Reviving an Art Deco Font

Options
Howdy Y'all! Over the past year I have been working on a display font inspired by the title lettering of a series of murder mystery novels from the 1920s. It is the second full font I have designed, and is tentatively called "Philo"; however if anyone has a better name, I'm all ears. The original novels provided the basis for most of the capital letters, but the lowercase, numbers and punctuation were extrapolated from there. This typeface will ultimately include 2 styles: a "title" style and "subtitle" style, but I would love your feedback and sage counsel on this initial title style before completing work on the second. Thank you!





Comments

  • Thomas Phinney
    Options
    Do you want to reproduce every idiosyncrasy of the original book-cover lettering?

    The original lettering had that capital “O” with a major top overshoot and an even more exaggerated bottom overshoot, almost like a bit of a descender. I would be curious as to how well it would work if you dialed back that bottom overshoot, making it only as much as the top. Which is still substantial, just a bit less.

    I would try a much smaller counter on the lowercase “e”; the current design tries to preserve the design theme of diagonal counters from the cap O, but at the expense of a broader theme of small counters. I would also make the counter of the ”a” ~ square by moving down that middle bit.

    Lowercase v w are a bit too wide. Perfectly reasonable to consider giving them a wider angle than their cap counterparts, but I think it has gone too far.

    The diamond tittles of i and j can be noticeably bigger. Especially if you lower the x-height...

    Which is my big question. Why is the x-height so huge? Most of the contemporary work from the 1920s and 1930s that deviated from “normal” x-height, went in the opposite direction of small x-heights. To me, the big x-height makes it feel more like a 1970s mock-revival of a 1930s style, rather than something more genuinely of that period. If you want flavor in the x-height, I would try going small. Average is also fine, if you like. But extra-large x-height feels weird to me, in this context.
  • Craig Eliason
    Options
    Feels to me like it can't decide if curves are allowed or not.
  • Andreas Stötzner
    Options
    Droping the oversize of O and Q will be beneficial for the font’s credibiliy. The bottom of J is too heavy.
    the @ and € sign …  ?
    the counter of ¶ wants to be wider.
    – Overall, a rather nice exercise.
  • Ray Larabie
    Ray Larabie Posts: 1,387
    Options
    Dots on the i and j=cool. Period, exclamation, question etc.=not cool. Diamond dots and a diamond comma are the way to go here. You don't need to be so conservative with your punctuation with art deco fonts. The @ will exclusively be used for email addresses in a display typeface, so you can push the boundaries. A Letter Gothic style simplified @ form usually works better in these cases. Numerals widths could use some visual balance.
  • ishtar van looy
    Options
    I like the vibe! Feels like some more similarity is needed between /p & /q. Specifically, it's the circular /q that's throwing me off a bit, especially with the very sharp angle in the /p.
  • Nick Shinn
    Nick Shinn Posts: 2,158
    Options
    Great job on the lower case, Ben!
    Ditto Ray on the period.
    The Euro could do with more heft.

    Favourite Philo Vance: Basil Rathbone in The Bishop Murder Case.
  • K Pease
    K Pease Posts: 182
    Options
    The vertical of the question mark is chonky. It could taper like the bang, but in this style I think it would be better not to be a vertical at all. There could be a small zag to the right or it could just end at the bowl.
    I do not think /p and /q need to be similar at all, but there is something to be said for /q being more different from /b/d/g. A hexagonal /q might be most harmonious, making an /o/q relationship that reflects the /O/Q relationship.
  • Ben Noe
    Options
    Thank you everyone for your feedback! These are some solid suggestions, and I've got a lot of work to do. @Thomas Phinney your advice to lower the x-height is well taken... however, given the heavy weight of this typeface I find that lowering the x-height on characters like /s and /a leads to congestion (even after reducing the counter of /a). The usual remedies of reducing stroke width and changing the angle of /s spine just seem to throw off harmony with other characters. Do you have any guidance on how to solve the x-height problem without creating more problems? Also, by lowering the x-height, the overall proportions of the lowercase become wider. Is this ok? Am I going about it all wrong?
  • Thomas Phinney
    Options
    If you lower the x-height, you need to reduce the letters in both X and Y directions; just taking the top and moving it down will make them too wide. You could shrink and then add weight back, or make a copy and scale it down just to see what the proportions should be, before manually adjusting.

    In a more conventional design, and at text sizes instead of for display, a smaller x-height might also need slightly wider (proportionally speaking, not absolute) glyphs—but not so much with this typeface. IMO, anyway.

    That said, the smaller the x-height, the more difference makes sense between cap stroke thickness and lowercase stroke thickness. The usual difference is something like 4-10%, probably 6-8% is reasonable here, if you have a more conventional x-height.
  • Matthew Smith
    Options
    Alternatively, if you’re going to reduce the x-height but like the overall density/color of the lowercase, you can just scale down the lowercase, and then adjust the weight of the capitals to match.
  • Thomas Phinney
    Options
    Agree with Matthew—but I figure the cap weight is probably what you want to keep, as the caps are based on an existing design.
  • Cory Maylett
    Options
    In addition to what others have mentioned, I would likely take the leg from the K and duplicate it on the R. The b, d, and q all match as mirror images of each other. However, the p is different. This isn't necessarily a problem, but to me comes across as an odd inconsistency.