Breaking point for Hairline weight?

Adam Ladd
Adam Ladd Posts: 263
in Technique and Theory
Based on your experience, is there a stem width threshold for hairline weights that is no longer functional (considering it's typically used for display)? In other words, what is the minimum width you've been able to create a hairline at before the line starts to distort or cause problems at smaller point sizes or on the grid?

I'm experimenting between 10 to 14 units, and though I like the thinner option, 10 starts to seem a bit frail in some applications (ok if set at a very large display size). Just curious what others have found.
Tagged:

Comments

  • James Puckett
    James Puckett Posts: 1,992
    I used 5 unit hairlines for the large size of my Barteldes family. I tried to do four but that was causing some terminals to end on undesirable angles.
  • John Hudson
    John Hudson Posts: 3,186
    I tried to do four but that was causing some terminals to end on undesirable angles.
    Perhaps that would no longer be an issue now that it is possible to use fractional UPM coordinates?

  • Chris Lozos
    Chris Lozos Posts: 1,458
    You might want to bump up your UPM to avoid kinking
  • Adam Ladd
    Adam Ladd Posts: 263
    Yes, a UPM increase may help. So far, in a few early drawing tests I haven't run into noticeable kinking problems. But that's at the 14 unit width.

    It's a large, display sans with a tall x-height, so the line still feels pretty fine but not frail at this size. If I dip more towards 10 (or 5, as James mentions), I suppose I'd see more evident kinking/jumping of handles as I get further along.
  • James Puckett
    James Puckett Posts: 1,992
    John Hudson said:
    I tried to do four but that was causing some terminals to end on undesirable angles.
    Perhaps that would no longer be an issue now that it is possible to use fractional UPM coordinates?

    I tried that but it caused other problems with overlaps interpolating into factional coordinates that didn’t quite match up.
  • Vasil Stanev
    Vasil Stanev Posts: 775
    edited October 2018
    Wasn't there a certain relation between ink drops/paper adhesion and stroke width? I have read so much about it that I have forgotten it completely.
  • Adam Ladd
    Adam Ladd Posts: 263
    edited October 2018
    Perhaps that would no longer be an issue now that it is possible to use fractional UPM coordinates?
    Do you mean decimal coordinates? Would be nice if they are generally acceptable, though I think I've heard elsewhere that they can be problematic with some printer outputs (but ok for the majority of situations).
  • James Puckett
    James Puckett Posts: 1,992
    Vasil Stanev said:
    Wasn't there a certain relation between ink drops/paper adhesion and stroke width? 
    It seems like that would vary by the paper, press, and ink used. 
  • Michael Jarboe
    Michael Jarboe Posts: 265
    edited October 2018
    I've used 5 and 6 unit values recently, and have seen 4 unit value hairline executed successfully, but I wouldn't be concerned with rendering below a certain large display point size, naturally. There are drawbacks to a larger UPM or fractional coordinates. I recently have come to accept the masochistic nature of merging font outlines on a 1000UPM grid in the lightest of weights. Merging outlines especially with angled intersecting stems requires varying degrees of compromise once the points snap to grid. These issues exist regardless of stem width, but of course are exacerbated the narrower the stems become or again at what angle they intersect.
  • Chris Lozos
    Chris Lozos Posts: 1,458
    Raise the UPM (of a copy)before you blend outlines with the hairline.

Categories