KDE do some caching (I don't know what the settings are in Kubuntu, but it is probably enabled).
It's like comparing Windows XP dans 7. 7 will max out your memory at anytime, but in fact use less tahn 10% more than XP. KDE do a lot of caching to make apps load faster. Memory don't need to be wasted. All modern OS do some caching to make the system feel faster. Of course Kubuntu is a terrible KDE distro and feel slower than Vista, but it's an other story. But please keep in mind that it does mather in a professional benchmark, your KDE stats are biased.
Only Gentoo with -Os in make.conf will give the real picture. At least you know that both DE have been compiled the same way, and not optimized with different settings.
Now I'm curious as to how well e16 - or even better, e17 - stack up to everything else.
If you believe a flame war started I think you are looking at the responses wrong. All I have seen is the fact that they are performing a test on a distro that has been known to bundle in extra software and startup processes. True test results should be completed on vanilla system or you are going to mislead people.
Originally Posted by bugmenot
Yes, it does.
Originally Posted by KAMiKAZOW
Btw. it seems this comparison is misleading:
i was gett'n the same bad vibes as bugmenot was from this thread, man. :hippie:
Originally Posted by migizi
darn. this forum needs more psychedelic smilies.
Well, what distro did you want him to use for the test? Most people don't install a vanilla distro and then add their desktop? They choose their fav. distro and pick the version they prefer (many distros have a DE version). I guess if you are talkin about Gentoo????
KDE 4 is said to be a resource hog now. I think it's tough to assess because both desktops use the library of the other for certain applications. Or am I wrong there?
I liked this article and test because I was curious about the DE since I was wondering what DE to use for my older laptop. I currently have AntiX and Debian Squeeze with LXDE on it. I like LXDE but a few minor annoyances I had was that it didn't seem to have many native apps so I picked and chose ones I already like (but, it then had to add the libs from either Gtk or Qt depending on the app). I am not sure how well that works and how well the lighter desktops integrate. Anybody know?
I'm a total noob regarding this stuff so if I sound silly on any of it, that's why but at least I admit it, right?
don't forget that KDE also provides a lot more features out of the box. Or does XFCE or gnome have nepomuk enabled?
And what happens if you start a couple of apps? Firefox for gnome/xfce, konqueror for KDE? Abiword for gnome, kword for KDE?
And if that test program was gtk based, the results are skewed and crap anyway.
The methodology used to conduct these "benchmarks" is *seriously* flawed.
1. You were NOT benchmarking GNOME vs. KDE vs. LXDE vs. Xfce as such, but rather their *buntu incarnations.
2. The last time I tested Kubuntu, KDE had a horrible memory usage there because of some crap applications running in the background.
3. When I restart my computer into KDE 4.4.1, my memory usage is approximately 190MB used by applications (that does not include cache of course).
4. Instead of forcing me and everybody else to read the totally cryptic paragraph describing the system&OS configuration, why not simply put those specs into a nice and readable table!!!
5. The sad truth is that many desktop applications are using too much memory these days. For example, KCalc in KDE 4.4.1 uses 3 MB of private memory (not including shared memory). Now tell me what those 3 mega-bytes could possibly contain in such a trivial calculator application! I understand what that memory is being used for, but in my opinion the facts are horrible and speak of inefficiency.
6. KDE uses Qt. The memory usage of KDE may be mainly caused by the fact that it uses Qt. So, first it is Qt to be blamed, only then blame KDE. A trivial graphical "Hello world" in Qt4 uses more than 2 MB of private memory. Am I alone in thinking that this is clearly a sign of a completely wrong software architecture?
yes, you are alone. Qt isn't very memory hungry.
But that does not change the fact, that the whole thing is complete bull.
Seems you have different standards than me.
Originally Posted by energyman
But let me ask you a question: Have you ever looked into the source code of some Linux executable or library? I have, and from what I have seen I can tell you that the amount of inefficiency in common Linux programs is breathtaking. Seriously, if you think about the algorithms used there and compare them to "ideal" implementations, the result is shocking. The number of totally pointlessly burned CPU cycles is massive.