Eau de Garamond — a sans distilled from the essence of Garamond

11011121315

Comments

  • I feel like you're trying to turn Ysabeau into something different than it is. It's a low-contrast sans; in fact, it started out completely monolinear in its earliest beginnings. I do have the (very, very) long-term goal of complementing it with a contrast axis, but it's behind a lot of more interesting things on my to-do list (such as IPA support and a condensed, high-x headline cut).
    Here's what /H/ looks like with a thinner bar. It feels to frail to me, especially for the extreme end of the heavy weights. I don't think /E/ is a good indicator for how light the horizontals should be in a heavy sans; they're necessarily optically compensated for density.
    Here's /M/ and /W/ together:

    While I think they mix well overall, up close the /W/ does have lighter joints than /M/. If anthing, though, I find the /M/ more attractive and might reduce the optical compensation a bit in the /W/. It definitely needs more contrast to match the /M/.

  • Definitely better:

  • Ori Ben-Dor
    Ori Ben-Dor Posts: 386
    edited February 2021
    I'm not trying to turn Ysabeau into something different, I'm just suggesting more consistent contrast. Some letters (E, M, U, new W...) demonstrate higher contrast than others (L, S, thicker-bar H...), in a way that, in my view, isn't justified by constraints, compensation needs, etc.

    To fix that inconsistency while maintaining as much of the current look and feel as possible, I'm suggesting reducing contrast in the most contrasted letters and increasing contrast in the least contrasted ones (keeping the average, if you like).

    Thin-bar H takes it too far. I'd go for something in between.
    The same with new W. I think the twin joints are now too heavy (bear in mind that they're closer to each other than in M, which means it makes sense to make the W ones a bit lighter), and you've also lost much of the positive stylistic effect of the tapering outer strokes.
  • Christian Thalmann
    Christian Thalmann Posts: 1,988
    edited February 2021
    I tried splitting the difference on /H/ and still preferred the original.
    I did lighten the joints in /W/ a bit; good point about the joints being closer together. I don't miss the stylistic effect of the flaring, though; it always looked a bit too disco for my taste. I'd rather go in the other direction.
    As for /S/, you call for it to be thinned after I just fattened and widened it for two other commenters on the previous page. :grimace: (I just compared the current version to the slimmer one from before, and prefer the current.) I don't currently have the energy to play a game of Pleasing Everyone (whole family has a cold 🤒, including the two-year-old, who can't go to daycare because of that, and my father's in hospital for a different reason), so I'm going to save the issue for later.
  • You don't need to please me. I'm just offering my subjective view. It's perfectly okay for you to reject it. It's also perfectly okay to save it for later, of course.

    I'm not calling for S to be thinned as a whole, nor to be narrowed. I'm calling for just the thinnest parts to be thinned, in order to introduce more contrast and bring it closer to the average level of contrast. 

    As for the tapering, I think you should consider coming up with a global policy rather than addressing it on a case to case basis. All diagonals, as well as the verticals of N, should probably have a similar amount of tapering. The latest version of a whole UC set I'm seeing is from July 2018, and back then X, V, etc. had much more tapering than new M, W have now. As for the desired global amount of tapering, I understand why you don't want too much of it, but I wouldn't get rid of it completely. You've managed to keep the quality of humanistic elegancy found in Garamond thanks to such details as diagonal terminals, generous extenders, etc.; tapering diagonals have a similar effect.

    I hope the whole family gets better soon!
  • Adam Jagosz
    Adam Jagosz Posts: 689
    edited February 2021

    I hope the whole family gets better soon!
    Human family or font family? :smirk:
    I do think the H, E, L, M and all the rest need different contrast, at least in the darkest weight. My goal there would be to equalize color, not contrast in each individual letter. In some of the lighter weights though, the contrast might get unnecessarily different while no longer contributing to equal color as much as in the blackest weight. This might call for glyph masters (unless you've already got these) to actually equalize the contrast again.
  • My goal there would be to equalize color, not contrast in each individual letter.
    Yes, that's what I've been doing.
    I do have glyph-level intermediate masters (brace layers) to keep contrast from setting in too early when transiting away from the monolinear Hairline, but it mostly affects lowercase letters like /n/.
  • So Ysabeau is scheduled for her first ingestion into Google Fonts at the end of March. The plan is to stick to the weights from Hairline to Bold at first and then add the Black master once I finish drawing its Italic. I've been putting that particular job off for quite a while because it looks daunting, but I've finally started and am making enough headway to wonder whether I should still aim at finishing it by the March deadline. No doubt it's going to look a bit rushed then, but as the Black Italic, it's both less commonly used and more tolerant to a bit of idiosyncrasy than all the other masters...
    In any case, here's what I have so far. Does that work? Anything that sticks out to you as in need of work?


  • Definitely better:


    Oh oh.
    these two tilt leftwards.
  • Christian Thalmann
    Christian Thalmann Posts: 1,988
    edited February 2022
    My eye doesn't agree. But I tilted them a bit to the right, is that better?
    What about the Italics?
  • Craig Eliason
    Craig Eliason Posts: 1,440
    edited February 2022
    On the italics:
    Cap /Schwa’s look wide. Overshoot of the rounds in the caps, at least in the bold master, looks too much. Inner contour of bold /C looks wonky to me, making the left part of the letter look too upright. Bold /X may be light. Bold /S too? Both /B’s look narrow to me.
    Was gonna try to talk you out of the quaint alternate form of /p before I realized that’s a /thorn!
    I think the bold lowercase is especially appealing!
  • So Ysabeau is scheduled for her first ingestion into Google Fonts at the end of March. The plan is to stick to the weights from Hairline to Bold at first and then add the Black master once I finish drawing its Italic. I've been putting that particular job off for quite a while because it looks daunting, but I've finally started and am making enough headway to wonder whether I should still aim at finishing it by the March deadline. No doubt it's going to look a bit rushed then, but as the Black Italic, it's both less commonly used and more tolerant to a bit of idiosyncrasy than all the other masters...
    In any case, here's what I have so far. Does that work? Anything that sticks out to you as in need of work?


    Some of this may be down to rendering issues, but a couple of things caught my eye.

    Black /c, the base seems to have a bulge whilst the top seems a little flat.  Appreciate you're pulling the contour in at these points, but maybe the nodes could be shifted left and right respectively to compensate and help smooth the curves?

    Both bold and black /æ & /œ look very similar.

    Black /M, the right diagonal appears to get thicker towards the junction.

    Black /C looks relatively upright to me.

    Black /N diagonal gets heavier left to right (but maybe this is intentional, as it's quite pronounced).

    Black /S, I'm seeing a slight bulge at the base.

    Bold and black /U, the apex of the bowls seems a little too far to the right.

    Bold and black /a, the curve to stem transitions looks a little abrupt.

    Bold /r, the junction of stem and arm seems too thick.
  • Thanks! Better now?



  • Much better, in my view.  I much prefer the new /kgreenlandic too.  I had a note to mention this last time, but forgot.

    Maybe you've discussed the rationale earlier in the thread, but I'm wondering why no tittles?  I've seen tittles in images posted earlier for the roman; did you remove them from that too, or just for the italics?
  • I was going to address accents in one go, rather than singling out the tittle. Rest assured there will be tittles. 😉
  • Christian Thalmann
    Christian Thalmann Posts: 1,988
    edited February 2022
    How's this?

  • Looks great, though I'm seeing the tittle on the /i drifting a little to the right.  Same for the umlaut on the /ö
  • /q and /a feel a bit short, could probably use some overshoot.
  • Does /t/ profit from a slightly steeper stem slope to compensate for its bend?

  • Yes, that's a nice adjustment.
  • Trying out a suggestion for /g/ from the German forum. I think I prefer the original, though.



  • @Christian Thalmann, again, not sure if this is by design (or rendering), but right leg of the /n is jumping out at me and, when I compare it to other glyphs, I see it's appreciably thicker than all the other straight stems (except the /h*), including the /m and the corresponding left leg of the /u.  

    It also appears to be thicker than the widest parts of the bowls of the /o and /p, but not of the /a, /c, /b, /d or /q.

    * the /h is less noticeable because of the way it thins
  • You're right, the right stem of /n/ was a bit much. Better like this?

  • Better, but probably because I'm looking for it, I'm still seeing the /n right leg a few points wider than the /u left leg, but it's doesn't jump out at me now.
  • Well, unlike the /u/, it's vertically diminished both on the top and on the bottom, so it might just profit from that extra bit of heft...?
  • Craig Eliason
    Craig Eliason Posts: 1,440
    I think that terminal of /h is really hard to get to relate persuasively to the baseline. (But I'm repeating myself from years ago!)
    Does the alt /h with the /n-like outstroke still exist?
  • Christian Thalmann
    Christian Thalmann Posts: 1,988
    edited March 2022
    I believe the design has come a long way since last you opposed it! :grimace: But rest assured, the /h.ss02/ is alive and well! (Though remind me to generate the matching accented forms and ligatures...!)
    EDIT: Never mind the reminder!

  • in this sample the main stem of h is too thick in comparison with its right part and the stem of the f shapes.
    In both ligatures ff and ſſ the left mainstroke is weaker than the right one.
    And the crossbar to the h ought to be much lighter since it is a diacriticum, not a part of the letter; look how much thicker it is compared to the bar of the f’s.

    I think you really need to watch and look more carefully what’s happening.



  • Craig Eliason
    Craig Eliason Posts: 1,440

    I think you really need to watch and look more carefully what’s happening.
    This is a 15-page thread of a designer listening to constructive criticism and making subtle changes to a design over years. The implication I get from your comment that he's blind or careless to subtleties of design really feels out of place to me.