Re: [ft-devel] Autohinter: stem darkening, first rough prototype

[Nikolaus Waxweiler]

> I used a zoom-in tool to examine the pixels in your before and after
> screen shots. The left sides of the stems were unchanged while the right
> sides were darker.

Ah, okay ;)

> Alexei has explained how this is the result of using the existing
> emboldener. I don't think it is a problem. It only caught my eye because
> the CFF stem darkener, which works on the outline, applies darkening to
> both sides. If we took the approach of darkening TT outlines before
> applying the autohinter, I'd expect the same: darkening on both sides.

Technically, the autohinter could apply emboldening *before* applying 
hints. That would require more fiddling and/or more emboldening calls 
that may drag down performance, I didn't seriously look at that yet. 
One-sided emboldening could be a problem then, I don't know..

> As for subpixel rendering, that's to be expected. For vertical LCD
> stripes, it should sharpen vertical stems.

Hm, yeah. I noticed that DirectWrite's grayscale output (e.g. in Edge) 
doesn't look too clean either. Oh well. It's high time high-DPI screens 
flood the market.