Caveat: I know little about the particulars of Cyrillic type design. However, I have been spending a lot of time looking at and evaluating how scripts [should] behave when mixed.

> Should I be using the small cap height for the Cyrillic x-height?

Which smallcaps? Aren’t you the guy who makes 7 sizes of smallcaps?  ;-)

Seriously, my view (which stems from some informal linguistic analysis) is that a smallcaps height which is notably taller than the x-height is indeed the best guide for the Cyrillic “x-height”. You can see the necessity for this as soon as you look at some text using the same guide for the x-heights of both Cyrillic and Latin: the former looks secondary, in apparent size.

On the other hand, you should know that this stance is somewhat anti-establishment. People like John Hudson (who I admit design aesthetically very solid Cyrillic fonts) are in the majority when they say that the x-heights should be the same. But they’re wrong. :-)

> maintaining a level of consistency with regard to x-height is probably not a big deal.

In fact that’s a clue as to the boundary between immersive versus deliberative reading: in the former (like setting a book) you need overall balance of texture/size more, while in the latter (like setting a poster) you need (or at least might prefer) visual alignment.

A reference:
http://www.themicrofoundry.com/ss_rome3.html

—

I personally go even further and state that characters that seem to have identical structures in both scripts should actually have different renderings! The reason is two-fold: emphasizing (even exagerating) cultural authenticity; and improving differentiation. For example look at the “Getty” sample on my link above: if it weren’t for those parentheses, the reader would be seriously thrown off.

As an example of what I’m talking about in real life, check out the Cyrillic of Quadraat. John doesn’t like it — I love it.

hhp

Hrant, I could not agree more with your last point: stressing inherent cultural differences by drawing (visually) common letters separately. This is exactly what I plan to do.

About the small caps: yeah, I’m all about quantity. It’s actually not that family that I’m working on, but another. I think I will plod along with a Cyrillic x-height matching the Latin small caps, slightly higher than the Latin lower case.

It’s been a while since I looked at Quadraat Cyrillic, I will have to check that out. What I would die for is a Cyrillic expansion to Clifford!

I love this stuff.

Thanks,
Alan

BTW, another reason to vary the rendering of [superficially] identical characters is notan (the balance of black and white): if one script likes its letterspacing looser than the other (for example as a result of a greater proportion of vertical stems versus rounds, like in Armenian), that would act to increase the internal white as well — and vice versa.

The tricky thing, however, is how to make the differentiation non-hoaky. I think it was John who said/claimed that Quadraat-Cyrillic* isn’t just “divergent”, it’s inauthentic. I might have to take his word for that, but I can’t help but remember a standard disclaimer of mine: what most type designers consider inauthentic might be just fine in the eyes of the reader. In any case, it clearly takes a greater-than-otherwise cultural sensitivity — when you opt not to copy the establishment, you’re deeper in the forest.

*
Here’s an image I used during my presentation in Thessaloniki:
http://www.themicrofoundry.com/other/quadraat_cyr.gif

hhp

Forgot:
On the other hand, look at how much smaller the Cyrillic text looks in that image. That’s what happens when you match the x-heights… Naturally, seeing their language set smaller pisses people off.

hhp

The Cyrillic may look so much smaller, partly due to the lack of ascenders. And to my non-Russian eyes, some of those letters look awfully snug.

But then how about the same writing system for a different language, like Ukranian or Serbian? They use Cyrillic characters, but add a few letters that have ascenders (or dots) that fill some of the space above the lower case.

The Quadraat Cyrillic Italic does look a bit affected in places. I think the gaps in letters like the zhe and kah are distracting for a lower case, and the descending zeh?! Is that what that is? Looks like it was just borrowed from the old-style three?

If John Hudson is around to chime in, that would be great.

Alan

I can’t comment on the individual letter issues you mention (I’m simply not familiar with them), but in terms of relative apparent size, it is in fact the proportion of extenders (mostly ascenders, yes) that causes the difference, assuming the color is equivalent. Interestingly, any difference in width (noting that the Cyrillic letters have much greater average width) seems not to matter — and I think I know why: lines are set horizonally, and line breaks “absorb” the width difference. But the Cyrillic text will generally run longer (especially since English is a very terse language), and you could say that’s where the greater width makes itself felt.

It’s also interesting to note that the Ukranian/Serbian factor affects Latin as well, in fact more so! I once did a study of the relative vertical span of 6 Latinate languages, and the two extremes (English and Polish) were pretty far apart. Ergo: Polish needs to be a set a little bit smaller than any parallel English! But to do this sort of thing you need to look closely at letter frequencies, not just the alphabet. The Greek alphabet for example has a greater proportion of extending characters than English, but in real text it’s even more confined to the x-height!

BTW, Maxim Zhukov & George Sadek’s “Typographia polyglotta” is a great resource for developing some linguistic sensitivity towards mixed typesetting. Just try not to squirm when you read the last paragraph on page xvi… to me it almost ruins the whole point of the book! But that slip of “fundamentalism” notwithstanding, it’s really a great little gem.

hhp

Yes, I have the book you mention, I’m searching frantically for it. In its absence, I’ve been referring to omniglot.com, among other resources.

The sample Polish text on Omniglot doesn’t look too riddled with extenders, but I’ll take your word for it. Then there’s a disaster like Vietnamese written in the Latin alphabet! For some reason I had always thought that Vietnamese had no written language until it was forced upon them in a Latin form, but clearly I was wrong. But given the complicated accents, you could see how one could assume that the written form was an afterthought. But that’s an entirely different topic.

Thanks for your insights!

Alan

(double post)

> a disaster like Vietnamese

“Disaster” is overly kind.
It reminds me of Carmen Miranda.



hhp

That would be a rather perfect comparison, I would say

I still can’t find my copy of Typographia Polyglotta. Do you know where I can get one? I know Emigre used to sell it (that’s where I got my first copy).

Thanks,
Alan

I got mine from Emigre too — do they not have it anymore?
If not, I’d try contacting Maxim Zhukov directly.

hhp

Alan,

I think you’re right with your attempt to design Cyrillic lowercase slightly larger than Latin.

I was advised to do so by Maxim Zhukov who also told me that, when mixing Cyrillic and Latin using a typeface that has all lowercase of nominally the same height, he often uses a fractionally larger type size for the Cyrillic.

Regarding Quadraat Cyrillic, it isn’t well readable for a non-native Russian speaker such as myself. Native Russian speakers told me that Quadraat possibly looks sort of “Bulgarian”, if like anything at all.

I definitely recommend you to read the essays by Vladimir Yefimov and Maxim Zhukov published in “Language Culture Type” (ATypI/Graphis 2001): http://www.atypi.org/10_Visitors/70_publications/50_LCT

Regards,
Adam Twardoch

Many thanks, Adam. I have a reminder to myself to buy that book. I was reading recently about Cyrillic from Bulgaria (or was it Belarus?), but now I can’t remember where it was. In any case, I’ll take your word for it, and assume that Quadraat Cyrillic is not the best model for what I’m studying.

Thanks for the link, too.

Alan

Go Maxim go!
John, take his advice, eh?  :-)

> Bulgarian

Now that’s interesting.
Any ideas why?

hhp

First of all, in addition to the articles in “Language Culture Type”, by all means, I recommend reading Vladimir Yefimov’s primer on designing Cyrillic letters (in Russian only, unfortunately):

Part 1:
http://www.paratype.ru/ru/e-zine/issue01/knowhow.htm
Part 2:
http://www.paratype.ru/ru/e-zine/issue02/knowhow.htm
Part 3:
http://www.paratype.ru/ru/e-zine/issue03/knowhow/typemak3.htm

Examining various approaches to cyrillization of popular designs such Helvetica can be very insightful.

Russian-style Cyrillic:

Helvetica Linotype (Linotype: Cyrillic 2001 by John Hudson and the Linotype staff) — my favourite
http://www.fontexplorer.com/FontStore/661573179/Product/View/16841171

Pragmatica (ParaType: Cyrillic 1987 by Vladimir Yefimov and Isai Slutsker) — my second-favourite
http://www.paratype.com/pstore/charset.asp?scode=PT_PRG55&refcode=123495

Helvetica Cyrillic (Linotype: Cyrillic 1992 by Jovica Veljovic and the Linotype staff) — not so favourite
http://www.fontexplorer.com/FontStore/661573179/Product/View/16111680

Helios (TypeMarket: Cyrillic 1993 by Alexey Kustov and Sergey Shanovich) — not so favourite
http://www.paratype.com/pstore/charset.asp?scode=TM_HLS55&refcode=123495

Swiss 721 Cyrillic TL (Tilde: Cyrillic by Andrejs Grinbergs) — not so favourite
http://www.tilde.lv/Fonts/fontsamples/pqrs/swiss721cyr.htm

Nimbus Sans (URW++: Cyrillic by URW team) — rather poor
http://www.myfonts.com/fonts/urw/nimbus-sans/l-regular/win-t1/97112/charmap.html

Bulgarian-style Cyrillic:

Helen (Hermen Software)
http://www.hermessoft.com/newproject/packages/helen.html

In general, you can notice that the “Bulgarian style” of Cyrillic letters usually involves more ascenders and descenders than the “Russian style”, e.g. in “ve”, “zhe”, “ka” or “yu”.

Also, the upright letterforms “ge”, “de”, “te” or “sha” often have the same styllistic features as the corresponding italic forms.

Some experts disagree that these “Bulgarian variants” really are representative. My suspicion is that Russian-style Cyrillic is perfectly acceptable to Bulgarian readers just as the Bulgarian-style forms, while the opposite isn’t true: the Russian-style forms are the only ones acceptable for Russian readers.

Serbian-style Cyrillic:

Similarly, there are some particular letterforms that are distinct in Serbian-style Cyrillic — especially in italic style. These include “ge”, “de”, “pe” and “te” that tend to be more influenced by calligraphy than in the Russian-style Cyrillic. Helvetica Linotype implements this in the OpenType “locl” feature.

Please note that, for example, Jovica Veljovic’s Cyrillic designs tend to be slightly more of Serbian style than of Russian style.

There are good samples of Cyrillic and Western character sets of fonts designed or adapted by ParaType designers and other Russian designers (especially the work by Vladimir Yefimov and Tagir Safayev is worth checking).

For example, check out Vladimir Yefimov’s Octava. Pay special attention to the Latin and Cyrillic K, and then to the Zhe.
http://www.paratype.ru/pstore/charset.asp?scode=PT_OCT55&refcode=123495
http://www.paratype.ru/pstore/charset.asp?scode=PT_OCT55&codingList=11&refcode=123495
Also, check ITC Charter Cyrillic
http://www.paratype.ru/pstore/charset.asp?scode=PT_CHR55&refcode=123495
http://www.paratype.ru/pstore/charset.asp?scode=PT_CHR55&codingList=11&refcode=123495
ITC New Baskerville
http://www.paratype.ru/pstore/charset.asp?scode=PT_NWB55&refcode=123495
and other serif designs
http://www.paratype.ru/asearch/sresult.asp?ftype=SE&enclang=enc&fenc=1&ffoundry=PT&refcode=123495

I consider Matthew Carter, John Hudson and Gary Munch as the designers who are most successful among non-natives in designing Cyrillic letterforms. Carter’s Georgia and Verdana, Hudson’s Sylfaen and Helvetica Linotype, and Munch’s Really are exceptionally well-designed Cyrillics. I also think Robert Slimbach’s (and Carol Twombly’s) designs for Minion and Myriad are well done.

Finally, I’d recommend contacting experienced and highly valued experts who could express some critique on your letterforms. I think contacting Maxim Zhukov, Vladimir Yefimov and Jovica Veljovic would be best choice.

Best regards,
Adam

Adam, we need to frame this post! Thank you.

> in Russian only, unfortunately

I once threw an automatic on-line translator at this, and the results were pretty intelligible, and totally entertaining!  :-)

hhp

Yeah, wow. This is great, thank you so much, Adam! “Save as PDF”

Adam, great! thank you have you any Veljovic cyrillic works to show? i can’t find it on web. please

hhp, Quadraat Cyrillic is awful, espessially italics… as for native-Cyrillic reader

> Quadraat Cyrillic is awful

:-(
I can believe it. But still, there must be a way to make Cyrillic different than Latin but usable nonetheless.

hhp

Andrij,

> have you any Veljovic cyrillic works to show?

I cannot speak about this _at the moment_.

Regards,
Adam

Adam, przepraszam.
i want to say ‘do you know any places on the web with Veljovic cyrillic calligraphy’

Hrant,
i am sorry. if you looking for ideal cyrillic, look on Minion, Myriad or Georgia

Alan,
can you show any samples? i am very interested.
x-height is same like latin lower case. cyrillic small caps burn a discussion, but generally not availaible. but SC not the same as LC anyway

–—

sorry for my english again. i need to learn it

> x-height is same like latin lower case.

Usually it is, but as we seemed to have arrived atr in this thread, it rarely should be, because that makes the Latin text look bigger (= more important) than the Cyrillic.

hhp

Hi! I’ve run into this discussion by chance, and obviously, too late. A few things got mixed up here.

1. That the small caps exist in Cyrillic, is an indisputable fact, impossible to deny. Good typography is hardly possible without small caps, either in Western languages, or in Russian, or other languages using Cyrillic. Just have a look at the digital fonts offered by ParaType that feature ‘SCOSF’ character sets.

2. That in multilingual typefaces that cover both Latin and Cyrillic scripts the same x-height is used, is another fact. One can theorise ad infinitum whether this is the best solution but that’s the way it is (challenging this practice has nothing to do with being ‘anti-establishment’: this is pure coquetterie).

3. In designing multilingual typefaces that cover both Latin and Cyrillic scripts it is standard to use identical shapes for different characters in both sets: for example, the capital A, B, C, E, H, I, J, [K], M, O, P, S, T, X, and the lower-case a, c, e, i, j, o, p, s, x, and y (in roman, and more in italic).

4. Yes, Russian text looks different from English when set in the same point size, on the same leading, track, with the same horizontal scale, etc. Of course, in the same typographic format English text looks different from Italian, Spanish text looks different from Danish, etc., despite they all use Latin script.

5. Adam, I never told you that Cyrillic lowercase should be designed with slightly larger x-height than Latin. What I shared with you was a typesetting, not a type design, tip. Manipulating the point size and the tracking is sometimes necessary for better visual balance of the Russian text with the text in English, or in any other language using Latin script.

6. The design of Quadraat Cyrillic I find pretty amateurish, and the typeface unusable. As to Minion, that was the very first digital Cyrillic with a good pedigree, and one of the first typefaces the young Robert Slimbach did for Adobe; however, I am certain that to-day he could come up with a much better design.

> Russian text looks different from English when set in the same ….

Yes, it looks smaller, no? Doesn’t that convey “less important”?

> What I shared with you was a typesetting, not a type design, tip.

But why not translate that to the type itself, instead of assuming the typographer is an expert? Aren’t “smart” fonts good? The recent Whitman comes to mind: in the elaboration in Kent’s PDF there are descriptions of many “tricks” he’s incorporated that work in spite of the typographer.

In fact, I see a good reason to move that feature (differing x-heights) “deeper”, to the type design itself: if you take a font with Latin and Cyrillic components (where the color has been matched for the same point size, as it will always be if it’s any good) and scale one slightly larger, it will look darker too.

I’m not disputing what the standards are, I’m just pointing out functional problems with them.

hhp

Dear Maxim,

thank you so much for your insightful comments.

I am very sorry that I incorrectly remembered your advice you gave me some years ago. What you advised was not to design Cyrillic glyphs with a slightly enlarged x-height, but rather to give some extra horizontal space by slightly increasing the sidebearings of the Cyrillic glyphs as compared to Latin. I’ve mixed these things up, for what I’m sorry.

BTW: I just returned from a lovely two-week vacation in Yalta, Crimea (Krim, the Ukrainian peninsula at the Black Sea). I had a lot of opportunity to speed up my Cyrillic reading, and to observe many good and bad examples of Cyrillic applied typography. Plus great wheather and great sightseeings. And amazing food

Adam

Dear Maxim,

thank you so much for your insightful comments.

I am very sorry that I incorrectly remembered your advice you gave me some years ago. What you advised was not to design Cyrillic glyphs with a slightly enlarged x-height, but rather to give some extra horizontal space by slightly increasing the sidebearings of the Cyrillic glyphs as compared to Latin. I’ve mixed these things up, for what I’m sorry.

BTW: I just returned from a lovely two-week vacation in Yalta, Crimea (Krim, the Ukrainian peninsula at the Black Sea). I had a lot of opportunity to speed up my Cyrillic reading, and to observe many good and bad examples of Cyrillic applied typography. Plus great wheather and great sightseeings. And amazing food

Adam

Hi Adam,

I hope you had a great trip. What’s the food like there?

Did you happen to take any photos of the lettering? Or was this a *real* vacation from everything!

Alan

Alan,

the Ukrainians (and Russians, I think) like to put mayonnaise in their salads, so one needs to be sparse with them. The strongest side of the local cuisine are fish and sea fruits. You can get salmon in numerous incarnations. Also, since the Tatars and Armenians were and are present on Crimea, you get tasteful shasliks and other meat dishes mainly involving lamb. You also get various ravioli-like specialties such as pelmeni or vareniki. Crimea has fine wines and Champagne-like sparling wines. And, of course, you get caviar that is incredibly affordable (less than $50 per 100g).

Of course, I couldn’t resist making some photos of the lettering, but I tried to restrain myself.

Adam

Heeh… I just wrote that $50/100g of caviar is incredibly affordable. Well, it isn’t — that’s the price you pay everywhere in the world (sometimes even more). What I meant was that on Crimea, you can get 100g of first class black caviar for less than $10, sometimes even less than $5
Adam

Heeh… I just wrote that $50/100g of caviar is incredibly affordable. Well, it isn’t — that’s the price you pay everywhere in the world (sometimes even more). What I meant was that on Crimea, you can get 100g of first class black caviar for less than $10, sometimes even less than $5
Adam

> What you advised was […] to give some extra horizontal space by slightly increasing the sidebearings of the Cyrillic glyphs as compared to Latin.

That was just an idea. No one ever tried that, so it remains a hypothesis to this day.

> give some extra horizontal space

That’s a great idea.
I guess the reason for it is that Cyrillic has more straight stems than rounds compared to Latin.

When I was designing Nour&Patria it became clear that even cognate forms (like vo/”n”) have to be spaced differently (looser for the Armenian) because of the exact same issue (although probably not as pronounced).

Where this becomes even more interesting is this: if you accept that a glyph’s black shape depends on the white (both inside and outside), then you realize that looser spacing means the black shape has to change too!

—

Adam, did you see the ghost of Khrimian Hayrig? ;-)
He was a famous Armenian priest from that peninsula. After seeing Armenian issues get demoted from 16th to 61st priority during two subsequent post-WWI conferences, he gave us this cherished proverb: “To drink from the soup of the world, you must have an iron ladle.”

hhp

had to tag on another link:
http://www.prodtp.ru/index.php?module=subjects&func=viewpage&pageid=92
anyone know of any additional character design standards help for cyrillics?

“To drink from the soup of the world, you must have an iron ladle.”
Also, you need a very thickly padded pot holder to protect you from the heat :-)

“repetition is the mother of learning”

where’d that come from, chris?

?