Re: [ft-devel] Rasterizer performance benchmark

[Werner LEMBERG]

> The rasterizer I should benchmark against for anti-aliased
> renderings is ftgrays?

Yes.

> From what I can read in that code there is no hinting done in this?

No.  This is done separately.

> And there is no performance decrease in setting the STANDALONE flag,
> right?

Not as far as I know.

> And lastly, the ftgrays rasterizer is truly _exact_ as it claims?

I think so -- it computes the exact pixel coverage (at least this is
what David says :-)

> After the setup I run the same tests on my rasterizer and ftgrays.
> That is I rasterize all glyphs in the font Vrinda.ttf 1000 times and
> measure how long it takes, and I only use font sizes up to ~40 px.
> The reason I ask for your help here is because from my test it seems
> that I get better performance, which I find highly unlikely.

Not necessarily.  I don't know how much David has tuned the
rasterizer.  We welcome any improvements :-) Note, however, that the
performance depends on both the platform and the compiler, so you
should test this extensively.  Does your rasterizer handle both second
order (TTF) and third order (PS) curves?

> So the main issue here is, is my test valid and fair?  [...]

It seems so.

> Is there any other flag or any other thing that I have missed?

Unlikely (I say this without looking at the source code).

> Is this as fast as any anti-aliased FreeType rasterizer goes if I
> don't want any hinting, bitmap allocation or rescaling or anything,
> just measure the time it takes for the rasterizer to fill a bitmap
> with an already scaled point set?  Could things like subdivision
> cutoff be tweaked or anything?

I don't know.  Perhaps David finds some time to chime in.


    Werner