New to Typophile? Accounts are free, and easy to set up.
Create an account
Typophile RSS | More Feeds
I see that in some fonts sharp angle between paths is padded. What do you call this thing? What's the purpose of this? How large should it be?
Thanks for any answer.
A piece of trivia: the Trapping Flower was used in the production of FF Ernestine.
Hrant, the question is about an acute angle that has been filled in, not one that has been opened up.
Well, I admit it's impossible to be sure. But note the following:
- That's actually what a -simplified- curve/curve trap looks like. It's an approach favored by some people (including Carter AFAICT) even for straight/straight traps, for its simple elegance.
- Many people put shears at the ends of their traps, especially when they [have to] go too deep.
- When I look at the darkness of that area, it seems a bit too light to me, which means the curves have been pulled in (like you would do when trapping).
- That "B" seems to have gentle thorns (the inverse of a trap) so it seems to be attuned to output degradation (where trapping is most relevant). But I admit that could just be a stylistic thing.
Here's a detail of the Ջ in Ernestine:
If it is filled-in, the reason is probably to avoid super-sharp corners, since they can cause rendering problems. I think Doyald Young recommended that any corner sharper than 30 degrees should have such a shear.
It would be very helpful to see other glyphs, like the Bold "v".
Thanks for the answers. The "B" in the picture above is from Microsoft's Candara font. I've checked the "V" of the regular and bold version of Candara, and their corners are filled too.
I want to add this corner filler thing to my non-Latin font. I don't know how large the fill should be, so i tried to google for more information. Unfortunatelly I don't know what this thing is called among typographers. Perhaps to determine the size of the filler, I'll just look and compare the filler in the professionally made fonts.
We do this partly to lighten joins (trapping), partly to avoid problems when shapes are outlined, and partly because FontAudit and all the seniors tells us to.
I do it to add definition to angles. Usually between one and ten units, depending on the design. Angles can look very sharp without.
Yes, look at other fonts to figure out what your preference is.
If you look at Candara's "v" you will see a clear trap there, even in the Regular. Now look at the "k": no shears on those sharp corners... Plus, like I said, Candara even has thorns. So since it's a very "pointy" design with at least some trapping, I think the shear in the "B" is a classic way to finish off a trap (which might be done to avoid it going too deep). But we can always ask Gary. :-)
BTW whether you want to add a shear to avoid pointiness or not, the Trapping Flower can do it: once you decide how big you want the shear to be (see below) you can figure out which measurement ring it should fall on, and use that ring as the cutoff for all the inside corners. You can even use the Flower to make thorns BTW - in which case it would be a rose... ;-)
To decide the size of your shear: look at some respectable fonts, but do mind the point size range they're intended for, since fonts for display will have smaller shears than fonts for small text.
This is best thought of not as an angle that is filled in but as a crotch in which the confluence of two curves or diagonals is spread apart a little. The purpose is, as Frode says, to lighten the join and stop the area becoming too dark where the bowl or stems come together.
[Ink] Trapping is something very particular, which I don't think applies in this case. The purpose of an ink trap is, as the term suggests, to capture ink during printing. It is a special treatment of an internal angle designed to compensate for the phenomenon of inkspread that would otherwise distort the letter shape at this point. As such, ink traps are properly designed relative to output technologies. They are related to but technically distinct from two other phenomenon: similar feature in designs for phototypesetting that compensate for photomechanical softening of corners, and more general, output-independent optical compensations of the kind used in Candara and many other designs.
This illustration shows how this type of construction enables one to open up the crotch while maintaining the shape of the curves further out. The same curves without the 'spreader' segment is in the background layer.
PS. These days, I usually put a spreader section in the crotch of two straight lines only if it is a particularly acute angle or if the type weight is heavy and the white area small. But I always put one where convex curves come together.
That's why I don't use "ink". :-)
I think the idea is the same, so to me "trap" is a solid term here.
A trap traps something.
I suppose if I had to call this kind of design feature something, I'd call it a spread crotch. But that's, um .. indelicate.
Maybe it's trapping darkness. Or you could look at it from the white's perspective, and say that it's trapping light.
BTW to me "spread crotch" is a bit vague. I would prefer "silicone cleavage".
Very sharp points in type outlines are, paradoxically, vague -- their appearance being determined to an extent by the resolution at which they are rendered.
By pre-empting the blunting that press gain, for instance, may produce, the designer produces a more consistent design, with a nominal appearance controlled to be closer to the set result than the really sharp points that may otherwise characterize the typeface when closely inspected in, for instance, online PDFs.
Well expressed, Nick.
Nick expressed my motives for choosing this construction.
Incidentally, I’ve always found it amusing that the term “ink trap” was originally used pejoratively for an attribute of a letterform that filled excessively on press, and it was precisely these ink traps that Griffith sought to *eliminate* and work around with the Linotype Legibility series of newspaper typefaces, ca. 1920s.
I understand the motivation (and Nick did indeed express it well) but I wonder whether this is another case of designers wanting more control than they can really have, resulting in the sacrificing of something more important than "consistency". Namely the quality of the form when it can be consciously appreciated (read: large) versus when everything just blends in (read: small).
Note that this thread exists because somebody noticed something they weren't supposed to.