Re: FT Performance Regression?

[Alexei Podtelezhnikov]

> > This is definitely not a bottle neck
>
> Maybe, but 1/3rd of a function call being spent not doing actual work is
> definitely concerning in a world where you only have 33.3ms or 16.6ms to do
> all your work that frame.
>
> It looks like setjmp in FreeType is basically used to do exception-like
> handling in C, in lieu of using and checking return values?
>
> Looking through the mailing list it seems to come up periodically, mainly
> when people are either missing setjmp, or have a really bad implementation
> of it.
>

Let not make quick judgement calls. Rasterizing time grows proportionately
with glyph complexity. Ultimately, the memory and CPU time is spent
generating scanlines of spans. FreeType is optimized for rather usual case
of of 2 vertical stems in a glyph which you can identify is absolute
majority of scripts. In letters like AHORU, eight spans per scanline is
quite sufficient. Letters like NMW need more and you can actually observe
more longjmp's.

If you do see too many longjmp's, you need to configure FreeType for your
font needs. Out of curiocity, which font gives you so much headache?

Alexei