How about a two-stroke Upsilon (like U-cy)?

Is this Upsilon acceptable at all? (Latin and Cyrillic below for comparison).

Do you know of a precedent for such (mal)treatment of this letter?
I am aware that the shape of Cyrillic U is most probably derived from a vertical ligature of Upsilon and Omicron, so using this shape for the Upsilon on its own might seem illogical. But that set aside, does it look just plain wrong?

Comments

  • No idea about the Upsilon, but: Cyrillization yay!
    Although I'm not sure about making it descend.
  • Adam JagoszAdam Jagosz Posts: 233
    edited November 5
    In fact, I was wondering if it was plausible to make U-cy descend as well. (In handwriting/cursive calligraphy, it can, right?). Maybe a compromise, a semi-descender like in Q, half way down. Also kind of like the serifs in Д Ц Щ Џ etc.
  • Maxim ZhukovMaxim Zhukov Posts: 60
    edited November 6






  • In (Standard) Greek the Ypsilon has to have a straight downward lower Part. In fancy or scriptish Greek the Y can feature swashy upper arms, but no bottom swashes. In Latin it is usually the same standard shape as in Greek, but also the assymetrical form is possible (right stroke going down to bottom left). In kursive or scripty Latin a downward-left swash is possible, as well as the rounded glyph variant.
    In Cyrillic the У has a tendency for a descender, hence the height of the stroke’s connection lies rather low. In a similar shape for Latin that connecting point might sit higher, quite at the middle between base line and cap height.
  • Adam JagoszAdam Jagosz Posts: 233
    edited November 6
    Yes. I am talking non-standard here ;) And yes, that would mean a Upsilon-hook with the hook in the ‘wrong’ place. Why? It just occurred to me, since I'm making the non-standard (for print) move for the Latin Y, I might try to reflect that in Greek as well.
    As far as I know a wrong thing to do would be to give a straight leg to /U-cy, since that would be a different letter — Cyrillic straight letter U. But the opposite for Upsilon, what's the harm? (I will need to wait for a native reader to stop me).
    How about this for a precedent? Ha ha.

  • Yes. I am talking non-standard here ;)
    Bless you.
  • John HudsonJohn Hudson Posts: 1,489
    edited November 6
    Adam, the 'modern Greek alphabet in the ancient style' image from Shutterstock that you posted is not an accurate representation of ancient or archaic Greek letters. As soon as you see that form of Epsilon, you should be aware that you're actually looking at a modern invention popular in Greek Restaurant signs outside of Greece.

    For reference, here are the forms of Upsilon that are recorded in archaic epichoric Greek inscriptions. The 'standard' forms of Greek letters, are the Ionian alphabet, which everywhere became the norm, replacing local forms, by about 400 BC.



    I do think the hook on the bottom is problematic in your design. The two-stroke construction per se is seen in informal. non-cursive writing, but I don't recall ever seeing it with a hook at the bottom.
  • Hrant H. PapazianHrant H. Papazian Posts: 1,141
    edited November 6
    > a modern invention popular in Greek Restaurant signs outside of Greece.

    Huh, sounds like a decent use-case for this typeface... Versus a facsimile of some object from 2418 years ago.

    The Greek script, like Greek cuisine, is living culture, thanks to people anywhere in the world who can enjoy and benefit. It is not confined to a museum.
  • John HudsonJohn Hudson Posts: 1,489
    This is a nice typeface that isn't in lack of decent use cases. It would be a pity to see it undermined by a single idiosyncratic letterform that doesn't benefit the design particularly, and isn't within the constellation of shapes that constitute that letter for readers.

    I didn't post the image from my lecture on epichoric Greek to suggest that Adam should attempt a 'facsimile' of any of the archaic shapes, but only to illustrate that the form in the image he had posted was not, in fact, what it purported to be. The simple two-stroke construction without the hook is seen in modern handwriting, and doesn't require spurious 'ancient' precedence.

    Adam, if you're interested in non-standard forms that would fit the style of your typeface, perhaps try something like this?

  • Maxim ZhukovMaxim Zhukov Posts: 60
    edited November 6
    In Cyrillic the У has a tendency for a descender, hence ⟨…⟩.
    I would say, in Cyrillic Garaldes and Humanist Sans-serifs…

    B.t.w., these (⬇︎) are the primary forms of the У, Ж, Ф, and З in Veljović’s Sava (Adobe, 2003).


    The cap-high hooky У is an option.
  • Hrant H. PapazianHrant H. Papazian Posts: 1,141
    edited November 6
    > It would be a pity to see it undermined by a single idiosyncratic letterform that doesn't benefit the design particularly

    Clashing with the whole is a valid but completely different issue, one not resolvable via historicism.

    > isn't within the constellation of shapes that constitute that letter for readers.

    Based on my contemplation and observations, I think it must be. In fact some of those shapes from 400 BC are mere nebulas today.

    > the form in the image he had posted was not, in fact, what it purported to be.

    Where do you see what it purported to be? Beyond somebody's interpretation of Greek. Which is what we all do anyway, simply with different degrees of moxie.

    > The simple two-stroke construction without the hook is seen in modern handwriting, and doesn't require spurious 'ancient' precedence.

    Which is irrelevant to what shape could work (versus what shape is known to work). Like how an upside-down "V" does not need the bar to work as an "A" most of the time. (And nothing works all the time.)

    --

    Sava BTW has long been a personal favorite for its Y/У/Υ.
  • Hrant H. PapazianHrant H. Papazian Posts: 1,141
    edited November 6
    Typefaces need focus, and versatility is increasingly over-emphasized.
    It makes great sense for Adam to shoot for a multi-script system flavored for use in the Cyrillic sphere; a Cyrillized Y/Υ helps do that. Like this Cyrillized "g" I once made for a Russian client:

  • John HudsonJohn Hudson Posts: 1,489
    Where do you see what it purported to be
    Mouse over the image: 'modern Greek alphabet in the ancient style'. Adam had suggested this — with a smiley, suggesting he knows it's bogus — as a possible precedent. As I pointed out, he doesn't need that kind of precedent, because one exists elsewhere, in actual Greek use.

    It makes great sense for Adam to shoot for a multi-script system flavored for use in the Cyrillic sphere; a Cyrillized Y/Υ helps do that.
    ???

    Like this Cyrillized "g" I once made for a Russian client

    But that g still looks like a g, and conforms to expectations of the shape of that letter. i don't think putting a hook the bottom of an Upsilon does, that's all.
  • Hrant H. PapazianHrant H. Papazian Posts: 1,141
    edited November 6
    > Mouse over the image: 'modern Greek alphabet in the ancient style'. Adam had suggested this — with a smiley, suggesting he knows it's bogus — as a possible precedent.

    Oh. The filename. OK, I can accept its misleading... in proportion to how many people actually notice. Although maybe they simply thought "-inspired-by-the-ancient-style-" was too wordy. But really, the visual precedent is there, for people who only need that for inspiration, historic merit or not.

    BTW your reading of that "Ha ha." (not smiley, but anyway) seems convenient; it's like me claiming it means: "Precedents are ridiculous". Which I don't believe they are, but they do have limits, especially in proportion to their age.

    (Me trying to read that "???" is even less reliable.)

    > in actual Greek use.

    Is there some geogenetic test that certifies this? A terroir of type? Really...
    Any rendering that makes people think it's the Greek alphabet becomes a precedent for Greek. And if somebody like Oprah promotes it, guess what, Richard Porson can turn in his grave fast enough to power Iceland and it would still become the precedent to die for, honey. Nobody has to like this for designers to take it as the cultural reality we must work with.

    > But that g still looks like a g, and conforms to expectations of the shape of that letter. i don't think putting a hook the bottom of an Upsilon does, that's all.

    Formal conscious expectation, I can buy. But can they read it? It's hard to doubt that. And does it give it a Cyrillic air? (Assuming that's desired.) Well how else could it do so.

    Whether it clashes with the whole remains a valid question; although clashing to the point of causing outright rejection seems unlikely. Ergo: it just doesn't seem overly risky.
  • John HudsonJohn Hudson Posts: 1,489
    edited November 6
    I didn't see anything in Adam's original post to suggest that he's trying for a 'Cyrillic air' across the scripts. So better to let each script be itself, and resolve the stylistic similarity in ways that are idiomatic to each.

    Personally, I don't think this form is working well in any of the scripts, including Cyrillic: the join of the two strokes is too low and too dark in all cases. In this design, I might be inclined to try this kind of construction for the Cyrillic:


    [That's rough: the join could be lower.]
  • Ray LarabieRay Larabie Posts: 786
    @Adam Jagosz  I like how it visually relates to a λ...like a mirrored version.
  • John SavardJohn Savard Posts: 199
    edited November 7
    Your typeface certainly appears attractive to me. Obviously, only a native speaker of Greek could really address the question of whether this type of upsilon might pose issues or not. One possibility that came to mind, though, I soon saw was not an issue. Lowercase nu in Greek is ν, which looks like a V, and is thus similar to your upsilon with its relatively small tail, but uppercase nu is Ν which, of course, looks just like a Latin N.

    And since lowercase upsilon is υ, without the mid-stroke, the change of form you are proposing is not applicable there.

    Thus, from an outsider's perspective, no issue arises that your new form of capital upsilon will make any two letters confusingly similar. But no doubt these considerations are things you already checked at the start.

    As well, if the two-stroke shape is used in Greek handwriting, and a faux-ancient typeface with one is found acceptable by native Greek restaurateurs, as noted in posts above, that should give you confidence.

    In addition, I did some searching, and found this:


    While the shape is different, a kind of two-stroke upsilon is shown in that chart to have been quite common in ancient times.
  • John HudsonJohn Hudson Posts: 1,489
    A two-stroke Upsilon exists in various forms and times. It just never has a hook on the bottom.
  • Adam JagoszAdam Jagosz Posts: 233
    edited November 8
    Thank you for your insight, John [Hudson :smiley:]. Indeed not only this dark join, but also other ones, like in /M, are not very successful. I brought this on myself by deciding to design a bold condensed low-contrast face. But I like the challenge. I tried to address this issue by working on the join some more. Of course it was not dramatically improved. Below are revised U, Upsilon and Cyrillic U, all descending. I'll be glad to hear your opinions on this one, no obtrusive hooks.
    I'll provide an alternate cut with flush variants for the use cases where the quirky descenders might be a deal-breaker:
    By way of explanation. I had one minor reason behind introducing the descender, apart from changing the mood: by shifting the arms down, it reduces the whitespace below them and around the letter, thus (I hope) helping preserve a bit more even color.
    I made a pass at a hooked variant (which sits in the Upsilonhook slot), but the result makes for the opposite of what I just mentioned and does not sit too well among the otherwise tightly packed letters.
    That's an interesting chart, John [Savard :smile:]. I see what I suspected: Upsilon has some record of looking like a V, which might be where V came from. I had this in mind while designing my weird Upsilon (my subconcious train of thought: all right, Upsilon today reads “ee”, it is transliterated “Y”, but it is basically a “U”, and “U” once was equivalent to “V”... let's make it look like a “V” with a strange disease stuck to the bottom, like the bottom hook on Textura “x”...) About confusion with nu, I though of it especially now that I removed the hook. Hopefully, it will not be an issue, like you said.
    Thanks, Ray! If I were too keep the hook, I might have to introduce one in lambda for consistence... or otherwise. Below is a pdf with the full image of what we're dealing with.
    I wasn't going for Cyrillic air, Hrant, though that is an interesting goal on its own. I see the ‘Cyrillization’ tag gave that impression. I put it in there because... okay, my explanation of the process above is bogus afterthought, in fact I just had this lazy/crazy idea and pasted reference to У into the Upsilon slot and just assumed “I'll ask the Typedrawers later”.
  • Hey Adam! :) some comments if you want to consider them, mostly regarding construction: 


    - I vote for the second /Upsilon. First one looks (don’t get me wrong :# ) just wrong. 
    - Third one looks very old-ish, especially for this context. You can indeed find it in early 20th century specimens, but not so often anymore, or not so calligraphic. You could try a more subtle approach. (Photo below from one of my specimens: Deberny & Cie, Le Livret Typographique, Caracteres Grecs, Paris, ~1910-1914). 





    Lowercase:

    - /tonos being so horizontal, might cause issues for /dieresistonos, since your design is so condensed. Maybe go with a more vertical one like the one you use for UC.
    - /zeta & /xi seem to be very-very narrow, I would try to make them more equal to /epsilon. Revisit /finalsigma when you have a solution; these three share similar design language.
    - this construction of /gamma, with the open loop, is seen only in handwriting. I suggest you try a different approach; this will look (amateurishly) wrong
    - you could use elements from /chi (in-stroke construction) for the ascender of /lambda and maybe on /nu and /gamma. 
     - UC /Theta and LC /theta are not distinguished enough. Usually a not-connected horizontal bar on /Theta does the trick. 

    Hope these comments help! :)

  • Adam JagoszAdam Jagosz Posts: 233
    edited November 11
    Hi Emilios! Thank you for joining the thread and thank you so much for the detailed feedback!
    I hope I get you right on /Upsilon. I slept on it and in the morning I realized it's probably like I tried to get fancy about Latin /U and this came out :sweat_smile:
    The default will for sure be the straight normal glyph. How about a two-stroke flush alternate? Not that it makes much sense — if anything, it only introduces excess whitespace. But maybe adds a touch of informality? I also tried out John Hudson's suggestion, bottom line. How's that working?

    Since I already had a critique thread for this project, and just wanted to ask a general question about /Upsilon here (using my project as an excuse and example), I responded to the rest of the issues here: http://typedrawers.com/discussion/2825/rywalka-a-condensed-sans-explored-and-extended#latest
Sign In or Register to comment.