[cairo] Re: Gtk performance issues from a user's point of view

[Kalle Vahlman]

2006/10/9, Kalle Vahlman <kalle.vahlman at gmail.com>:
> 2006/10/9, Federico Mena Quintero <federico at ximian.com>:
> > In particular, the torturer makes it easy to plug arbitrary widgets and
> > get timings when they are exposed/resized/etc.  You may want to plug
> > your whole complex window and see what it tells you.
>
> Yeah, that's the next step I guess.

Just a quick note on what I got from that:

# 3 x 240 iterations -- Single widget torture
# Widget        Expose           Resize
GtkButton       0,00731258       0,0553553
ScwView         0,025231         0,154395

The boot times are meaningless as ScwView needs a model to draw and
basically a GtkScrolledWindow to keep it from asking for all the space
in the world so the generation of 200 row model is there too. Also the
expose and resize include the scroll bar of the scrolledwindow (but
this is pretty much ok as it usually is the case anyway).

So, looks like one view matches three buttons on resize. Thus the test
case (with 11 buttons IIRC) would be about right given that the
buttons occupy roughly 30% of the area on average: 1/3 area * 3
buttons per widget * 11/3 "worth of buttons"-ratio == 11/3 ~ 4 which
is only one off of the original percent ratio (24%/8%).

And if you believed any of the last paragraph, I have some really cool
stuff to sell you. Just send me your credit card number and wait for
"surprises" ;)

Perhaps it's time to sleep over it...

-- 
Kalle Vahlman, zuh at iki.fi
Powered by http://movial.fi
Interesting stuff at http://syslog.movial.fi