Re: Floating point in pango

["Sean Kelley" <sean sweng gmail com>]

Hi

On 7/16/06, Matthew Allum <mallum openedhand com> wrote:
> Hi;
>
> On Sat, 2006-07-15 at 18:14 -0700, Carl Worth wrote:
> > >
> > > Carl, what are your thoughts on this ?
> >
> > I haven't looked closely at the issue yet.
> >
>
> Please spend some time checking Jorn's posts + profiles, hopefully they
> clearly show the issue with floating point causing slowdown in the font
> rendering case at least ( on ARM ).
>
> > But my general take on this is that we are not ready to put new API
> > into cairo for "performance reasons only" since we haven't
> > sufficiently tried to optimize the implementation of the existing API
> > yet.
> >
>
> What do you see as downsides in extending API like this ? Would it even
> be that bigger addition ?
>
> Id be interested your thoughts on optimising FP to perform when you dont
> have an FPU. Maybe the amount of FP calls in pango-cairo could simply be
> heavily reduced ?
>
> If Cairo cares about performing well on embedded type hardware ( maybe
> it doesn't? ), use of fixed point seems pretty much mandatory on lower
> end hardware - For example OpenGL ES, Playstation and Nintendo Handheld
> graphics API's all use it.


I think many embedded folks like myself are in a very serious bind
because I need pango and Gtk's evolving features  But more importantly
I need pango-cairo to be fast.  It can be a double whammy because
there are aspects outside of pango-cairo that I need in Gtk+ 2.8.  So
do I maintain massive patches to pango-cairo and Gtk+ 2.8-cairo
stripping out floating point?  Plus then do I fork cairo's api to add
fix point?

These are tough questions because, honestly the landscape is like a mine field.

Sean

> Many thanks;
>
>   -- Matthew
>
>
>
>
>
>
>
> _______________________________________________
> Performance-list mailing list
> Performance-list gnome org
> http://mail.gnome.org/mailman/listinfo/performance-list