I got this image from the printer of a page using my typeface SeriuosSans. At the press some characters magically disappeared, while they were correctly displayed in the pdf and in the proofing print. I suppose it has to do with my font, but does anyone have any idea what could have caused this?
Ouch!
What software did you create the font with? Check winding order and -- I think I've read this as well -- if there are two exactly overlapping copies of the glyphs.
What Theunis said about the exactly overlaid duplicate path is the most likely culprit. I have seen this happen before with Adobe's Times Ten font back in the nineties.
I made this in Glyphs and Fontlab. No overlapping shapes. In the revision I checked all directions of the outlines; there were some irregularities but not exclusively to the charaters affected in the image… Right now I am testing: I hope the corrected outlines will do the trick. Never seen this happen before.
What type of font is it: TrueType or Postscript (CFF)?
I assume the printing press is full digital. Was the proofing print done on a device different from the actual press? If so, is the proofing device controlled by a software or print server different from the press?
The font is OTF (postscript flavoured), the printing press is offset, the proofs were made digital. This is an insert I made in a magazine production of someone else; I do not know the printer's specifics, but proof and execution were seperate. My big question remains: if it is in the font, what did I do differently/ wrong from the fonts I produced earlier which never gave any trouble of this kind? Riddle me that…
I kind of doubt that there is anything wrong with your font.
You say they are printing offset, so:
1. They are using film to make the plates, or
2. They are going direct-to-plate.
So I believe the problem might lie in the imagesetter or printer software since your proofing process came out correct. It may be that they are using a Postscript emulator rather than true Postscript, which can certainly cause problems.
A workaround would be "flattening" the outlines before printing: http://typophile.com/node/76217#comment-439419
have you tried to uncheck "use subroutines to compress outlines in CFF table"
Hi Diederik,
Did you find the source of the problem?
There are scripts that can easily spot if there are overlapping contours.
Captured from web.archive.org on 1 Jul 2017. view this capture