Agamemnon 27! Now with Greek, Cyrillic, and Cherokee

cuttlefish's picture

edit: For those of you joining the thread in progress, the most recent PDF files are at the bottom of the list of attached files in this starting post. They all have version numbers, in case you lose track. I'm keeping the earlier ones around for historical documentation of the project, but you should feel free to skip ahead to the most recent PDFs and come back to the old files at your leisure.
--JP, 4-18-07

edit: This project has come a long way since I started sharing it with the community here. As it nears completion, at least in this weight and style, I need just a little more advice to turn that last corner. Eben Sorkin has been an immense help, far greater than I have a right to expect from any one person who I am not paying. The constructive advice of other experts (and I know you're hiding around here somewhere) would be greatly appreciated.
—JP, 2-14-07

----------

Here is yet another project I abandoned long ago. I'm trying to retrace my steps to how I created it and why I gave it that name. It looks more like a slightly circusy wood type with some modern attributes than anything Macedonian.

Indeed, it needs a lot of work yet. My initial output is what Fontographer expells as an EPS of all characters, which for some reason breaks some of the points apart so I can't apply a fill without tediously fixing everything.

I welcome any and all suggestions and comments, including on finding a new, more appropriately evocative name for the thing.

AttachmentSize
Agamemnon-outl.pdf53.92 KB
Agamemnon2-sample1.pdf20.98 KB
agamemnon4-sample1.pdf21.75 KB
agamemnon6-sample1.pdf22.75 KB
agamemnon7-sample1.pdf27.03 KB
agamemnon8-sample1.pdf27.48 KB
agamemnon9-sample1.pdf27.43 KB
agamemnon10-sample1.pdf28.82 KB
Agamemnon10-layout test-c.pdf78.54 KB
Agamemnon10-pangrams.pdf27.29 KB
agamemnon11-sample1.pdf29.05 KB
Agamemnon11-pangrams.pdf28.68 KB
agamemnon12-allcharacters-2.pdf41.38 KB
agamemnon12-pangrams2.pdf28.38 KB
agamemnon13-allcharacters.pdf41.75 KB
agamemnon13-pangrams.pdf28.76 KB
agamemnon13-randomtext.pdf18.02 KB
agamemnon14-allcharacters.pdf41.88 KB
agamemnon14-pangrams.pdf28.76 KB
agamemnon14-randomtext.pdf17.91 KB
agamemnon15-pangrams.pdf29.01 KB
agamemnon15-allcharacters.pdf42.01 KB
agamemnon15-randomtext.pdf18.07 KB
agamemnon16-allcharacters.pdf43.18 KB
agamemnon16-pangrams.pdf29.9 KB
agamemnon16-randomtext.pdf18.01 KB
agamemnon20-newer.pdf53.86 KB
pr-AgamemnonTwentyOne.pdf131.4 KB
pr-AgamemnonTwentytwo-chrlst.pdf158.22 KB
pr-AgamemnonTwentytwo-12pt2.pdf151.61 KB
pr-AgamemnonTwentySeven-ffd1.pdf298.85 KB
pr-AgamemnonTwentySeven-dis1.pdf348.33 KB
ebensorkin's picture

It looks like there could be refinements to the weight distribution and the heights etc but this looks really fun. I like your swooshy 7!

cuttlefish's picture

Here are a couple new ligatures from the expanded set.

ebensorkin's picture

It's hard to judge them outside of their real context but I like what I see so far. You must have been reading the same blog as me! I recognize the idea!

Choz Cunningham's picture

I like the new text figures. They have a broader feel in general, and the larger ball terminals compliment them well. The 2's diagonal seems a little too forced at the bottom, perhaps.

You have such a grasp on ligating the f that your new ones are beautiful on the first go. :) Once again, nice work. How far do you feel you are from completion?

Who's blog have you both been reading?

cuttlefish's picture

There was a blog entry referenced in another recent Typophile thread, regarding the proper desigh of eth and thorn, and toward the end it suggested creating ligatures such as those I made above. I wish I could find it but I can't remember what it was.

edit:
I found it!
http://font.is/?p=166

and the thread referring to it:
http://typophile.com/node/33006

Miss Tiffany's picture

Jason, without having to read this entire thread, are you considering stylistic sets? Your typeface has loads of character and potential, but in some spots, for my taste anyway, I'd want a more regular option here and there for some settings.

cuttlefish's picture

Oh, definitely there will be stylistic sets. I already have two Qs, and I may add an even plainer one. Is there anything else in particular that strikes you as too fancy for regular use?

cuttlefish's picture

OK, over the last month I've kind of gone crazy with extending the character set, possibly neglecting some more basic design features that will probably cost me in the long run. I have a whole bunch of ligatures, small caps, old style figures, diacritics, alternate forms and all that rot. I should have an image up in a little bit, once I figure out how to best showcase the new stuff.

cuttlefish's picture

Here it finally is. I skipped a few versions bringing it to this point. Still a lot to do, along with some glaringly obvious errors.

The PDF on top is the same as the above image, though the last line repeated for some reason. Small caps are coming soon. I had been working on them in the same font file, but that was getting way too cluttered so I copied them off into a new file.

Enjoy.

ebensorkin's picture

Wow. This is polishing up nicely. I think you have made a good case for this flavor in type. I really like the lc r, the Delta, and the UC DE&Z.

I have not looked in some time so I think my eye is fresh - here is what I am noticing now:

The weight of the c seems to light compared to the other glypshs. I think the top might be built up a lttle too. When your a zomed in it's not that noticable but when the letters are smaller it jumps out at me. The Cap C is also less poshed & strong that the Other Glyphs. On the f thorn ligature I might let the arc start slightly higher on the left. The join on the ct ligature could be a bit weitier as it joins the t. There seems to be a litle toomuch weight on the right side of the u. The top left serif on the AE lig might have more weight & flourish. The join in the middle of the a is bothering me. What if the join was straighter or more narrow? I think either thing would help it read as an a better/faster. The diacritics seem too close to me -especially over the d & h. The double story g needs much much more work. But I expect you know that. It's a good idea to have though! The Ogoneks sem clunky - too much weight in their tops I think. I amnot sure about the blob end on the acute/grave. I like it's style but I think they should be more narrow & pitched a bit hgher maybe. Have you read Adam T on french vs polish for this diacritic? The C in your copyright should be smaller. Why not go with a purely circular zero in the old style numbers? That would be cool. What abour harmonizing the ball weights in the old style a bit more too? They all look cool but they don't play quite well enough together yet. They are all grabbing too much limelight from each other because of the over-variation. What if the leg of the Cap l was shorter & had more flair & weight? What if the Join on the left side of the cap Y tapered slightly as it joined the base? The top right leg's Join on the Cap X might look better is it was slightly lower. Maybe that Breve should bend more/go higher. The Paragraph Mark seems a bit too heavy to me. The Q options will be a big selling point for this face. I think if the tails on the 1st two were heavier as they ended it would make the read as Qs more effectively. How about an Alt lc k.

I hope these ideas are useful to you. You have put in a ton of effort and I think it might pay off. I wold take Tiff's idea seriously too.

You know... The x height you have now is great for smaller text but the intensity of style is more appropriate for larger sizes Headlines and Display. Do you see what I mean? In a way I could imagine taking the design in 2 directions now. One for text that would be slightly softer in it's character and made for smaller sizes like you already have but maybe with some additional optical adjustments - Then you could have another one for Display & headlines that would keep it's current character but be for larger sizes and so have longer ascenders and decenders and maybe some of the joins would become less bulky - stuff like that. Have you sen font families that do that kind of thing? If not. I can point you at some.

Cheers!

cuttlefish's picture

I know what you mean about the weight being a little uncomfortable with the height. At text sixes I see it as kind of a medium or demi-bold weight, and a lighter version will be needed for the bulk of text use, but a somewhat reduced lc at this weight would be good for display.

As for the Polish accents and ogoneks, you mean the instructions here:

http://www.twardoch.com/download/polishhowto/intro.html

right? I thought I had followed the instructions, but I see what you mean about the ogoneks being a bit chunky. Something gave me the impression they should be that way, but it didn't look right to me either. I haven't set up the Polish kreska accented letters yet, but I have perpared a mark for them (between the cedilla and the ring). t- and d-caron need the abbreviated apostrophe instead of the regular caron/háček.

ebensorkin's picture

BTW I was reminded about my display vs Text comments when I looked at this today. Have look - maybe it will be helpful.

http://www.dutchtypelibrary.nl/PDF/M504/DTLFleischmann_M504.pdf

( page 9 to start )

cuttlefish's picture

A PDF of version 21 is up now at the bottom of the top post.
This is what comes out of the Print to PDF function of FontForge.

I've gone a bit crazy with some of the features, especially the automated accented glyph assembly, so there may be more remaining work to do than there was before I did all of that.

Perhaps I should pare it down for initial release?

Might anyone have suggestions on making an italic to complement this?

ebensorkin's picture

It's hard to compare with the glyphs sitting off on their own little spaces...

cuttlefish's picture

True, I output this one in a bit of a rush. I'll try to put together something resembling the earlier samples for better comparison.

ebensorkin's picture

Cool.

litera's picture

carons on Z and z are off. The same goes for S and s. I live in the language area that uses Č芚Žž. And this kind of caron positioning is quite unnatural to me. And of course I'm missing the Ccaron and ccaron. :)

cuttlefish's picture

I'll get those Ccarons in there and correct the others. Most of the accented characters were made using the automated assembly feature (element> build> accented glyph) and still need some adjustment to be centered properly.

cuttlefish's picture

OK, there are a couple new sample PDFs loaded to the bottom of the list again. I am getting the impression I totally have to rework the spacing and kerning, but FontForge's metrics window is infruriatingly slow, at least compared to Fontographer's, and you can't load in a list of kerning words as far as I can tell. Anyway, there are a plethora of new characters now. I've nearly filled out the Latin B area, and some of the IPA and Greek are getting filled in since some of the Latin B lowercase or derivors are in those areas, not that it matters a whole lot, or maybe it does.

I'm not sure where I'm going with this anymore. I feel like I've lost focus, yet I'm driven to keep working on it.

Choz Cunningham's picture

There are two ways you can go, I think. Either treat it as a product, and you need to make it happen, or as not one, and you don't have to.

If this is a product you want to sell, each additional part demands a multiplication of refinement. In software-land, feature sets get locked well before a product's ships. Even really great ideas get shelved for later versions because otherwise, the first version never gets out the door. Cut out the most belaboring aspects, add what is needed to stay competitive, and keep track of your hours. :)

If it is a labor of love, your magnum opus, and all about the journey...
Take time, make more notes, get feedback and keep adding things. Or go for a walk if it isn't too hot. Have a life. Let things stew a bit in your brain and see what, if any, is the purpose. Get back to it when your vision takes you there. But don't rush or over-analyze it until you are sick of the project.

It's a good face, and everyone here "can't wait" to see it released. But we actually can wait. So figure out how far this needs to go for you, and go. Then stop. :)

litera's picture

Sorry to say that but I think that both your "S" and "s" are a bit off leaning right. Or at least the curves aren't as good as they could be. I'd try to correct that.

Otherwise I agree with Choz. He's right. If you feel it's sufficiently usable, release it. If you think some major things are still missing, then finish them and release later. But be a good judge about this.

I'm in software development and I know how can details distract you from releasing an already usable and fine enough product.

ebensorkin's picture

Choz makes some really good points.

My guess is that when you are making the Latin B etc you are most likely working in an abstract way in the sense that you are taking an (unfamiliar?) shape and deciding how to interpret it. It's great - but fairly abstract.

If you want to evaluate what you are up to and gain focus - I suggest that you apply your type to a design project. Even a faux one - but ideally a real one. That will help you notice what is & isn't working - and ground the work in a 'reality'.

What did you make of my display vs text shapes comment? Relevant? Irrelevant? Why?

cuttlefish's picture

OK, maybe moving this into FontForge wasn't the best idea. It's a fine enough program for designing, but might be a nigh bid more powerful thatn I can handle in metrics handling.
In Fontographer, you could load in a text file of kerning words and shuttle down the list while making adjustments in the metrics window.
I cannot figure out how to do the same thing in FontForge, if it is possible at all. It appears that one has to type in whatever letters one wants to see in the metrics window. So inconvenient.

But anyway, while I've had trouble working on the spacing, I haven't been idle. I've expanded the character set to include support for Vietnamese, lots of obscure languages, as well as Greek and Cyrilic. Those last two are mostly just an experiment, since I have no idea how those are really supposed to be. but I'll have pics pf them up soon so you can tell me what you think of them.

cuttlefish's picture

I have expanded the character set into a very large chunk of the Latin Unicode space (including coverage for Vietnamese and a buttload of other stuff don't know), as well as limited inclusion of Greek, Cyrillic, and Cherokee. Actually the Cherokee is pretty complete but I have no idea whether or not it would be legible to native readers (same with the others too). I'd be happy if these scripts just don't look too stupid or inappropriate. I'm sure to get flamed for deriving too much from the Latin for these other scripts, but hey, I'm learning here!

There are some bugs involving missing or incorrectly ordered glyphs in the new output files (pr-AgamemnonTwentySeven-dis1.pdf and pr-AgamemnonTwentySeven-ffd1.pdf, attached in the first post), but those are FontForge and/or Acrobat issues and I've been talking to George Williams to get those resolved.

cuttlefish's picture

Here are some pictures.

Number3Pencils's picture

Some of the foreign-language characters aren't looking good.
Russian:
Zhe should definitely point outward. zhe could probably use mirror, not rotational, symmetry, much as it makes the chirality seem wrong.
Left leg of El/el should be light.
Sha and Shcha are too narrow.
Ya should probably not have a light left leg.
be should have that flourish on top. This one looks very incomplete.
ef, strangely, should probably be rotationally symmetric. (Russian is full of paradoxes, isn't it?)
I'm not a Russian, but I think Russians would agree with me on this: Never ever make ascending myagkij, tvyordij, and yery.

Greek:
Lambda is upside-down.
Xi should have weight in it somewhere. Most people make the strokes thick to solve that.
Pi doesn't need serifs on its serifs on top.
Almost all the lowercase forms look very amorphous. Try drawing a "normal" text face in Greek, then use that study to make the Greek here less erratic. These letterforms are too all-over-the-place to read with anything like smoothness.

ebensorkin's picture

There may be things going wrong with some of the unfamiliar characters but the overall flavor of the thing still seems really nice to me.

cuttlefish's picture

I already knew about the Lambda. That one switched around spontaneously somehow while I was doing the Cherokee stuff. Thanks for pointing out the other mistakes. I knew there would be many. I'm working on fixing what I can right now.

Number3Pencils's picture

I forgot: that sharp corner on the cruved part of the De/de is really distracting.

cuttlefish's picture

Yup, I got the De, and spotted how that shape relates to the El in most instances in other fonts, so those are corrected. I finally saw what you were talking about with the Pi too. I'd always thought of it as a T with two legs (hence the seriffed serifs) but it's really more like an H with a raised crossbar.

Number3Pencils's picture

Precisely. Also, I hate to rag on your Greek more, but there's a basic misunderstanding of, I believe, the nu here. The alphabet is out of order, so I can't be sure, but if the glyph that's after eta and before xi is meant to be a nu, it's wrong. A nu is v-shaped, not y-shaped.

cuttlefish's picture

I see what you're saying, but I was trying to distinguish the nu from gamma, which, to my non-Greek reading eyes, also looks like a v, and upsilon, which seems to be in-between a v and u shape. What exactly are the distinguishing factors between γ, ν, and υ?

I may have figured it out just by typing that, but what's your take on it?

Number3Pencils's picture

gamma has a straight stem descending below the baseline, which neither upsilon nor nu have. upsilon is curved, and nu has a corner at the bottom, like U compared to V.

I've drawn up a conjectural evolution of the gamma and nu. I think the nu evolved along a similar path to the mu, so I showed the mu for comparison, because it's more straightforward. I know I've seen the second form for the nu, in GFS Complutum, which is from the 16th century. These evolutions are my educated guesses, not confirmed historical stuff, so don't quote me too exhaustively. I don't know how we get from the Upsilon form to the upsilon form, but I'd guess it was quite a bit of lazy handwriting among scribes.


Bow to my mad MS Paint skillz.

cuttlefish's picture

Has it really been over six weeks since I last posted an update? I did make all those corrections, I really did. I think I even made the Greek lower case almost usable, though it does still need work.

But before I post the next sample set, should I go and remove some of the old PDF samples? I've put a lot of them up here over time and it might be overwhelming people new to viewing this project. Which ones do you think show changes so subtle that they could be safely excluded from the history? Should just I start a new thread to give you critics a fresh start?

ebensorkin's picture

I would start a new thread and link back to the old one if this one is starting to feel unwieldy.

It's good to be able to follow process if you want to.

Cheers!

Quincunx's picture

You can also start adding the date or month to the PDF filenames. That way people know what the newest samples are.

cuttlefish's picture

The existing PDFs are all version numbered. I thought that would be sufficient.

cuttlefish's picture

OK, folks. We can pick this up again over here:
http://typophile.com/node/41105
I started a fresh new thread for the thing with some descriptive text. I hope it isn't too windy.

cuttlefish's picture

I've gone goofy with all these new currency symbols cropping up these past few years. Now it looks like I'll have to come up with one for the Icelandic kronur soon. Anyway, here's what I have so far:

riccard0's picture

Aren't the €'s bars a bit too far apart?
By the way, t's a pity that such nice glyphs like your spesmilo sign would be rarely (if ever) used.

cuttlefish's picture

I see it as not so much being that the euro's bars are too far apart, but some of the other horizontal double bars are, out of necessity, too close together.
And, though it's a rarely used glyph, I know a few Esperantists who might appreciate having the spesmilo.

Rob O. Font's picture

The Sm thing you show, I had to warn a designer recently away from a feature that was a different scale and color than the rest of the font. It had been a while since I had to do that, but it used to be a common occurrence. Scaleable type just can't deal with that kind of thing very well. This hardly seems like a face so shy one couldn't swell the bottom of the S to University Roman proportions to widen and weighten the m?

Also, the bars on the d thing, seem rather harsh.

I'm really impressed with the breadth of progress on this face, what a lot of good work.

Syndicate content Syndicate content