Revival - the latest

My Garamond effort

Stempel Garamond

Finaly I’ve digitized the lower cases from the tracings I talked about in my previous “revival” post. I’ve tried to remain as close as possible to the original drawings but there where some major mistakes in my drawings and I had to redraw some of the shapes from scratch. You can compare for yourself and tell me if you think this is a good revival. Don’t be too tough on me as this is my first ever attempt at type design and there is still loads to be done to improve it.

Appart from getting used to the slippery bezier curves I had the pleasure to find out all about the “metrics window”. The metrics window is where you tell every letter the exact space within which it will live. That means that at the same time you define the space between the letters. A good way is to start with the ‘n’s and the the ‘o’s. Make a line of ‘n’s (nnnnnnnnnnn) and space them so that there is opticaly the same amount of white between the letters and inside the letters. Then do the same with the ‘o’s. Now you can start spacing all the others by placing every letters between the ‘n’s and the ‘o’s. So if you start with ‘a’, you will look at: nananoaoao. Then you go to ‘b’ and: nbnbnbobobo. You must go through all the letters and you should get evenly spaced characters.

Tracing

The project that is taking most of my time is the “revival” of the Stempel Garamond. I have scanned a few pages of the book and I’ve tried to quickly digitize the letters in Fontlab but the results were disappointing. I was for a long time confused as how to transfer the shapes from the printed pages into vectors in Fontlab. The shapes on the printed pages are very rough and I must somehow redefine the outlines of the typeface. This is rather difficult since most of the details of the glyphs have been completely lost by the effect of ink trapping and the porosity of the paper.

I have blown up every letters to exactly 4 cm in height (x-height + ascender or descender) This I’ve read is the ideal size for sketching. I have done my own test and I have to admit that it’s a good size although it is quite hard to get sharp details at that size. I have traced all the lower cases quite roughly but I am already making some changes by ironing out the most prominent irregularities. I don’t know if at this stage I should perfect the drawings a little more or going straight for digitization? At this stage the glyphs are looking a little bold (or maybe semi-bold). Since I don’t know much about Fontlab I don’t know if it’s easier to thin the glyphs down in the program or on paper.

One might wonder why I am tracing all the letters and not just the ones from which I could generate the others such as the “o” and the “n”. If I traced the “o” and the “n” I could easily generate “p”, “d”, “,b”, “i”, “j”, “h”, “m”, “l”, “c”, “e”, “r” and “u”. But the point of this exercise is to make a revival and that means getting the flavor of the typeface I am reviving and not finding the quickest way to end up with a digitized typeface. If I generated most of the shapes from the “o” and the “n” I would end up with a very different typeface in flavor that that of the original. This happens because the idiosyncrasies that might be present in the “o” and the “n” will be repeated over the whole typeface. These idiosyncrasies might be the result of errors of interpretation of the original shapes.

Revival or survival

My new assignment is a revival from a letter-pressed book I bought from a second-hand market. The type-face is Stempel Garamond and I am going to digitize it in Fontlab. I guess there are already plenty of garamond out there and Stemple garamond has already been digitized. But this is not a commercial venture and it should only be looked at as an exercise to learn how letter-shapes work by mean of research and comparisons. I will also use Fontlab for the first time and I will undoubtedly learn an awful lot about bezier curves, kerning, metrics and all the rest.

The first part of the assignment was to identify the type-face. It was Stempel garamond but at first I thought it was Sabon. Only the book was printed in the 30’s and Sabon was only designed a few decades later. Anyway, I started looking for books about Stampel garamond and I could gather from “Anatomy of a Typeface” that Stempel garamond was adapted from a specimen sheet of garamond types printed in 1592 in germany by Konrad Berner (Egenolff-Berner). After much confusion in the library of Meermano Museum in Den Haag, I got more lucky searching the net for good quality scans of the specimen sheet.

I compared all sizes of the specimen sheet to the 12pt Stempel garamond from the book and unsurprisingly the “cicero” matches almost perfectly to the Stempel garamond. Some of the garamond types from the specimen are a little wider and the counter bigger and more open. The “m” and the “o” is definitely wider than the Stempel and the upper counter of the “a” is more open than the Stempel. The ascenders and descenders are longer than the Stempel and interestingly they vary hugely from size to size. You only have to compare the Canon “d” with the Petit Canon “d”. Some of the serifs are strikingly different like the upper left serif of the “p”. It is a little bit higher in the garamond specimen sheet. The lower right serif of the “u” is longer and no so angled as that of the Stempel. The “g” is more squished and the counters more even in the Stempel.

I still have some ground to cover with this comparison. In the end I hope to be able to identify the kind of decisions taken by the designer at Stempel. Hopefuly, this will help me in taking decisions when digitizing the Stempel. I could even use some shapes from the garamond sheet in order to be closer to the original garamond design. But I already understand that this is a polemical subject in the type world.