Roy Schestowitz wrote:
> Ubuntu, Kubuntu, & Xubuntu Benchmarks
>
> ,----[ Quote ]
> | With Ubuntu, Kubuntu, and Xubuntu each deploying a different
> | desktop environment (GNOME, KDE, and Xfce respectively), how
> | do each of these distributions compare performance-wise? We
> | have taken two systems -- one with dual Intel Clovertown
> | processors and the other an AMD Sempron -- and tested out
> | each of these distributions in some of our commonly used
> | Linux benchmarks to see how the performance truly stacks up.
> | In this article we will be presenting part one of our results.
> `----
>
> http://www.phoronix.com/scan.php?page=article&item=650&num=3
>
> Command-line-based, CPU-intensive tasks. Maybe part two will be more
> interesting...
So, I sit at my Suse with Gnome, open a console and time a compression
routine.
A mate sits at a {any linux you care to mention} with KDE, opens a console
and times that same compression routine.
Call me ignorant if you like, but in what way are we comparing Gnome to KDE?
The next tests shouldn't show any particular difference either, we have a
common graphic interface CGI. The only difference could really be if one of
the GUIs takes up enough space in ram that a bit of memory shuffling has to
take place dueing the game.
Phoronix have some interesting bits and pieces on their site, but this isn't
one of them. The chosen tests have to be tests that use items of the GUI
API:-
So in Gnome do something with GLIB/GObjects/GTK/GnomeVFS
In KDE, what would it be, kdeui maybe?
The thing is that the Gnome API covers much more ground than the KDE API
does. KDE tends to still be pretty much a graphical interface for the Linux
utilities and libs underneath, whilst Gnome does some of that but adds
improved programming interfaces via it's own libs.
I might be being unkind to KDE there, it is a long time since I knocked
together any code that was specifically for KDE and it may well have
changed a fair bit. But a quick look at the API manual just now and I
notice some additions but otherwise I can't see that it is really any
different.
|
|
