Power & Memory Usage Of GNOME, KDE, LXDE & Xfce

[KAMiKAZOW]

I'll probably get flamed away for this, but if GNOME (Ubuntu) would have had higher memory usage compared to KDE (Kubuntu), then most people would have come and said how this proves once again why GNOME fails for some many reasons. Now that KDE, which generally gets quite a bit of positive support in these forums, comes out as the "loser", it's all but but but and this is flawed.

Ok so KDE uses more memory. So? Doesn't make it a bad DE or GNOME the greatest DE ever. But instead it's "But on my special-uber-plasma-pwn machine KDE actually only uses -15 MB of RAMz. You fail!!!!11111"

The benchmark methology is flawed, even if GNOME had lost. As I explained on Page 1 it's intransparent. And as I also explained, it's well known that KDE SC's initial RAM consumption is higher than GNOME's, but it also levels out once actual applications are used.
Read my posting, because I don't want to repeat myself over and over again.

I'm tell'n ya man. Those green wallpapers have to go. It completely flavors peoples opinion of the distro. Green highlights are good but taken too far...

There is a strong possibility that openSUSE 11.3 won't be green. It's not decided yet but the artists said that green is a bad color for LCDs. Don't know, don't care as long as the result looks good.

I never could understand that. I keep installing version after version of openSUSE (since 10) only to delete it after being completely fed up each and every time. The biggest issue is that they bastardize the KDE UI to fit some kind of ugly, perverse design

openSUSE 11.2's Plasma theme is just a slightly modified Air theme. It's 98% the same as vanilla KDE SC.
openSUSE 11.2's KDE themes were modified by KDE's Oxygen artist team itself.

The second biggest is that they consistently manage to make the distro underperform. The last stroke: >40'' boot time on an Intel SSD (when Ubuntu 9.10 boots in ~10'' on the same hardware).

That's because openSUSE is a more conservative distro than Ubuntu and Fedora. Both are bleeding edge (=kinda beta) by design. Fedora and Ubuntu for example both use Upstart, while openSUSE still uses classic SysV init.

While KDE by itself is both usable and pretty (although I disagree with a few specific design choices), openSUSE somehow makes it feel ugly and inconsistent. Kubuntu is closer to the original KDE feel and Arch is, of course, vanilla KDE.

Kubuntu is known for its half-assed backports into the mainline distro. Even the Kubutnu project itself acknowledges that. Read the "Project Timelord" announcement.
I'm curious to see Kubuntu 10.04, because it promises to fulfill the goals stated in "Timelord".

That said, Kubuntu has a really bad history, especially when it comes to localization. Check out http://www.flickr.com/photos/1961688...7608562200171/

But that's not Kubuntu's only flaw. Another major one is that KDE SC 4-based Kubuntu releases never ever shipped with a stable graphical package manager. One release shipped Adept 3 beta and the ones afterwards ship KPackageKit which within KDE's repository resides in the "playground" branch which means that it's somewhere between alpha and beta status.

Usage of Kubuntu does provide an automatic entry into the main game though. Where there are pre-built binaries for Ubuntu, they're also easily installable for Kubuntu as well. There is a tonne of stuff in the repos for Kubuntu though. Also it's usually easy to find pre-build packages for Ubuntu where a project isn't currently shipped in the standard repos.

(K)Ubuntu isn't the only distro with many apps in the repos. Fedora, Debian, Arch, and openSUSE all have someone to package an app for you into the repo.

The KDE 4.0 release was a BIG, GIGANTIC, incredibly MASSIVE mistake for any distro to ship and Kubuntu was in that list. They certainly weren't the only one though.

To be fair, Fedora was the only mainstream distribution that made that mistake. Kubuntu 8.04 had two flavors: One with KDE 4.0 and one with KDE 3.5.9.
Same was also true for openSUSE 11.0, btw.
Everybody made it clear what KDE 4.0 will be beforehand. Only the Fedora packagers seemed to not know it.
Even the KDE project itself made it very clear and that's why KDE 3.5.10 was released after 4.1 came out. In fact, the KDE 3.5 branch is still maintained, btw.

[deanjo]

I think BlackStar had a bad experience once that will forever taint his opinion of openSUSE. The general consensus that I've encountered is that openSUSE provides a high quality KDE world. Kubuntu opinions are all over the place but generally are a bit lower others.

Ya we all know that Blackstar has an issue configuring his VM's.

[dlang]

the real benifits would have show up with limited memory

the real battery and power savings figures would have shown up if the ram had been limited to 256M or 384M, then the DEs that used more memory would have been exercising the hard drive for swap, using considerably more power (even if it was a SSD just writing to flash)

I do find it interesting that on ubuntu KDE takes more ram/processor than Gnome does.

[BenderRodriguez]

190MB for KDE, something's wrong with your setup mate

My Gentoo KDE 4.3.5 setup eats just 89MB of ram and that is including wicd network manager which is gnome based.

The ONLY thing i omitted from building was nepomuk (though i've seen it build but it is not activated).

Additionally the DE is really snappy (only systemsettings gets never cached and it has to read all those options every time).

[bytor4232]

Memory is difficult to measure on Linux systems. I tested LXDE and Xfce on my blog:

http://goo.gl/tB8m

On two radically different laptops had a difference of about 15 megs. That was on a command line only Ubuntu with just xfce4 and lxde installed, nothing else.

[deanjo]

In fact, the KDE 3.5 branch is still maintained, btw.

Packages may be still being built but 3.5.10 was the last maintenance release and hasn't been maintained for over a year now.

[bytor4232]

Memory usage amongst the various desktops is dicy at best. I did a study on my blog comparing LXDE and Xfce.

To do my test, I installed a command line only Ubuntu using the Alternate Installer, then installed:

apt-get install lxde
apt-get install xfce4
apt-get install gnome-core # using metacity, not compiz

This essentially only used debian packages with no Ubuntu modifications or additions. Xfce only used 15 megs more than LXDE, and GNOME only used 12 megs more than Xfce.

It all depends on the implimentation of the desktop. Now ubuntu-desktop uses a lot more memory than just gnome-core, and xubuntu-desktop uses almost as much resources as ubuntu-desktop. lubuntu-desktop looks to be a bit better on resources, but to provide a full modern desktop for everyday users is a resource heavy job, if you know what I mean.

[bytor4232]

Sorry, my blog is at http://leanubuntu.blogspot.com/ if anyone is interested.

[smitty3268]

What method of measuring memory usage are you using? There's a ton of different ways this gets reported, and IIRC most of them are misleading.

Also, I agree that Ubuntu + KDE and Ubuntu + XFCE isn't nearly as interesting as testing Kubuntu or Xubuntu directly. I'm not actually sure there's any difference, but I could easily see the former including a bunch of extra software running in the background which wouldn't share memory as well.

I'd be very surprised if KDE didn't still use the most memory though. It's got so much more functionality built into it that it's kind of a no-brainer - a tie in memory usage would be correctly interpreted as gnome having serious memory problems.

[mugginz]

The benchmark methology is flawed, even if GNOME had lost.

Must agree here that it absolutely was. If you want to be able to glean clear and accurate metrics in order to form well based opinions it's pretty important to get your methodology right. Especially when you know people will hold your findings in a higher than normal esteem.

That's because openSUSE is a more conservative distro than Ubuntu and Fedora. Both are bleeding edge (=kinda beta) by design. Fedora and Ubuntu for example both use Upstart, while openSUSE still uses classic SysV init.

Any distros use of "stable" packages has its own issues though and then it becomes a cost/benefit analysis issue for a given situation. Because the Linux desktop is behind the commercial ones in a few areas there can be a strong need to be using the latest possible versions in order to provide strongly needed functionality. Where you can forgo that enhanced utility it's safer to stay away from the bleeding edge.

It then becomes more an issue of do you need stable vs latest versions rather then which distro does the better job of packaging upstream.

Kubuntu is known for its half-assed backports into the mainline distro. Even the Kubutnu project itself acknowledges that. Read the "Project Timelord" announcement.

I think you have a point with regard to translations, and for those who this affects in a really negative way should look elsewhere. But for many this isn't an issue. It's good to see though that it's on their radar.

I'm quite disappointed in many aspects of KDE though with regard to robustness. There doesn't seem to be much in the way of fault tolerance in quite a bit of it. If KDE is so sensitive that absolutely every aspect of a KDE deployment must be incomprehensibly accurate with no room for any error, otherwise completely random behavior is observed without clear error reporting then I think the KDE team need to work on that. Only some parts offer decent enough fault reporting that an end user can easily find a solution to a given situation.

....But that's not Kubuntu's only flaw. Another major one is that KDE SC 4-based Kubuntu releases never ever shipped with a stable graphical package manager. One release shipped Adept 3 beta and the ones afterwards ship KPackageKit which within KDE's repository resides in the "playground" branch which means that it's somewhere between alpha and beta status.

And while you have a bit of a point here, there's no viable or palatable solution for Kubuntu right now. I would argue it made sense to ship the Ubuntu/Gnome based package management with Kubuntu until a viable solution was developed, but if they were to do something so bold as that you'd have KDE zealots crying foul because as everyone knows, you can't have Gnome software on a KDE desktop or the world implodes. On that subject a lot could be said for KDEs current bluetooth integration. I find the need again to use Gnome based parts to fulfill this role in a satisfactory way.

So the Kubuntu team are between a rock and a hard place. The benefits of a Debian based package management solution seem to rule out some of the other KDE front-ends that may of been satisfactory if not tied to RPM, etc.

(K)Ubuntu isn't the only distro with many apps in the repos. Fedora, Debian, Arch, and openSUSE all have someone to package an app for you into the repo.

That's right, but these days Ubuntu (and therefore any Buntu) seems to be the biggest blip on the radar for projects not in a standard repo though. So thats a valid reason to select one of the Buntus for a distro, and the best KDE based Buntu is Kubuntu.

It was previously stated though that Kubuntu didn't have much in the way of packages relative to the other distros and I was just highlighting that that didn't seem accurate to me, not to mention Kubuntu's easy access to the whole Ubuntu software suite.

To be fair, Fedora was the only mainstream distribution that made that mistake. Kubuntu 8.04 had two flavors: One with KDE 4.0 and one with KDE 3.5.9.
Same was also true for openSUSE 11.0, btw.
Everybody made it clear what KDE 4.0 will be beforehand. Only the Fedora packagers seemed to not know it.
Even the KDE project itself made it very clear and that's why KDE 3.5.10 was released after 4.1 came out. In fact, the KDE 3.5 branch is still maintained, btw.

It could be argued fairly easily though that KDE 4 still wasn't really viable for 8.10. So any distro that was shipping 4.1 as there flagship desktop was never going look good from a functionality or reliability standpoint. However, thankfully now that's just water under the bridge and usually only gets gets brought up in KDE vs Gnome arguments.

I still think it was a boneheaded decision to give a pre-production software suit a point 0 version number. They can say all they want about how they were clear in their communications that KDE 4.0 wasn't end user ready, but I saw official KDE 4.0 release statements that didn't include warnings, so depending on what info you were exposed to it was fairly easy to form the conclusion that 4.0 was something you would want to run.

The 4.0 version number thing is really the domain of a different thread though, and if someone does start one I'll be sure to be including my 2c worth.