Read Roberts (Adobe)
Moderator
Hero Member
Karma: +6/-0
 United States
Posts: 27
Fontlab Studio 5.0.4, built 2741, Mac OX 10.5
|
 |
« Reply #1 on: 2009-07-27, 14:44:01 » |
|
The count vs stem width reported needs to be interpreted with reference to the glyphs. If a particular stem width is has a very high count, that doesn't mean that you should automatically include it in the list of global stem widths, as it may not be a stem that you care about. Likewise, a stem width with a very low count may still be one that you want to include, if it controls an important stem in a glyph which is used a lot. This is why the reports also lists the glyphs with which each stem width is found, so that you can look at where it occurs.
It is actually a good idea to start using stemHist very early in your design, as it is a good idea to be consistent about stem widths for the same kind of stems in different glyphs. Having the same kind of stem be 3 units different in one glyph than in another usually makes no design difference, but makes it hint differently. Also, if you are consistent in your design, then the stems widths with a high count are the important ones, and the important ones all have high counts.
Another consideration is that a stem width is actually associated with a plus/minus delta value, and will capture other stems whose width is within the delta of specified global stem width. This delta is calculated during rasterization, and depends on point size. The formula is "(0.35*72)/(ptSize)." If the rasterizer is hinting a glyph at 10 points, the delta is 2.52 design units. At 12 points, the delta is 2.1, at 16 points, the delta is 1.575 design units, and at 25 points, the delta is a little over 1 design unit.
For the specific example you cite, I would choose 128, which will cover stem widths at 127 and 129 up to 25 pts, which is good enough coverage for most fonts.
- Read Roberts
|
