Reinette, having fun with the Renner Type

ivan louette

A trial with an Art nouveau design : the Renner Type from Bruce Foundry, 1899 found in De Vinne catalogue 1914.

Christian Thalmann

The original is less skinny than your rendition, which helps it work better as a text font. For instance, the internal stroke of the /a/ is laser-thin, and the /f/ is too light as a whole. (I'm not sure how historically accurate you plan to be, but if it were me, I'd draw a wider default /f/ and keep the original as a contextual alternate before high letters. But that's just a matter of taste I guess.)

I also note that your counters in /b/d/p/q/ are very pointy toward the stem, which makes them look agitated, whereas the original looks more or less elliptical.

Craig Eliason

Apostrophe looks too dark and maybe too severe. Agree that overall this feels very slight—what’s the intended use/size? Agree that the middle part of /a tapers to too fine a point before disappearing. Pointy serifs at x-height like /u’s might need some blunting and/or more overshoot—those letters look short relative to neighbors with round tops. 

Charming model and you’re doing good work with it! I love faces like this that are quite functional even when bursting with character. 

ivan louette

Thanks for your comments. Of course you are right about every point. I began last sunday night and I still hesitate about what I will keep or change. In this specific case contextual alternates seems a good solution for what looks outdated or even not efficiently designed (for example I am not convinced by some capitals).

André G. Isaak

It looks nice, though the thread title made me think this would be about Futura.

ivan louette

I have tried a less skinny version and corrected the /a and /bdpq problems. Tomorrow I will try some changes to /f and other ones which share the same to narrowly curvated ends (/j /y).

I must change diacritics too and improve punctuation.

Andreas Stötzner

there are Art Nouveau typefaces which deserve resurrection. This one doesn’t.

ivan louette

@Andreas Stötzner Thanks for your comment. It's your point of view and I do respect it. Did you see the originals I pointed in my first post ? Here are the pages (223 to227) of De Vinne catalogue for your additional information. Except George Auriol types it seems there are few Art nouveau text fonts while many are nice for titling (however I can miss some of them). But that's only one of the things which lead me to play with it. The main one is that I falled in love with it at the first time I saw it some years ago. It has also the taste of an unachieved project interesting to investigate.

Andreas Stötzner

well, my notion was very harsh, there is always something also in a crude thing. If you love it – do it. The special idea of such a design is one matter, the quality of execution another. With such a peculiar design, one is in danger to stumble over many things. A closer comparison of the scan and your samples reveiles: the Renner type is much more Jensonian than your design (maybe that’s what you want); but: your lower ending of e and c are not quite what the original shows (not better than the original), your s is too small whereas in the original it is not (only a quick pick at 1st sight). You have to be terribly cautious when you want to transport the interesting ideosynchrasies of Renner but avoid slavishly copying the originals pitfalls. Have you had btw, ever a closer look at Jenson? – Good luck with it.

Nick Shinn

What’s your method, Ivan?

ivan louette

@Andreas Stötzner The quality of execution depends on time. Sometimes the first step in designing is fairly fast but it needs tons of small adjustments afterwards. there is nothing terrible in that  Then the eye must be trained day after day. In this new step I address some of your helpful remarks and drastically change the spacings to match the original ones. I am also trying an alternative to /f top end (and did the same for /j and /y).

ivan louette

@Nick Shinn My method is artisanal and composite. It's widely based on the vectors copy and paste facilities between Inkscape and FontForge on Linux (it works in the two directions).

1° finding an image (a portion of text is the best) with a fairly good resolution of the font ; 2° enlarging it in Inkscape up to the font size of 750 pts, which correspond to 1000 units in FontForge by default (I made a canvas in Inkscape with a precisely positioned base line) ; 3° manually drawing every character in Inkscape (no autotrace used) after searching how to rationally repeat different parts of the characters ; 4° copying and pasting the characters to FF with a mark (a simple quadrangle) drawn at the top left of the character area for a constant vertical positioning.

After that I can type characters at 750 pts in a blank document in Inkscape and convert them to shapes to use Inkscape drawing tools to improve my drawing, or I can also use FF drawing tools for what is unavailable in Inkscape. However I use Inkscape a lot because I know its tools fairly well after many years of practice (and some contribution to its development, even if not in the vector drawing area).
And for spacing or tiny other positioning adjustments I can type text on top of the original bitmap of the font in Inkscape.

ivan louette

Here are samples of the two versions of the font for today : the first one with its original /f, /j, /w, /y and the second with design changes for these letters inspired by the top end of /r and the shape of the original /v for the new /w. The characters and lines spacings are the original ones. In a few cases kerning pairs will be needed.

Paul Miller

The second one is better, the /w and /v are more harmonious, the /f doesn't look cramped at the top and the /j doesn't look cramped at the bottom.

ivan louette

@Paul Miller Thanks for your advice. The first one is closer to the original. I was always intricated about this so classic and anachronic /w within more stylish other glyphs. The second one is my own trial to put homogeneïty inside the texture. But I like also the "kitsch" feeling of the first one.

Craig Eliason

I notice from your catalog link that the foundry reverted the /v to a more conventional form not only for the smaller roman cut but also for the bold. The curvy /v is lovely but its "distractingness" needs to be managed (and the same true of course for your added /w). I might increase the gap between the terminal of the curve and the left diagonal to make sure the /v doesn't optically close up. It also looks small (as does the original)--that's maybe inevitable from the form, but you might try more overshoot at the bottom.

Relaxing the /f is a good move IMO, but I wonder if your new /j and /y go too far in the generic direction. 

I think the bottom of /p's bowl gets too light.

ivan louette

@Craig Eliason Thanks for your comments ! You are absolutely right about curvy /v, /w and /p. I tried to address that. And some kerning has been added. Testing some different combinations or design changes for /j and /y seems interesting too ; I will do it.

Claudio Piccinini

ivan louette said:

@Andreas Stötzner The quality of execution depends on time. Sometimes the first step in designing is fairly fast but it needs tons of small adjustments afterwards. there is nothing terrible in that  Then the eye must be trained day after day.

This is the main reason for which I am grateful that I did not started to get accustomed to bezier curves before becoming familiar with curves. While creating curves with bezier (or similar spline tools) offers specific room for experimentation, if you are trying to "understand" a curve nothing beats observation and manual hand sketching.

As far as your project goes, I think the best thing would be to carefully redraw the original curves, and then – and only then – proceed with your own experimentations/changes/new typeface.

Thanks for the link to the De Vinne catalog. I had to browse it yet in full, and it has many important things relevant for my current digital version of De Vinne.

John Savard

Perhaps the criticisms of the typeface as being too light had some validity. When I looked at the print sample of the first version, it appeared to be a revival of the typeface noted as "Elzevir or Cadmus", on the facing page of the DeVinne catalog link, rather than of the Renner type.