On Mon, 2005-08-29 at 09:11 +0200, Stefan Kost wrote: > hi, > > just read http://www.rasterman.com/index.php?page=News (Sunday, 29 May 2005 > E17 is being optimised) > Good to see that Metacity does quite well. Still the first test can probably > go > faster. Good to have a stress-test tool available with > http://www.rasterman.com/files/wm_torture-0.1.tar.gz. > > Any comments from the metacity devs? Take the challenge ;) >
Spent a lot of time when that came out going around saying "metacity ownz you in window mapping throughput!" and getting funny looks. Seriously, these benchmarks have nothing to do with anything that matters... when was the last time your application launch time was primarily dependent on the WM, or when was the last time you opened 100 windows all at once and were all DAMMIT WHY DOES MY WM TAKE 1.3 SECONDS INSTEAD OF 0.7 SECONDS TO FLOOD MY SCREEN IN 100 WINDOWS??? ;-) But if you want to improve the benchmark anyhow, http://bugzilla.gnome.org/show_bug.cgi?id=309567 shows one way. It should speed up some other stuff and clean up the code as well so it isn't entirely useless. Havoc _______________________________________________ gnome-devel-list mailing list [email protected] http://mail.gnome.org/mailman/listinfo/gnome-devel-list
