makes my stomach churn a bit.

Yeah, it’s pretty badly drawn! ;) I find the ’r’ difficult to get right.

@Jos:
Thanks for the visual aid, I will see if I can improve those points. I didn’t see it before, but the join of the ’k’ is probably too low indeed.

@satya (or whoever can enlighten me on this point),
I would like to do lighter weight(s) as well. This is so bold because I started this only to make certain words, but I extended it to a whole alphabet. Is there any relatively easy way to intrapolate this weight? Is there a trick for it? I understand that any intrapolated shapes need a lot of ’repairing’ afterwards.

@James:
* what kind of uses do you have in mind for the type? Be as specific as you can.

* will it be a display type or a text type, or capable of straddling both display and text work?

This was mainly started as exersize for myself. And I see it as a display type. Although I don’t know what will happen if I make a lighter/text weight, it may very well work for smaller sizes with some refinement. I think a lot of the problems will be more apparent when I make a lighter weight. It may be good to make this lighter weight, and then extrapolate it to get a new, better bolder weight.

* are the curls (tails) functional or stylistic, or a combination of both?

Purely stylistic. I just liked the look of it basically. ;)

About your ’Refinement’ points (too much to quote all now), they are noted and I will check ’em out. The battle for the ’s’ continues. Although making it less ’italic-looking’ might already improve it drastically.
I agree on the ’w’ part, they always seem to stand out a bit. I am starting to like it the way I have them (with some added improvements). A cursive-style ’w’ sounds nice. I will try that as well. With rigor you mean something like contrast?

I like the image you posted a lot. Where you have experience building text-sans serifs, I have none at all. :)

Okay, when you’ve developed it enough you’ll have a better impression of its potential.

Purely stylistic. I just liked the look of it basically

Cool :^) Curls are very trendy at the moment as a stylistic device, but there is a practical side to putting them on ’a d i l u’ in a sans serif. They function as semi-serifs, helping to make up for the absence of serifs, making sans serif fonts more readable.

The battle for the ’s’ continues. Although making it less ’italic-looking’ might already improve it drastically.

Same thing. The italic look will go away when you’ve mastered the s structure.

One way of determining where to place the join on k is to try integrating it with x. Overlay x on top of k and adjust k until they match, or as close as is practical to make them.

To make the lighter weight: run a bold filter with negative parameters on a and e. You’ll need to experiment with different values until it looks like the weight you want. Test render those letters at large and small sizes to make sure of the weight. Then run the whole font thru the bold filter with the negative params established by testing a and e. Finally, restore the stressing and lost detail. It means adjusting all the characters, but this method is still quicker than building the whole thing again from scratch.

j a m e s

Thanks for the reply. I was at the very moment you were posting experimenting with the negative Bold filter. It works quite well, and I like the results so far. I’ll put up an image later.

Preview. I haven’t changed anything after intrapolation, so most of it looks stupid. But I think it can work after adjustments.

For a better interpolation you need atleast two extreme weights. i.e one light and one Black. With only one weight you cant get other weights easily. I love interpolation but never used Bold command in Fontlab. Though it can be useful sometime.

Yeah, I understand two weights would be better. But I think I can manage making a lighter weight with the Bold filter. What I will do with any additional weights, I don’t know yet.

satya, what do you normally use then, to make other weights?

I was just going through some revisions, mainly the ones suggested here, when I thought I’d try another version of the ’g’. It’s a more ’regular g’
Which one would be better? I like the left (original one) better, because it has more character. However, the right one may be more balanced and may also be more recognizable/legible? The top-right corner might be too pointy?

p.s. It is a very quick edit.

Jelmar,
I really like the kinks. Don’t be afraid of them. They may be a bit too timid. Look at thje left of the “g”. Where the vertical curve meets the top horizontal curve on the inside.

Good gutsy work!

ChrisL

With too timid, you think I should actually make them more distinct, as in the ’g’?

Thanks for the reply. When I started this face, the kinks weren’t there. At one point I thought, ’let’s take large chunks out of those letters’. I wanted to do something like it for a while now.

Yes, more distinct, even more than the g,

ChrisL

...I thought, ’let’s take large chunks out of those letters’. I wanted to do something like it for a while now.

This is good. Anything different is good.

Chris, I love that you encourage designers to pursue their instincts like this, and give voice to their individuality. That’s what’s beautiful about people—-that we are all different and unique. Type design is beautiful too in that there are so many different types.

Like baseball, it doesn’t have to mean anything, but can be just beautiful to contemplate.

:^)

Jelmar, if it turns out to be a display type the kinks can certainly be retained with refinement and systemetization.

j a m e s

I’ll see how the kinks turn out. I’m going to try to incorporate the changes suggested so far that I like into the face first. Then do a intrapolation, and see how it looks. The test I did with a quick lighter version, shows that it actually might work for smaller sizes, if I loosen up the spacing.

What about italics? (not that I’m going to make them now, but I was just wondering). Draw from scratch, or slant characters and alter them to their italic shapes? Some of my characters already have a italic shape, like the ’g’ (and I also have a ’y’ with a g-like shape).

Making an italic. First off, to save yourself extra work later, perfect the regular upright font. Make sure the side bearings and any kerning are well-sorted—-test it thoroughly to be certain.

Then use the upright font as the basis for the italic. Apply a slant filter with a value between 5 and 10 degrees. Experiment with different slant angles until it looks right to you. Post the results and get feedback.

Once the ideal slant angle is established, start converting key characters to their cursive forms. How far your go with this depends on the final look you’re aiming for. Typically:

First narrow every glyph by about 15 to 25 %. Slanted uprights that haven’t been narrowed just don’t look right (like an italic font) unless the upright originals were condensed forms to begin with.

a > monocular a
b > curl ascender tip backwards slightly, remove baseline spur and smooth bottom left corner, like b in Minion italic
d > curl ascender tip backwards slightly, point baseline curl upwards
e > make it cursive, eg: Minion italic
f > drop stem below baseline and add back-facing curl
g > already done
h > curl ascender tip backwards slightly, add curl to right stroke, pointing upwards
i > add curl to bottom of stem, pointing upwards
k > curl ascender tip backwards slightly, make straight upper arm curved, or alternatively convert to a loop k
l > curl ascender tip backwards slightly, point baseline curl upwards
m, n > add curl to right stroke, pointing upwards
p > curl descender tip backwards slightly
q > replace with italic b rotated 180 degrees, curl descender tip forwards slightly
r > curl baseline stem tip backwards slightly
s > adjust curves and stressing for cursive flow; alternatively convert to cursive handwriting structure (see Bickham script and French script for example)
t > point baseline curl upwards
u > point baseline curl upwards
v, w > convert to cursive forms, study several good italic fonts, either seriffed or sans, for examples
x > study Minion italic x, convert, convert!
y > match to italic v form
z > study italic z in Minion and Charter for ’standard’ italic form; alternatively give it a cursive descender tail—-see French script)

c and o you may want to adjust the curves and stressing for a flowing cursive look. Severity of this depends how much angle you use. The gentler the angle, the less need to adjust. Same applies to curved parts of h, n, m, u, and the bowl letter pairs b-q, p-d.

Capitals: caps with curved elements will need adjusting too, but otherwise most italic fonts make do with the same capital structures as their matching uprights. If you want to add a bit of flair to the caps; replace E with a curved structure (examples: Liberty script, Phyllis italic, Piranesi italic); give V W X and Z the same treatment as their lowercase italic forms.

j a m e s

Jelmar,
Regarding italics. You might start out by slanting your upright regular and test the slant angle. Some people like different degrees and it can vary with the nature of the family. My current preference for a sans is 11 degrees but that is NOT a rule.
After you make your slant, copy the whole font to the mask layer and use it only as a reference. Then, draw your italics as they should look in the glyph layer. I personally find drawing italics to be much more work than I expected so don’t become frustrated if it takes a while. Italics are a different world and need care and feeding :-) Spacing them is another adventure.

ChrisL

”...encourage designers to pursue their instincts”

James, I hope designers would pursue their individuality and not be locked in by what has already been done. It can be very difficult to be on unchartered ground but you can learn more by treading beyond the 22 mile limit of nautical vision :-)

ChrisL

:^) Yep-yep-yep! Uncharted/unchartered ground is where I’ve always tried to be. It’s fun. There can be no art without taking risks.

j a m e s

Chris, is the advice I’ve given on making an italic okay? The last thing I want to do is give Jelmar bad advice. I documented the method I know. Yours sounds good too; using a slanted version of the regular glyphs as a mask layer for reference.

j a m e s

James,
There is nothing wrong with your advice. Most people would like very specific directions as you have given. I prefer to allow the person some time to play without specific guidelines at first. This allows them the freedom to explore without a model to measure against. After they get beyond this open and exciting time and are looking for fine tuning is when I would give more specific suggestions. This is just my way and my hope that people will not trap themselves in to convention before they take a good, healthy naive swing at it without preconceived notions to hamper them. There is always time to dial back afterwards if things have gone too out of hand.

ChrisL

Thanks for the replies. I think I will just go ahead with that ’naive swing’ Chris mentions, and then use James’ descriptions to fine tune the results afterwards.

I’ve been studying several italis of sans serifs to see how italic shapes work. I don’t like slanted romans, so I want to try to make a proper italic.
But I will first finish the roman font, with all it’s diacritics and such. But I was just curious about the italic.

Unfortunately the coming week will be quite busy with other work, but if or when I have something new to show, I will post it. :)

Chris,
Thanks for the feedback, you’ve confirmed what I was thinking. The changes I recommended are specific to a sans italic I made based on the cancellaresca italic model from 15th century Rome and Venice, so it’s fairly standard stuff.

But I agree Jelmar should explore without preconceptions first, and discover what happens with his own instincts. This is very good, I like the way it’s going.

Jelmar,
Take as much time as you need.

j a m e s

James,
Historic models are very valuable and should be studied. I would hope a student would first make an effort to do some designing without them at first so that they have a healthy respect for the historic and a chance to fly a bit first :-) Again, this is not dogma, just my opinion and probably minority opinion at that.

ChrisL

Not at all Chris—-I don’t think it’s dogmatic or a minority opinion. There’s at least three of us thinking this way.

j a m e s

for more tips on italics, see:
obliques how-to

Really nice work.

The ’i’ looks a little odd to me with the curling tail. Suit’s the ’l’ though.

Thanks.

Yeah, in my opinion the lowercase ’l’ needs something to differentiate it from the capital ’I’. Not so much if it’s a sans serif with slabs on the cap ’I’, but otherwise... So consequentially I put it on the ’i’ as well. Without the tail I think the ’i’ will look too harsh and square compared to the general feel of the face.
But thanks for your reply, all thoughts are welcome. :)

Check this out: Nomad

That’s nice. I had been searching for typefaces with kinks in it, but no luck.

I added a new PDF (rough-cut-4.pdf). A couple of pangrams, and a longer text setting. Changed all kinds of details and added characters. Crits and tips would be welcome.

I think I’m slowly winning the ’battle for the s’. I have shortened some of the curly end strokes, but I’m not sure I like them now. Need to draw them again. And I’m really looking forward to making an italic version, and lighter weights. But I want to finish this one first of course.

I don’t know if I have incorporated all the previous critique already, but all in due time. :)

both k’s are too narrow (K and k). So is E a little bit.
And I think your letter spacing is too tight to make it decent for text setting.

Thanks for your reply.

Really? The lowercase ’k’ is too small as well? I hope someone else can also give their opinion about that, before I change it. I agree that the ’K’ and ’E’ are too narrow though.

About the letterspacing, yes, it’s too tight. It’s a remnant of when I was designing this purely as a display face. At this point I’m planning to make a lighter weight as well, which should be more fit for text setting. Somewhere in the process of doing so, I will also loosen the spacing. I didn’t want to put time in to loosening it right now, but I might sometimes in the coming week. :)

Here is a (hopefully) helpful hint about the g:

You can (and should) cheat the bowl at the baseline a little higher. Instead of making it overshoot as far as the rest of the round glyphs, have the bowl just touch the baseline, or even a little above. It makes the letter balance better between the bowl counter and the space between the tail. Especially when you have short decenders. Right now it’s a little top heavy.

You’d think moving it up would make it stand out, but it’s the opposite.

R

Alright, I checked that out in a couple of sans serif fonts (like Fedra), and you seem to be right. The ’g’ (and ’b’, same shape) often have smaller overshoots than the ’n’ and ’o’ etc.

Good tip, didn’t know that. I will try it out. :)

Suprised to hear the b is shorter in some fonts. The reason I learned to shorten the g is because of an optical illusion caused by the tail and the space between the tail making the bowl look long. Wouldn’t be an issue for the b (i don’t see a problem with yours). It would depend on the font. Overshoot is not a matter of math, but of optical equality which changes at viewing size/distance.

Again, good work!

Yeah, I checked Fedra Sans’ ’b’ and ’g’, and they both have less overshoot than other roundish characters. I don’t know why he did that on the ’b’, really. Maybe because the shapes are virtually the same, and he wanted it to be equal.

Well, I have been changing the ’b’ and ’g’ yesterday (I don’t know if it’s for better or worse), which I will post in about 5 hours or so (or tomorrow). I can’t right now, because I’m at school.

And thanks for the compliment. :)

The squared bottom left corner of Fedra sans’ b means it needs less overshoot that Fedra sans’ d. The bowl of p in that font follows the outline of b. Neither needs much overshoot at the baseline because the bottom curve is closer to horizontal. Peter Bilak’s q works the same way as his b but upside down.

Rough-cut sans: The lowercase k does look a bit narrow, but that’s a subjective impression. I don’t think it’s a big deal as there are many highly-regarded fonts with narrow-ish k, some of them good text performers. Try widening it a small amount and see how it works in context. I overlaid your x on the k and the angles are quite close already. If x is not going to change, use it to set the arms on k. Then, to widen k lengthen the short bar connecting the arms to the stem.

A “rule” of thumb for determining capital widths: most Latin fonts either have proportional caps, or so-called “even width” caps, which are not really even width but work on the principal of even width. The best example that springs to mind is Giambattista Bodoni’s original roman type. Have a look at Bodoni’s caps and compare widths of H, N, E, U, D, B, K. Rough-cut’s H and N are the same width (or very close). Try making E and K the same width as H. They don’t have to be exactly the same width as H. Adjust them until everything looks right in context.Think of H as a guide.

S look too wide. a stands taller than o, c, e etc. Bowl of Y—-match to top half of X? The serif on C stands out a bit.

j a m e s

Your comments about Fedra Sans make sense. I’m also experimenting with the the square-corner ’b’ shape for my typeface, because it (and then also ’g’ and ’q’), don’t seem to look right as they are now.

With the caps I was mostly aiming at proportional caps, but that might not look right. I liked the large, almost-round capital ’O’, so that sort off forced me to go for proportional with the rest of the chars. I set the angles on the ’k’ and ’x’ and such more or less by eye at this point, I was going to do a check later, but thanks for your tip about using the ’x’ for the ’k’. I don’t know if the ’x’ will change, probably not a whole lot. I didn’t really think the ’k’ was too narrow, but the ’K’ is. I might edit the ’k’ a bit to fix any optical illusions that might exist that makes it look smaller (because I believe it’s fairly the same width as some of the other characters).

Using H as a guide sounds good, it does stand out a bit at the moment, because it’s so vertical and massive. So adjusting the rest to it will make everything look in context.
The caps I have in the latest PDF, are first drafts only, with a lot of issues still. Your tips will probably help me fix those a bit, especially since these are practically my first capital designs. :) I was partial to the serif on the ’C’ as well, but without it it didn’t look quite right. Have to try something different.

Thanks for your input so far, James. :)
Btw, what do you think about the lowercase ’s’? It still doesn’t quite look right in my opinion, but it’s starting to get better, isn’t it?

Hey Jelmar :^)

I’m also experimenting with the the square-corner ’b’ shape for my typeface, because it (and then also ’g’ and ’q’), don’t seem to look right as they are now.

That’s probably because the square-cornered b and square-cornered monocular g are common solutions in sans serif types. What looks “normal” in a font depends on what you’re accustomed to looking at. But disregarding all that for a moment, I can think of a reason why the square-cornered b, g, q would look natural in Rough-cut—-it will harmonize with the inner kinks.

Experiment with the width of k and post some tests, get feedback, then decide. It’s up to you. Your k is narrow but not extreme, so I don’t see it as a major problem.

Try a shorter serif on C with the inside vertical angled to the right (rhombus shaped). Lowercase s has improved. I think the problem is the angle of the spine, it’s closer to horizontal than usual. Try and tilt it clockwise away from horizontal to make the angle steeper. I can’t guarantee that will work tho.

j a m e s

I’ve made quick versions of the square cornered ’b’ and ’g’ (PDF in opening post). I think (in principle) it fits better with the style and the kinks. I guess you’re right about finding something ’right’ is based on what you are accustomed to. Which I find no problem, I’m not a great fan of doing something different just because it’s different (it always sounds a bit pretentious). ;)

About the ’s’, that was what I was thinking too. The spine is too horizontal. Making it more vertical would make the character significantly narrower as well, is this a problem? I guess not? I also don’t really like the entry and exit strokes in the ’n m’ and ’u’. On the ’p’ it somehow doesn’t look too bad. Maybe change it back to like it is on the ’d’? I’m not sure what I should do about it yet.

I’ll change the ’k’, ’s’ and ’C’ (and the rest you mentioned) tomorrow. The sample I posted in my opening post is more for the ’b’ and ’g’. I curved the top right corner of the ’g’ down by a couple of units, so that it mimics the top curve of the ’d’ more (instead of a straight rotation of the ’b’ + a tail). Otherwise those characters didn’t look right together. What do you think?

I’m not a great fan of doing something different just because it’s different...

Difference for its own sake can be tedious. But if a design is different for practical reasons, or new ideas that make sense—-as long as it’s practical and makes sense.

Narrowing s won’t be a problem. I’m sure you can handle it.

...entry and exit strokes...Maybe change it back to like it is on the ’d’?

Try that. Also try a gentler curl with less lateral movement, more like a bend.

b and g look good now. The bowl of g can go even shorter to make the space enclosed by the tail less cramped.

j a m e s

Agreed. Higher on the g. Try the bowl at maybe 95% of it’s current height.

Ok, I’ve made quite a lot of changes (rough-cut-5.pdf)

g - Shortened the bowl vertically (it now rests on the baseline, no bottom overshoot).

h, m, n, r, u - I have ditched the entry/exit strokes for now, and rounded the shoulder, so it joins with the stem at a lower place and looks more natural.

s, S - Redrew (or redrawn?) the s/S, I think it has improved drastically this time. They are both narrower than before too. I doubt I can make them more narrow/vertical on the spine, without making them look weird and too narrow.

C - Changed the serif. (Is this more what you meant, James?)

E, F, K - Made slightly wider.

Y - Raised the bowl, where the two arms meet. (It doesn’t exactly mimic the X as James suggested, but it does approach it more now.)

I think that’s it for now...

I like the entry stroke on the ’p’, I’m hestitant to change it, eventhough the entry stroke isn’t on any other character anymore... which I think might be somewhat problematic style-wise? I’ve also slightly loosened the spacing, which will be even looser when I have lighter weights. Well, new critiques are welcome! :)

I also only just noticed my previous PDF’s were huge because I accidently exported them CMYK + profile. New ones are smaller. :)

Looking good. Right now I’m burried under a pile of work. I’ll post later in approximately 12 hours.

j a m e s

Ok, that’s fine. :)

g: whitespace is better distributed now, no need to change it any more.

h, m, n, r, u: lower shoulder is an improvement. See drawings below for suggested entry stroke.

s: spine can still be tilted further clockwise, without narrowing letter body. It’s difficult, perfecting it seems impossible, but you’ll get there with persistence.

S: fine.

C: yep, that’s what had in mind. See drawing below for specifics. I made the serif a bit shorter and angled the bottom to harmonize with the branching strokes in lowercase.

E, F, K: much improved. K could go a bit wider, possibly with a shorter curl. Don’t be shy.

Y: approaching X is fine, it doesn’t have to match. It’s the spirit of what you’re doing that counts, not the letter ;^)

p: leaving the entry stroke as it is makes p the odd man out. Nothing wrong with that, but check the drawing below for an alternative.

The increased spacing is working nicely.

r suggestion: More mass at stroke end, modified entry stroke. Arm end treatment may look exagerrated when you’re looking at it close-up in the glyph window, but at practical rendered size it’s a different story. For r context is very important; think of it not as an individual letter but in terms of how it works with other letters.

j a m e s

r needs some overshoot too.

j a m e s

Thanks for that elaborate visual aids, James. :)
The entry strokes, I actually have something similar somewhere in the mask layer of some of the characters. I didn’t like it before, but that was 3 weeks ago. I think I might give it a try. Although I do think the left indent will probably have to be a bit higher, because there seems to be a concentration of black in your examples (because the stroke becomes more or less diagonal). But that might also be because the images are so large. :)

And the angled bottom of the C-serif is just one of those things that don’t pop up in my mind yet, because of little experience I guess. I will probably incorporate it. About the ’s’, I’ll try to edit it even more. I like yout ’r’ suggestion.

By the way, James (or someone else), what Encoding is best to select in FontLab? I’ve been switching back and forth between different encodings (Standard Encoding, MacRoman, Adobe OT PS standard names CFF, MS Windows 1252 Western, etc.), but I would like to choose one that is smart to choose. :)

Uploaded new pdf (rough-cut-6.pdf), which has some (not all) of the changes I wanted to make.

Iteration #6: the entry strokes/bent stem tops blend in seamlessly. s geometry is right but the spine looks too heavy. Serif on C is shorter than my eye (mind) expects; apologies if I misled you on that one.

I normally build a font with 1252 Western or Mac OS Roman encoding viewed in Names Mode. Before compiling (generate font), here’s a sequence of steps courtesy of Ray Larabie, who certainly knows what he’s doing:

1. Switch the font to Unicode mode by clicking the red UNi button.

5. Switch back to names mode. I use MacOS Roman but I don’t know if it matters. Choose glyph/sort glyph/by Unicode. Not only does this make your font look tidy in the font window but it sorts out the Index table. If you switch to Index mode, you’ll see that all your ducks are in a row. This will make your font look tidy in the Adobe glyph table, Windows character map and for nosey people looking at the index table in FontLab.

After that it’s safe to compile.

Can we have more advice on encoding from the technically adept please—-Paul Hunt, John Hudson, Hrant? Thanks guys.

j a m e s

Yeah, I think the entry strokes work pretty well now. I will leave them alone for now, and revisit them in the future if it’s needed. It looks much smoother now when applied in words/text. I had to slightly narrow the top stroke joining the stem on the ’p’, otherwise it didn’t look right.

I think the C-serif might be a bit short too. In the Glyph Window and short preview strings it looked alright, but in the longer text in the PDF, it shows it’s a bit too short. No problem ’misleading’ me, I don’t see it as such. It’s always good to try different approaches. I will just make it longer. :)

I’m quite satisfied with the ’s’ now, but I can shave off some weight of the spine without changing the geometry.

Thanks for the encoding tips. I will have to try that Private Use stuff out in a new file, because I have no idea how that works. More advice on encoding would indeed be appreciated. :)

just fyi: adobe discourages usage of the PUI.

I guess PUI is the Private Use? I don’t really need Private Use characters I think, so what ’standard’ encoding would be wise to select in FontLab at this point? I want to make it OpenType with some basic features (I’m teaching myself basic OT coding at the moment).

I’m not aiming for massive language support, and maybe only want a couple of ligatures.
And if I want to make things like smallcaps, should I make a seperate SC font? Or just a very large typeface with incorporated SC?

Thanks for that Paul.

Jelmar,
I forgot to explain the sequence Ray describes is solely to ensure all extra glyphs, eg: alternates, ligatures, smallcaps, have a bona fide Unicode index number. This helps ensure they will survive cut and paste operations between different apps. Step 5. “Switch back to names mode...Choose glyph/sort glyph/by Unicode” takes them out of the private use area. So ignore the Private Use thing as it is only an incidental step.

I’ve made fonts with Mac OS Roman encoding and other fonts with 1252 Western encoding, and both encodings work perfectly on both Macs and PCs. I think the safest bet is Mac OS Roman.

Incorporating smallcaps in one OT font requires separate kerning pairs for the smallcaps to get them to space properly. Technically it’s not hard, just more kerning work.

j a m e s

Ok, I’ll look into it.

Uploaded new PDF. Changed all kinds of small details. Made the serif on the C longer. Made some of the inktraps a bit more subtle. Widened the k/K, widened the e (only with about 5 units on 2048 em), and probably some other small changes here and there.

“t” looks a bit heavier from the rest. So do “k” and “w”. Text is way too tight. “s” looks a bit too narrow. UC “S” looks more “open” compared to lc “s”. “y” looks like it’s falling to the right.

UC looks very nice. But I’m having some problems with lc because there seems to be too much style variation between different letters. Maybe the too tight letter spacing is the problem, maybe the wrong kerning or maybe there is too much style variation going on. But I think it’s the combination of all three.

When looking at the pdf UC seams lighter to me than lc. There should be little difference between them so that they look the same weight (because caps are taller).

Thanks for the replies.

@Robert:
I will check the weight on ’s k w’. And it is set too tight, because the metrics are still from when I wanted it to be purely display. I was going to loosen it.
The ’s’ too narrow... I just made it narrower because people found it too wide. I will see what I can do, but the critiques are contradicting unfortunately. But I think I liked it better too when it was a bit wider than it is now.
The ’S’ is more open, because there is more room for negative space. I can’t really make the ’s’ opener, because it would become too light. I think I will have to wait what happens when I intrapolate it for a lighter weight, and reverse engineer any changes I can make then in to the bolder weight (this one).
There is no kerning. What do you mean with style variation? I would like to fix that, but I’m not sure what to fix. The letterspacing will be loosened with the next incarnation. :)

@Jos:
Yeah, I also noticed the UC looks lighter, I think I might have to make them a couple of units heavier.

Jelmar, I think you should continue shaping this in the direction you feel is natural. I don’t see the stylistic variation as a problem, more like a characteristic of the design that makes it what it is.

w does look a bit heavy, but t and k look fine. y falling to the right could be due to the ink trap veering left. A lighter weight version will put a whole new perspective on the font. I’ll wait until then to make further comments.

Robert, are you looking at this on-screen or in print? You may be seeing rendering anomalies on-screen. There isn’t much point comparing weight density and counter sizes of the same letter between the upper case and lower case—-most fonts have these inconsistencies once they get to a bold weight like this. What you’re picking up is a limitation of the dual case Latin alphabet structure.

j a m e s

Well, I realise I’m somewhat doing it the wrong way around. I think it usually is ’better’ to begin with a lighter weight, and then use that to create bolder weights. But I wasn’t really planning on putting as much time in it as I’m doing now. It was going to be a bolder display typeface, but now I want to make at least a normal weight as well, both with italics. :)

I’ve been changing and planning to change all kinds of things, I’ve slightly increased the weight of the caps for example. I’m not going to upload a new PDF at the moment, that’s for later this week.
I’ve also tried removing the curly tail from the ’k’, I like it, but then I don’t like it. I might put in one of them as alternate or something. I’ve also changed the ’y’ drastically, and I don’t think it’s falling over anymore

Sweeeet.

I’ve been there too, started on a font and discovered it was much more work than I thought. But you don’t seem fazed by it, so that’s a good sign.

Have a good week. :^)

j a m e s

> But you don’t seem fazed by it, so that’s a good sign.

Hehe, no I’m not really. I know (and knew) how much work designing a typeface is. But I never went any further than designing one weight. But I thought that with enough work this might become something nice, so it’s worth spending more time on.

You have a good week too. ;)
I have a week off school, but I had planned to help my parents paint the inside of the house (or at least the living room for starters), which is quite a lot of work. :) A little break from sitting behind the computer wrangling with bézier-handles is good.

Addressing your original concern that ’The ’s’, ’v’, ’w’, ’y’ and ’z’ are the characters that trouble me the most...’

Perhaps this is because these letters (I include the ’x’ as well) either don’t have any ’kinked-curves’ in them or they, in the case of the ’s’ and ’y’, are so minimal that their appearance becomes too regular as compared to the others. This, I think, violates the design intent of your font, which I take from the name ’rough-cut’.

I would suggest that exaggerating the ’kinked-curves’ in the ’s’ and/or adding some into the otherwise ’straight’ ’v’, ’w’, ’x’, ’y’ and ’z’ might bring these characters more into line the rest.

Not to be presumptuous but I made these, to try and illustrate my point.

Hi Jelma,

sorry that I am actually late for this discussion..

You have made a nice and serious work!
I like the good feeling that you seem to have for the pen stroke. As in the ’g’ above (I’d prefer the right one).
Personally I am a (Super)Light-freek. Seems you are on the contrary ;-)
But it is in fact nice!

The problem of the (very) light versions (but also the joy/thrill of refining it) for my eyes seems in general to be that you have to cover (relatively to the design size) longer distances between two curve points and define them via tangents. That makes it sometimes difficult to keep up with a warm and differentiated penstroke-curve without making it to fragile and weeky.

I am also curious about the whole interpolation story for usually I have (as you can guess) the converse problem, making my designed glyphs thicker in the end for using them in text.
I am a great fighter for charme and warmth in typefaces. Usually I have to invest a whole bunch of work refining everytime I used the Bold filter in FontLab. But it is in fact already so much better than all we had in the past..

Keep on with your typefaces! Great!
Many links of you to other design work had shown me in the past that you have great taste.

Stefan

Rick: Thanks for your reply. I have experimented with such shapes for the characters you mention, and I wasn’t sure if I liked it. I might try to find back those sketches, but I don’t know if I will be succesful. ;)
However with most of those experiments it became apparent that the characters in question quickly became too different. But I welcome all input, especially elaborate ones with images, so thanks. :)

Stefan: No problem with the late entry in the discussion, as I said above, I welcome all comments. :) You’re right, I love the pen stroke, and wanted to incorporate it in the design, without going so far as FF Legato or Beorcana (which are two great typefaces).
About the ’g’, I have chosen the right one, which also translates to the ’b’ and the ’q’. Yeah, I’m a great fan of bold type, which this design was going to be initially. But I decided to expand it to more weights, so in the near future I will try to tackle a lighter version as well. I’m curious as well as how it will work out. I think I too will have a lot of cleaning up to do when I use the Bold Filter (with negative values in my case). And thanks for the compliment. ;)

>And thanks for the compliment. ;)

No prob
you earned it!

I uploaded a new PDF (no.8). Changed a couple of things, but not too much at this point.

h, m, n, u - Made them a little bit narrower, they stood out too much in my opinion.

y - Redrawn.

6 and 9 - Changed.

acute/grave - Changed. More horizontal now.

k - I’m thinking about doing the ’k’ like this, without the tail. But I’m not sure. I think it fits with the rest better.

h, m, n, u - Made them a little bit narrower, they stood out too much in my opinion.

Check! I was waiting for that :^) When you make a lighter weight their width will become more critical.

Acute and grave are improved. Both k’s work okay with the team. Probably simpler is preferable. Try k with half as much curl as interation #7.

j a m e s

Hehe, I thought they were too wide for a while now, but didn’t feel like / have the nerve to change them. But since I had not opened the file in about 2,5 weeks since the last pdf, I can look at it somewhat differently now (for a couple of days at least). ;)

I might indeed try a ’k’ with half the curve of #7, maybe it works. If it does I will leave both in for now.

And I forgot to mention that I also slightly narrowed the horizontal bars (the ones on the ’t’ and the ’f’, etc.) and I loosened the metrics quite a bit as well. That should be apparent in the section of ’1984’.

I also drew em-dash/en-dash, ( ), !?, and some more like cedilla, caron
(to make composites). But I forgot to update the character-map bit in the latest pdf.

I uploaded a new pdf (ultrablack.pdf) with an attempt to a ’morbidly obese’ (Mr. Schwartz’ great way of putting it) extrapolation of my typeface.

I was planning on finishing the current weight and make a lighter weight first, but I just needed to fiddle around with a ultra black weight now. Because I never did before. :)

There’s also a pangram in the pdf with the current weight, for comparison.

Jelmar, I have nothing to contribute except encouragement — this is is shaping up nicely. And it’s really cool to see the Typophile critique forum working as intended. James Arboghast’s generosity and thoughtful feedback are really inspiring, even to someone who has yet to design his first glyph.

Thanks for the encouragement. I really like it that the critique forum is working as intented too. Quite surprised as well, with the amound of replies and such. A lot of people, especially James, have been very helpful. Also the fact that someone (I’m not sure who) handpicked this thread, is very cool, specifically to give the critique board a bit more attention.

I’m still working on the typeface, eventhough progess has been a bit slower lately. I have to work on it on my own time, which is scarce. ;) Too much freelance assignments to finish first; I have to design and typeset 2 poetrybooks and 2 photobooks.

Marc,
yours is the kind of feedback typohile’s critique forums benefits from. As contributions go it counts for as much as practical advice. If my contributions encourage or inspire you to design your first glyph or even a first typeface we would love to see it—-when you’re good and ready :^)

Jelmar,
as I said before, take as much time as you need. It will come, it will come, it will surely come. The book projects sound like fun. I’ll echo that about the thread being handpicked. Thanks for that, mighty indefatigable typophile crew.

j a m e s

I’ve been fiddling around a bit lately, and I’ve now set-up first drafts of a ’regular’ weight as well. And the aforementioned Black weight.

So it now consists of Regular, Bold (the one this thread till now) and Black. I’ve uploaded a PDF called ’weights.pdf’.

I already saw that the M,N,W, etc. in both lower- and uppercase are too heavy. But other tips are, as usual, welcome.

’weights.pdf’ acquired. I can’t look at this in detail until after the weekend.

First impressions:
* for the regular font the spacing looks just a little too wide and you may have to bring the side bearings in to optimize. The ideal, according to Hrant, is to have equal areas of whitespace / stroke. 50% black balanced by 50% white. But that’s an idealistic plan and I don’t believe it applies to every text font design. Getting the spacing right is matter of experimenting until you hit the sweet spot. The sweet spot for Rough Cut will be unique to its design and characteristics. The regular reads a bit too slow for me.

* The regular font looks overweight. The difference between regular and bold, for most linears like this, is usually more pronounced. Since you’ve done all the work of making the regular what it is now, consider keeping it as a robust chunky regular, and make a thinner book weight that will give a more effective contrast with the bold. The bold weight is about right so I think stick with that as is.

* Other pros please post your impressions of the above aspects to help us nail this with confidence. Chris, Nick, Carl, Dino, Anyone else with eyes.

* Fatboy is mean and chunky. Will make a great display variant, ideal for vegetable soups, magazine covers, car insurance ads, vehicle signage, and a heap of other stuff too.

j a m e s

make a thinner book weight that will give a more effective contrast with the bold

you know, a hairline and a thin would also be nice. :^p

Thanks for the replies.

James, I agree about the spacing. It partially has the original sidebearings of the bold weight, so I will reset those to fit the lighter weight.
And the regular font does look a bit overweight, I thought that was the case anyway, but I didn’t mention it because I was hoping someone else would. To see if it’s really the case. :) Your suggestion for making another, somewhat lighter book weight sounds like a good plan. I’ll do some experiments about what weight looks right.
And I’m quite pleased with how the Black (or Fatboy ;)) is working out. It needs more work, but I’m a sucker for black weights.

Paul: That would definitely look nice. I was just thinking about that yesterday. I think I will see what happens if I make a very thin version, by making a very quick interpolation.

Yep-yep-yep. With this latest thinned-down version we’re starting to see how interesting things get when those kinks get close to the softer outside lines.

Make a few glyphs at a thinner weight and post a bitmap rendering and get some feedback. Say, A, O, T, R, V, a, b, d, e, f, h, i, p, o, s, v. That should be enough. When we think it’s on-target for a book weight and the sidebearings are about right you can make the rest. I don’t want to see you do too much work.

Cheers Paul :^)

j a m e s

I’ve added a new pdf called ’weights2.pdf’. It has all preliminary versions of the weights, and I’ve probably changed quite alot of the characters since the last pdf as well.

January 2008 iteration
Regular is very impressive, almost ideal except for the over-generous spacing. The weight will “bite” the whitespace and give improved immersive reading preformance with a small reduction of the sidebearings.

The weight will make a fine book type. The modulation and kink logic is so well refined now I can visualize a natural extension of the design. Let’s call it “kinked stressing”. You’re halfway there already. Would it be okay if I post some drawings to show the concept?

j a m e s

This anim shows a spacing treatment aimed at forming more positive word shapes or “bouma”.

Nothing at all to do with the next generation — Rough Cut Voyager :^) with its advanced kink serif design.

j a m e s

Hey James, thanks once again for the critical eye on my work.

I think you are right about the spacing, it should probably be a bit tighter.
I already artificially (as in; in InDesign) decreased the spacing a bit. But since I still have to make the final sidebearings, I will take that in account. The sidebearings of the Regular and Bold are a hodgepodge of remnants of earlier versions. Only the Black weight have better defined sidebearings, which are rediculously tight, which I think is nice.

Once the evaluation period at my art school is over (where I will show this preliminary version of my typeface), which is coming friday, I will start on the kerning as well, which of course will also improve the word-image (or bouma’s, if you will). And of course italics. And other details, like some characters that are too heavy (e.g. the ’t’ in regular).

And you’re welcome to post any images to show the concept you were talking about.

Can you decrease the tracking further with InD to tighten things up for the presentation? Just as a temporary means of refinement.

From what you’re saying you will need to create a new side bearings plan from scratch for the regular. Study a high quality commercial sans serif font of similar parameters to yours. In practice there is no point trying to kern a font until the side bearings are well and truly sorted. So don’t be in a hurry to start kerning.

Also, you should try to acquire good word form with sidebearings alone, optimizing later with kerning.

Rough Cut Voyager concept: I’m picturing vertical kinks in the stems, forming asymmetric stem structures. They should be subtle like the fractured entry strokes. Fractured vertical stem stressing.

j a m e s

I have posted a sample with decreased tracking called ’regular-spacing.pdf’. Decreased with -15. I’m not sure if that is enough? Or too much?

I will study a couple of faces indeed. And you’re right about kerning; I wasn’t planning on doing any before I’ve established the sidebearings.

I’m not sure if I’m picturing the ’Voyager’ concept right. Can you illustrate it? But I guess it should be called SevenOfNine or Janeway maybe? ;)

:^) Yes I was thinking 7 of 9 would be a good name. Hold on I’ll try to get some drawings done soon and post them. Sorry for the delay.

j a m e s

Ok that is cool. No rush.

I will have to take a short break from the typeface anyway. I have to get another monitor, since this one (which is already a backup) seems to show things a little bit stetched horizontally. I never noticed this, since it looks normal, but I noticed it because my typeface printouts looked compressed.

With other words: once I have a new monitor, I will have to re-establish the hight-width ratio of all my characters. Since they are actually more condensed then I had designed them on screen. So, not too great. But since I have to make italics and such, a little extra work isn’t such a big problem. And I might come to new insights while inspecting all the characters up close again.

This is the kind of thing I have in mind, with spacing correctins. The cuts to the type outlines are rough. Some familiar ideas there, namely fracturing and folding, most of which is aimed at generating more line contrast. The vertical kinks on the stems take up standard semi-serif positions. If you build a few characters like these and slant them the forms should lend themselves well to italicization

Your regular spacing PDF makes for an improvement in word formation but the space “character” is buckling under the strain. Setting will work much better with a wider space.

j a m e s

Ah yes, I like those cuts, I understand what you mean now. When I have a new monitor, I will definitely try something like that. The typeface had more quicks like that when I just started on it, and I might have smoothened it out a bit too much along the way. I might make a alternate version with rougher verticals like in your illustration. At least I think I’ll find it interesting to try.

I was planning on making my italic a bit like that indeed, for which I actually already sketched some letters a while ago. I will see if I can find those, and post them here.

Cool!