Re: Scrolling performance



On Tue, 20 Jun 2006 15:26:51 +0200, Clemens Eisserer said:

> Where does all the time go to? I've done some sysprof-profiling and
> its shows than pango is not the evil. Are maybe the repaint
> algorythmns not optimal, repainting too often too large areas/hidden
> components?

So where *does* the time go to, if not pango?  The more detail you can
provide, the easier it is to track down.

For what it's worth, I'm using an older Dell C840 laptop (1.6Ghz CPU, 768M of
RAM, and an NVidia 440Go chipset), hardly a screamer, and it's able to keep up
just fine. (Although I admit that if I grab one 700x1000 window and start using
the mouse to move it back and forth as fast as I can move the mouse, in front
of 2 other 700x1000 windows, and the window manager is doing the moving window
in translucent mode as well, that *will* spike the processor up to about 100%
and only 8 or 10 redraws per second.

I'm not saying it's *not* an issue for you, but simply pointing out that
not everybody is seeing the problem - on my machine, GTK2 *is* a bit heavier
than GTK1, but not to the point where it's a major issue.  And in general,
it's really hard to chase down performance regressions in an open-source
project without somebody actually seeing the problem and willing to work
in some detail to chase it down (been there myself - there's been more than
a few times when I've been literally the only person who was both seeing
a given bug in the Linux kernel and in a position to help debug it...)

Attachment: pgpI3fttnJ7Ob.pgp
Description: PGP signature



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]