Archive through March 11, 2003

hrant's picture

> the particular pattern of ink spread, e.g. at junctions, is not necessarily something to be revived

Not to be revived, but often something to simply be incorporated, even these days. Certain areas of type design, like newspapers, phone books, photopolymer letterpress, small sizes* and third-world presses benefit as noticeably from trapping as they do from good kerning for example. Especially if the design in question calls for a "crispness" of character.

* In fact trapping isn't just about fighting gain, it's about fighting optical aberation too (which happens at small sizes, irrespective of gain).

The best designers do incorprate trapping (sometimes very subtly - like look at the lc "g" in Georgia), and sometimes its absense is completely unforgivable (like in Majoor's Telefont).

hhp

bieler's picture

I dragged out my Specimen Book of Linotype Faces (aka Big Red) and took a look at the Electra specimens. Revealed there, in 8, 10, 11, 12, and 14 point. The ink spread is held to an absolute minimum to allow the clarity of the letterforms to show. These are much "thinner" that the printed specimens show in this thread. Also the baseline disturbance is more prominent in the smaller sizes and is progressively adjusted thoughout the range. The 14 point baselines are hardly disturbed. The hanging or jumping lowercase characters mentioned previously seem to have been brought in line. These are quite vivid in the 8 point though. I'm suspecting this was a means for creating a certain legibility to the smaller sizes? It is fairly characteristic of many of the Linotype text offerings.

hrant's picture

Gerald, some scans would be really nice.

> The ink spread is held to an absolute minimum

Would WAD have approved (had a chance to approve) this or the work in "Emblems and Electra" more?

> the baseline disturbance is more prominent in the smaller sizes

This actually favors the "technical limitation" theory, as opposed to the "intentional design" one.

hhp

bieler's picture

Hrant

I pulled out the 1948 supplement to Big Red and there is a similar showing, even cleaner but with the inclusion of the Bold.

Also have the 1954 Distinction in Type Design specimen showings of the various Linotype text faces and the Electra is revealed at 48 and 96 point (partial fonts).

I'll bring these in for you to look at tomorrow. Big Red's a pain to lug around.

hrant's picture

I appreciate it!

> Big Red's a pain to lug around.

Yeah, and to think she was such a hottie before the revolution.

hhp

bieler's picture

Yeah, and to think she was such a hottie before the revolution.

Yeah, she was.

Gone a bit further with this. I set the digital Electra to the same size as the 96 point shown in the specimen. Ha! Near identical. There is very little structural change, and these are very subtle, in most cases a slight strengthening of the character! Stem widths are exact. In many cases, crossbars, a curved stroke appending to the main stem, secondary stem weights, etc., the metal version is a bit THINNER!!!

I'll bring these in as well.

hrant's picture

So either the digitboys took a scanner to the largest size at hand, or the photoboys took a camera to the same, hoping to get the "cleanest" transfer? If so: stoooopid. At least they could have started there but taken some global measurements at the smaller sizes to know how to compensate down from the "hi-res" source.

hhp

William Berkson's picture

Hrant you write "> The classic standard for beauty
No such thing, my friend!"

Ok, it's neoclassical. Francis Hutcheson, in the first book on aesthetics (1725) wrote that beauty is experienced when we are confronted with a "compound ratio of uniformity and variety." He influence the neoclassical movement.

When I did a google search on "unity, variety, balance", it came up with mainly pages from courses on design, so I'm not alone in thinking these concepts are useful.

As this is a thread on logic and intuition, I would suggest that questions such as 'is the design unified? is there enough variety? is it balanced?' are useful. They are a way logic can critique intuition and spur more creativity.

When I read between the lines of Bringhurst, I have the feeling he thinks that a lot of the digital versions of the classics were done without enough care or thought (under commercial pressures?) and as a result are a comedown from the orginals.

The discussion of Electra here seems to confirm that, big time.

hrant's picture

> neoclassical

Well, the Papazian school says that Beauty is first and foremost something mankind in general and any individual in particular can never really comprehend. It's not a recipe.

hhp

bieler's picture

"When I read between the lines of Bringhurst, I have the feeling he thinks that a lot of the digital versions of the classics were done without enough care or thought (under commercial pressures?) and as a result are a comedown from the orginals.

The discussion of Electra here seems to confirm that, big time."


Depends on what you choose to hear (or read) apparently. If a digital reproduction of Electra looked liked one of those poorly reproduced letterpress specimens, would you really be interested in it?

kentlew's picture

Gerald, you undoubtedly have a more discriminating eye than I do
with regard to the quality of letterpress impression. And, unfortunately,
I don't have a copy of Big Red (yet). To be clear, however, I do not base
my judgement solely on the one example of the Mergenthaler booklet.
It is based on a number of encounters with both fine and trade editions
on various papers. I am not suggesting that digital Electra be made to
look like a poorly reproduced letterpress specimen. I just wish it retained
some of the substance and vigor of the original.

Let me see if I can offer a counter example:

I feel similarly about the digital version of Caledonia (at least Linotype
chose to call it "New" Caledonia). I feel that it is too light and contrasty
compared to the original metal Linotype, especially under modern
printing conditions.

However, Matthew Carter created a digital version of Caledonia for
Time magazine in 1993. Now this puts some of the ink back on
the bones and reinstills vigor in the face. It is no historic mimic of the
letterpress specimen; neither is it an anorexic specter. As could be
expected of Carter, he has done a thoroughly modern job, but with an
eye for the quality and spirit of the original design. He recently did the
same thing with Linotype Monticello for the Princeton University Press.

I wish the same might be done with Electra. It would be a great service
to the book publishing field.

-- Kent.

hrant's picture

Speaking of Monticello/Oxford, is that a great "g" or what?

hhp

William Berkson's picture

Gerald, here is Bringhurst on Electra: "In their original Linotype form, Electra was the liveliest of the three [compared to Fairfield and Caledonia], though in digital form this is not necessarily so."

Tactful, but the implication is that the one who digitized Electra messed up.

I would much rather read a book in what you call 'poorly reproduced letterpress specimens' than the digital version any day.

I think John Hudson had it exactly right, above.

bieler's picture

William

Well, let see it I have this right. I have a letterpress specimen produced by Linotype itself. Very well printed, with minimum impression on a fairly receptive stock, with just the right about of ink control given to it. Representative of what they would want the market to see. The 96 point they present would most likely have been photographically scaled from the 14pt and reproduced photomechanically. My scaling of the digital type to that size indicates a very near match with only the slightest variations built in to correspond to a new printing technology.

So if I print the digital Electra with photopolymer plates I would expect a fairly accurate representation of the original, as a product of letterpress.

If the letterpress process itself is really what is giving Electra its liveliness, then it really isn't a problem with the digital rendition as much as it is a preference for the way letterpress reveals the face. But, impression and ink gain will alter any typeface.

I've printed with letterpress for near three decades now. I obviously prefer the printing process. But I also have a great deal of respect for the work being done with digital typefaces; the impetus "to" revive classical faces, the concerns for optimization and legibility. I don't believe the digital version is the travesty some of the folks here wish it were.

William Berkson's picture

Gerald, yes it may be that the digital version of Electra, if printed by letter press - assuming your photopolymer plates have similar ink spread as metal did - would correct the problems of the digital version, and get a similar superb result. My point from the start was that the digital version should have taken into account that almost all the printing would be from computer to photo to offset press, and so should get the supurb result by the current common technology. However, if Kent Lew's observations are correct - and he seems to have made a deep study of the matter - the digitizer also made the font more uniform than Dwiggins had intended, and made other adjustments. If so, I suspect that the result will still be not as good, even though the problems with the weight, particularly of the thin strokes, will have been corrected by the printing process.

I am not saying that letter press is inherently better, only that the change in technology sometimes wasn't taken into account properly in the digitizing of old typefaces. Now letter press can get 'sock', the slight embossing of the paper, which is nice. And you tell me, is there any difference in the density of the ink in letter press vs offset? To me these advantages may be real, but aren't that big. The main problem is suiting the font to the printing method.

Miss Tiffany's picture

I have to agree with Kent on this. Electra is not a soft round old-style typeface. It does sparkle more than does the average Garamond (given the same printing conditions).

The biggest problem comes with how the user "sets" the typeface, and second what printing conditions they will use.

I truly think that Dwiggins would have whole-heartedly embraced the technologies of today. He'd have tested his designs under all printing technologies too. He'd have wanted his snap and pop by-jiggidity and he'd have had it too.

bieler's picture

"I truly think that Dwiggins would have whole-heartedly embraced the technologies of today. He'd have tested his designs under all printing technologies too. He'd have wanted his snap and pop by-jiggidity and he'd have had it too."

Best not to hold hands with the dead. They might not appreciate it.

Miss Tiffany's picture

Okay, you have my attention. Tell me why you think this is.

bieler's picture

William

Well, I think the designers did suit the font to the printing method. If they would have emboldened it to create a faux letterpress look, would not that have been the wrong thing to do? Electra is a spindly face, it was as a metal face as well. It is just more revealed as such on digital printing devices.

Problem with trying to replicate a metal face with inkspread _in sito_ is that the gain is not uniform and also varies considerably based on presswork, substratum, printing press, etc.

Yes there is significant difference in the inking of offset vs letterpress. Letterpress is a continual gain that must be monitored continually. Once the ink grows out beyond the letterform outline (not a significant factor in offset) it progressively gets worse as the ink is viscous and continues to draw. There is good letterpress and bad letterpress. But in any printing medium you would want the clarity of letterform to be revealed as the designer intended. In this case it would be as the revival designers intended, not Dwiggins.

Fournier said "the letter should not be blamed for the fault of the ink." This in reference to the punchcutter versus the presswork of the printer. I'd suspect any type designer would like to see their creations presented in the best light, would they not?

hrant's picture

Gerald, would you say that over-inking/pressing is much more common than under?

hhp

John Hudson's picture

Perhaps the best approach is to consider the 'spindliness' of the digital Electra independently of the metal version. I think we're too focused on whether the weight matches the letterpress, and not considering the more basic question: is the digital Electra too spindly to work as a text face in typical offset printing situations? Yes, I think it often is too spindly, especially on white or very light paper stocks. If the visual sense of a typeface is that it is being overwhelmed by the paper, then that typeface it too light to be used at that size on that stock. I've seen numerous books set in 11pt Electra that I would not want to read on account of the weakness of the type. This is irrelevant on the colour of the metal type, or the digital printed letterpress via polymer: this is something that can be judged soley with reference to the digital type as printed offset. Personally, if I intended to use Electra for anything printed offset, I would probably spend a couple of days beefing up the outlines in various ways. I wouldn't do this in reference to any letterpress printed specimen: I would do this in reference to what would read best at the type size I intended.

Miss Tiffany's picture

Gerald. I'm still waiting and very curious to hear your answer.

hrant's picture

Very good point.
But would you also make the outlines somewhat bumpy, and the baseline somewhat jumpy?

hhp

Miss Tiffany's picture

Hrant,

How can you really know, except in extreme instances, when you've over-inked or under-inked? Could some of this be considered a matter of taste?

William Berkson's picture

I would echo Hrant's question to John Hudson, but more positively. Do you agree with Kent Lew that the 't' and 'w' were deliberately designed to dip more below the baseline than usual? Does my theory that the digitizer forced more uniformity on Electra than Dwiggins designed hold any water?

The slight fuzziness of the letter press outline I doubt can be imitated, even if you wanted to.

hrant's picture

I'm a pygmy at this stuff, but since Gerald might have already started
his trip towards Pasadena (class tonight - yay!) let me show you something:
http://www.themicrofoundry.com/other/joanna_c2.gif

This is a lousy scan of a novice proof on lousy paper that I did in class a couple of weeks ago. It's Joanna 14 at two impression levels: the right side is 2 mils* deeper (the difference of one page added to the stack). You can see how much difference that makes in the letterform: the body on the left is not black enough, but the contour is great; on the right the thins and the serifs get chunky, and the cusp between the stem and the arch starts filling in (long live trapping). Conclusion: you need the impression depth on the left, but with more ink. But my hunch is that when you're in a hurry (saving money), you're more likely to just press it harder than regulate the ink.

* A "mil" is 1/1000th of an inch.

hhp

hrant's picture

> The slight fuzziness of the letter press outline I doubt can be imitated, even if you wanted to.

About 3 years ago, motivated by the hunch that slight (I mean really very slight) fuzziness helps make letterpress more readable than offset, I had this crazy idea that all you'd need to do is unfocus the laser in your imagesetter by a hair... Anybody willing to donate their Avantra for experimentation? :-)

hhp

trajan's picture

"About 3 years ago, motivated by the hunch that slight (I mean really very slight) fuzziness helps make letterpress more readable than offset, I had this crazy idea that all you'd need to do is unfocus the laser in your imagesetter by a hair..."

Not a crazy idea, if you were dealing exclusively with letterforms with no halftones. You wouldn't "unfocus" the laser, though, you would merely set the base laser intensity a little higher than you would in a normal base linearization. That would give you slightly bolder letters and would probably come closer to approximating letterpress process with offset plates than anything else I can think of.

You'd need to find a pretty adventurous prepress house, though, and most -- needless to say -- are not.




trajan's picture

"I have a letterpress specimen produced by Linotype itself. Very well printed, with minimum impression on a fairly receptive stock, with just the right about of ink control given to it. Representative of what they would want the market to see...My scaling of the digital type to that size indicates a very near match with only the slightest variations built in to correspond to a new printing technology."

Was this printed on coated stock? I would use that and/or "Big Red" as definitive examples, since your ink spread is going to be far less severe than on uncoated stock. (Was the original Specimen brochure on coated?)


hrant's picture

> That would give you slightly bolder letters

No, that's not fuzziness, that's just weight, which can be done in the font - no need for any hardware change. What I'm talking about is very much fuzziness (albeit at the subvisible level), and you can't do that with [current] software.

hhp

Syndicate content Syndicate content