[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-core-devel
Subject:    Re: icon loading (app startup) speed (long)
From:       Marc Mutz <Marc.Mutz () uni-bielefeld ! de>
Date:       2001-08-10 19:26:52
[Download RAW message or body]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Friday 10 August 2001 18:37, Antonio Larrosa Jiménez wrote:
<snip>
> These are the times I got by running "time konqueror" :
<snip>
> ## Modified libraries to a "best (aka impossible to get) case"
> real    0m5.915s
> real    0m5.996s
> real    0m6.182s
> real    0m6.104s
> real    0m6.108s
<snip>

mean real 0m6.06s

> ## Normal libraries
> real    0m7.036s
> real    0m7.122s
> real    0m5.948s
> real    0m6.219s
> real    0m5.942s
<snip>

mean real 0m6.4534s

You might want to compare the wall-clock times, because that is what 
the user experiences. User and system times are completely unimportant 
from the user's POV (though they're interesting for the admin of a 
multiuser system)

Suddenly the difference is ca. 0.4 seconds. Granted, this isn't much 
either, but the full second on the first measurement would be quite 
worth it, no?

> ## Modified libraries to a "best (aka impossible to get) case"
> real    0m1.690s
> real    0m1.761s
> real    0m1.768s
> real    0m1.782s
> real    0m1.731s

mean real 0m1.7464s

> ## Normal libraries
> real    0m2.452s
> real    0m1.775s
> real    0m1.808s
> real    0m1.787s
> real    0m1.945s

mean real 0m1.9534s

> That makes 1.442 seconds for the best theorical case, and 1.518 for
> the current case, which makes a difference of 0.076 (user) seconds.
>

... or 0.207 (wall clock) seconds or 10% speedup.

Marc
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE7dE083oWD+L2/6DgRAlzRAJ4+0plMLV40oK+apc4lPg45Ab1HzACfdpQe
8Z4xgU6k4Mx2h8BcnwvkZTg=
=/Jvs
-----END PGP SIGNATURE-----

[prev in list] [next in list] [prev in thread] [next in thread]