Round only on-curve points to grid

RoboFont › Forums › Enhancements › Round only on-curve points to grid

This topic contains 6 replies, has 2 voices, and was last updated by  frederik 8 years ago.

Viewing 7 posts - 1 through 7 (of 7 total)
  • Author
    Posts
  • October 17, 2012 at 10:38 #5151

    Jens K.
    Participant

    I think it would be nice to have an option to only restrict on-curve points to the upm grid, allowing curve control handles to be positioned freely.

    I usually want stems of a defined (integer) width, but want full precision when adjusting curve handles so the curves can flow smoothly and control handles can keep an exact angle. Restricting only on-curve points to the grid would IMO be a worthy compromise.

    October 17, 2012 at 10:50 #5152

    Jens K.
    Participant

    That being said, is it possible at all to generate fonts with floating-point coordinates with Robofont?

    October 17, 2012 at 12:01 #5153

    frederik
    Keymaster

    floating points in binary fonts is not possible at all :)

    October 17, 2012 at 12:16 #5154

    Jens K.
    Participant

    Hi Frederik, there seems to be a way to put floating point coordinates in OTFs, because Glyphs does it when you turn off the rounding to upm grid there. I verified with ttx that the font actually contains floating point coordinates, and ttx has no problem decompiling and compiling such a font again.

    October 17, 2012 at 12:19 #5155

    Jens K.
    Participant

    There’s a related discussion here: http://typedrawers.com/discussion/123/maximum-upm-for-otf-fonts/p3

    October 17, 2012 at 13:28 #5157

    frederik
    Keymaster

    Havent tested this out. The main issue which rasterizer is supporting floats. During development this was possible but the binary fonts freaked out in lots of cases.

    Why worry about half a unit? scale up the ump or hint the font properly.
    Otherwise I know RoboFont allows you to draw with floating points.

    October 17, 2012 at 13:32 #5158

    frederik
    Keymaster

    And to reply to your first questions (different rounding behavior for off curves) :

    This is maybe useful when the offcurves aren’t extreme points and the bcp length is rather small (1-2 units). In all other cases, where the points are placed at extreme points this is not really an issue.

  • Author
    Posts
Viewing 7 posts - 1 through 7 (of 7 total)

You must be logged in to reply to this topic.

← Previous
Next →