Archive through August 27, 2003

hrant's picture

I think I've found some good evidence for my position on trapping of metal fonts in general, and Tetterode's Pascal* in particular.


It's from "Dutch typefounders' specimens" (Lane and Lommen). The photo is about 11 years prior to the production of Pascal, but I suspect not much had changed in the meantime.

I think one of those young'uns in the first row simply messed up that "b".


gerald_giampa's picture


"1) You couldn't make it a touch wider on Monotype..."


How much is a touch?

Gerald Giampa
Lanston Type Company

gerald_giampa's picture


When was Sabon developed?

Gerald Giampa

kentlew's picture

Gerald --

According to my references, Sabon was developed in the mid 1960s. Bringhurst says Stempel released the foundry version in 1964, and Linotype and Monotype released machine-composition versions in 1967. Other accounts simply date the design to 1966 or 1967.

Thanks for quoting Mike's letter. This jibes perfectly with what Matthew and I discussed.

Hrant --

I didn't quite follow your second point about Sabon 'a' above. You're right in point 1, he couldn't make it a "touch" wider. Examining Sabon more closely, it appears to be the result of the combined constraints of
Monotype and Linotype together.

The Sabon 'a' is allotted to 9 units, which is wide compared with the 'e', for instance, which is on 8 units. If Sabon had been designed solely for Monotype, the roman 'a' could have been put on 8 units, which would have given a more traditionally Garamondesque proportion; while the italic 'a' could have been placed on 9 units, so as not to be too narrow.

If instead, Sabon had been designed solely for the Linotype, an intermediate width could have been designated that struck a compromise between the duplexed 'a's, with the roman maybe just a tad wider than ideal and the italic just a bit narrow.

But Tschichold had to adapt to both constraints -- duplexing for Lino, unitizing for Mono. So he made the choice to make the two 'a's fit the 9 units (which yields a noticeably wide roman 'a') instead of cramping the italic in the 8-unit slot.

NB: This is just my analysis/reconstruction/conjecture; I don't have an actual account of Tschichold's process.

-- Kent.

kentlew's picture

Okay, I spent time I probably shouldn't have creating illustrations. Hopefully this will be edifying.

Here is a quick-and-dirty adaptation to show what Sabon might have looked like if Tschichold had chosen to assign the 'a' to 8 units instead of 9:

Sabon adjusted

The narrower roman 'a' looks elegant, but the corresponding, duplexed italic 'a' looks obviously cramped. Since the counter of the italic relates more closely to other characters in the italic, the discrepancy is quite noticeable. The roman 'a' relates less obviously to any other character, so any deviation from ideal proportions is less noticeable.

A narrower italic 'a' is perhaps not especially objectionable in and of itself, and this design might have been accommodated by adjusting the 'b d p q o' etc. *But* because of duplexing these adjustments would not have been possible to reconcile across both faces.

So, Tschichold had to make the best compromises he could, given the combined constraints. Here is the original to compare with:

Sabon original

After all is said and done, I think Tschichold's results are quite understandable and commendable under the circumstances.

-- K.

anonymous's picture

It would be a mistake for any persons reading these posts or posting their views here to assume that any good pantograph is a second-choice method of cutting a type. Neither is the pantograph a substandard method of making a type. A good pantograph is a precise, marvellous piece of engineering, that properly handled will give you an accurate translation of your original drawing, just like a good digitizing tool will.

As one who has just spent all of Sunday and all of today so far, attempting to re-cut a punch for a damaged lower case f, (normally this takes a few two to three hours) I can assure you that it takes care, dedication and determination to cut a letter on a pantograph, just as it does by the hands of a punchcutter.

After cutting four type designs by hand,I now choose to use the pantograph. Granted, one must make a pattern. Once the pattern is completed it can within some size constraints be used to cut other sizes of punches with the proper scaling. In addition, other weights can be achieved with the same pattern. I do not suggest that a person can make a book and a bold weight, rather, it is possible to compenstae for weights between, say, 12 and 16 point.

I use the pantograph for the same reason that large foundries turned to it in past decades: even with the task of hand cutting master patterns, and then the metal working pattern, I can complete a typeface in one size much quicker than I could if I attempted to hand-cut it.
I admire and respect those who can and do cut punches by hand. Dan Carr of Golgonooza Letter Foundry, and Stan Nelson of the Smithsonian Inst. both do this. I am told that they in return respect my work, which is pantographed.

It is possible to hand-nick out tight corners with a graver once the punch is cut, and I have done this to every patrix and punch I have cut over the past 25 years.

If you are ever fortunate enought to be in the position where you can afford to have a type cut for your private press, take a tip from me: either way is not to be slagged off as inferior. In the end, a good design, well cut and well fitted will be a joy to print with. Incidentally, not a soul has mentioned a thing about a good fitting.

The elitist attitudes, among letterpress printers at least, is not limited to punches. there are those who will not use type that is not cast on a Barth foundry caster, and blow off Monotype as though is were "unclean". In my world, any type that is well designed and is "type high" is worth printing with, and that's that. Hell, there's nothing wrong with the Linotype machine either!

Look--I'm not a crank, and I won't get into a lot of fuss about type. Type is a joyful thing. Some people who post to this forum give me the feeling they just ain't having a good time unless the are calling somebody else "stupid".
Like the sign on my shop door says: "Be nice or go away".

Come visit my foundry/shop if you are coming to Vancouver. See some type being made. Like all Canadians, I have "unlimited beer and wine".

Jim Rimmer

anonymous's picture


You've posed some good points. Let me see if I can answer them.

The adjustments that one can make on a pantograph are: Weight Adjustment, which is done by alteration the size relationship between the width of the cutter and the stylus which is used to trace the pattern. If a cutter is a fixed size, but the stylus which is used for one point size is .100" (or one hundredth thousands of an inch), you will get a certain width of cut removed from the punch blank.
If you use the same cutter again, but decrease the size of the stylus by say 20%, the amount of metal removed will be increased by a relative amount. Since you are cutting the neck of the punch to leave the face, the face will be thinner by that amount. If the scale of reduction is large (like 20 to 1) the relative amount will be very slight, so that for every adjustment for weight there is a fine control over the slimming effect on the face. Sorry this sounds fiddly.

The other adjustment is for condensing, extending and obliqueing. Obliqueing would be reserveved for sans. This is done by making the vertical and horizontal settings for scale on the machine of uneven amounts. One way pulls the letter to extend it. The other way pulls the letter in height to condense it. Obluqueing is a setting where only one of the four adjstments are thrown "off equal".

You are right, and I would never argue otherwise. Nicking out corners IS because of a limitation of the pantograph.
The proper employment of the pantograph requires the hand finishing steps of cleaning up corners and pinches. There have been many punches cut by Monotype (compositioin sizes) where hand finsihing of corners etc was not undertaken. The resulting round in the corners was very small; two-one-thusandths of an inch. A few days ago I printed some pieces which I had set in 10 pt Garamont Small Caps. If you look at the type under a stereo scope, you can see a slight softening of the corners. However when I printed it, it looked very sharp and precise; no soft look, but I'm a fussy pressman. If somebody really dumped the ink to it, it would look bad whatever had been done to the punches.
If you consider that the pantograph saves the lion's share of the cutting and the time, it is a small price to pay to need to finish the punches/patrices by hand. I don't know for certain, but I assume that that is what the great foundries of Europe did in their making of punches. You will have read that there was a great outcry from punchcutters in the early part of the century when the pantograph was finding its way into the craft. The same sort of resistance was met when the Computer hit the graphic design business. Not preaching. Let me sum it up: the pantograph saves the time, and does a perfect job. The corners need a human hand. Remember that Arion Press and Pennyroyal Press and most of the fine presses of the UK use Monotype. Nobody could exist if they had to have Stempel type. You need volume.

>Concerning the use of Monotype fonts etc>

1. The quality is variable only if a bum is casting the type. It was no more a DIY deal than the type being made at ATF. You need a good operator to make type. If it were a DIY affair the caster would only run for about five minutes before the DIYer burned his butt off. Certainly the metal delivery mechanism of the Monotype is not as robust as that on a foundry Pivotal caster or a barth Caster. However, if properly adjusted and maintained the Monotype (Comp*, Orphan Annie, Giant, or Super will deliver very good type that will last a long time. * Type from a Comp caster was never intended to last as foundry type. It was thought of as transient in the way that Lino and Intertype was . . . use it and melt it down. No distribution, and no lost labor. People who buy Composition size fonts from existing foundries in with the idea that they are buying permanent foundry type need to be aware that they are getting soft metal type. This is true of all faces from 6 to 12 point, and a few faces where the foundry has large comp mats up to 24 point. It's disappointing to have that happen, and it gives Monotype a bad name unjustly.

2. On all Monotype machines but the comp caster it is possible to use the hardest of foundry metals. I do this all the time.

3. You may be confusing the italic restrictions with Linotype and Intertype, where the Italic "a" would ride along on the same matrix as the roman "a", and had to be the same width. Monotype had no restrictions artistic restrictions regarding its italics, so in that matter it had the same freedom as foundry type. This applied to the composition as well as the display sizes.

Of course Linotype could not be handset. I sing its praises only in that it could compose very fine books, at least in my opinion, and it prints beautifully. Liontype did overcome the lack of kerning capabilties by making a suite of logotype (kerning pair) matrices of some of ists more artistically based faces. If you bought them, you could set good looking type.

I think Photopolymer has been one of the best things to hit letterpress since its decline as trade in the fifties. Photopolymer has brought the letterpress process into fashion again, and has made it possible for printers to satisfy exactly the wishes of designers. But look--I have been doing this almost continually for fifty-five years. I'm a REALLY old fart, and Photopolymer doesn't do it for me. There is just something about the actual handling of type that makes me feel good. Still.

I do not aim the word "elitist" at you. I just cringe when I get in roomful of likewise old letterpress farts who pull out all their theories about how only a blacksmith (blacksmith is one of their favorite words because its easy to pronounce) would use other than Barth-cast type, and only a blacksmith would use a certain kind of paper. It goes on and on. And without exception they are usually awful printers.

OK, this got long. It's not easy to cover things like this without taking some time; and I am sure this drivel will be of absolutely no interest to most people who read the form . . . but YOU are interested so I responded.

Jim Rimmer

anonymous's picture

Gerald Lange

You're right, Pennyroyal did do their big bible with photopolymer, so in their case I should have said "have done" regarding Monotype setting. I can appreciate that if a person has to buy Monotype composition, it's a big expense. Arion Press of course have all of M&H Type at their disposal, so I am surprised to hear that they make use of photopolymer. It was my understanding that Arion Press Monotyped its Bible in Romulus.

Do you pay a lot less per page if you make your own cuts in place of having Monotype pages set?

I know I am a spoiled guy, having three casters for my own use, but without them I wouldn't be able to produce The Adventures of Tom Sawyer. In any event I cling to the old ways because it makes me happy. Too many things in the world change. This stays the same.

Jim Rimmer

anonymous's picture


An 18 unit system could be very flexible for a designer, because a particular letter (for instance an h) need not always be assigned a value of 9 units. If for artistic reasons the designer wanted the h to be wider or narrower, that letter could be assigned to another unit row. This did not throw off the positioning as far as the caster was concerned, because the keybuttons on the keyboard could be linked up to accommodate the h's new position by the use of a different set of keybars. The keybars wa a sort of hardware/software that linked up the keybutton with the new position in the matrix case.

Another thing that made the Monotype flexible for unit values was the direct result of FW Goudy's pushing for some expansion of the unit value system. Lanston's resident designer/genius Sol Hess, was responsible for re-engineering this set of new unit values. The first example as far as I know was Goudy's Kennerley. Gerald Giampa would be the one to confirm this since he for many years held the Monotype material and matrix making equipment at his large printing plant/foundry in Vancouver. Kennerley was based on a 20 unit system. There was also a 15 unit system. British Monotype's Univers is, I believe, 15 unit. As Gerald Giampa has noted it was also possible for a keyboard operator to add or subtract units from a particular letter. Of course this is not to do with the design process, but it was a hand feature.

On the matter of italic constraints, there were absolutely no limitations put upon the artistic freedom of italic, other than that the designer had to work within a grid of fixed units. With the ability to pick and choose what amount of units a designer wanted a character to occupy, there was lots of room to move. In addition Monotype characters are made to kern in the matrix case. When I made my own matrixcase of large comp (18 point Hannibal) I had to stick with the precise positions allotted to me by the use of a British Monotype matrixcase layout (and caster wedge which makes these widths in the casting) for British Monotype Garamond, I had no trouble in altering my own characters to fit the spots with an appropriate amount of space each side of the letters. I had the advantage of being able to test these fittings and kernings since I digitized a computer version of the intended hot metal type that I could test and re-test as work progressed, without having to cut a single master patrix. I think Lanston would have loved to have the Computer back in those days.

I am aware that Linotype hot metal had no hard and fast grid of units for a new typeface, but it is my understanding that they did establish grids when they entered the field of making fonts for new technology. I am not informed completely on this, but I remember Rod MacDonald, who did a recreation of Cartier (Cartier Book) mentioned in a previous writing about Carl Dair, that when he visited the Linotype company with the proposal that they do the production on his drawings, he was handed a set of grids to work by. Perhaps Rod could shed some light on this.

One has feel sorry for the poor designers who had to design a typeface for a typewriter back then. Talk about unit constraints!

Jim Rimmer

anonymous's picture

For a few years beginning about 1956, I operated Linotype machines in and around Vancouver. My first stint was on the Vancouver Province, a daily paper. I was fresh out of my apprenticeship, and I have to admit I was not the best Linotype operator to ever sit at the machine. I can recall that in the first two weeks of my time on the Linotype at the Province, I gave the proofroom something to do. The Linotype and the Intertype are both really marvelous things to see in action. Although they were intended for speed and efficiency, and the Monotype was a more quality-based system, the linecastrs were and are wonderful. I am sorry to have sold my rare Pygmy Linotype.

Jim Rimmer

hrant's picture

Mostly for the purposes of clarifying a post I've made to
the ATypI discussion list where I linked to this thread:


hrant's picture

One more...


hrant's picture

I know it's been a while, but the "tetterode.jpg" image in the original post having gone missing is enough excuse to potentially give this thread new life. It's a great shame that we've lost both Gerald G. and Jim since.

Without further ado:


Syndicate content Syndicate content