Announcement

Collapse
No announcement yet.

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

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • #61
    I still think E17 would be a good test to run with, especially considering the last announcement of being used by Ubuntu with the ARM netbook flavouring.

    Comment


    • #62
      Originally posted by Melcar View Post
      Most of the complaints with Kubuntu center around the fact that the distro seems to get a lot less attention than it's GNOME counterpart. All Kubuntu releases so far are very lackluster compared to Ubuntu releases.
      But given the alternative where they over-dilute their resources by spending less time with their Gnome release and instead put those scarce resources into a KDE one I'll take the current state of affairs anyday. And I'm usually a KDE user. If you're treating the KDE vs Gnome thing as a kind of sport or war or something, and you've got your money on KDE then I can see why you'd want more resources spend on Kubuntu, but it's not the way things are working out just at the moment. I believe Canonical are expending more resources on Kubuntu as time goes by though. That in itself doesn't justify a lot of the anti Kubuntu sentiment in my view. It just reeks of a bad attitude.

      Originally posted by Melcar View Post
      Coupled with a rather anemic program selection compared to other KDE LiveCDs out there, and Kubuntu often gives the impression as a project that isn't that well maintained.
      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.

      Originally posted by Melcar View Post
      However, to be fair, Kubuntu does tend to be a bit more "bleeding edge" is the desktop arena compared to other KDE distros; Kubuntu releases tend to include newer and often experimental packages/programs as part of their KDE desktop.
      Well I certainly appreciate access to pre-built stuff from the PPAs that provide brain-in-bucket access to new stuff removing the need to be compiling stuff all the time. I don't like to pollute a production machine with dev stuff that's needed to build from source. 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.

      Originally posted by Melcar View Post
      Stability wise, Kubuntu seems a bit more flaky than other KDE distros. Could very well be because of what I mentioned earlier about the more experimental packages and programs, but when it comes down to it Kubuntu just has a less reliable execution.
      I don't find this myself though and that's why I ask for the specifics. If Kubuntu is systematically and continuously broken where others aren't, I think specific reporting of this may result in some positive action.

      Originally posted by Melcar View Post
      Whenever I use KDE with OpenSuse rarely does it crash even after I abuse it; Kubuntu, I'm lucky if Plasma doesn't crash after changing the desktop color scheme.
      I've not found that OpenSUSE has a stability benefit over Kubuntu myself but am happy to test where there are identified situations where feature X in Kubuntu is broken but works well in OpenSUSE. Specifically, at certain points of KDE releases, parts of Plasma are broken, and then updates fix that. Depending on when you conduct a stability test you may get varying results.

      The KDE team seem more prepared to ship broken functionality than are the Gnome team. I find that this reflects badly on all KDE based distros. On the flip side KDE has an edge in certain areas over Gnome though and at the end of the day it's up to the individual whether they prefer Gnomes stability over KDEs feature set.

      Comment


      • #63
        Originally posted by yotambien View Post
        Well, memory consumption nowadays is obscene. Firefox+Thunderbird+Acroread together--the first three programs I open on startup and never close afterwards--take something like 400 Mb given enough time (1). Add three KDE applications (basically text editors), which for some reason bring with them 11 extra processes (literally), and we are already talking about 500+ Mb. Fire up Wine, close it, and witness how wineserver stays there sucking memory like there's no tomorrow (2). Give it a couple of weeks and X already demands 300 Mb you never get back. I've got 1 Gb RAM, as soon as I hit ~800 Mb the whole thing starts to swap and becomes unusable. Which is funny, because I've got an old laptop as a backup server with 196 Mb RAM that is capable of running KDE 3.5 normally. So even though I have as much spare RAM as that machine's total memory the system for some reason desperately crawls.
        Do you claim that on that 196MB machine you were running Firefox+Thunderbird+Acroread together, and then also run Wine and then give it a couple of weeks, and it all works fine with the only difference being that you run KDE3 there?

        Are you kidding me?

        Comment


        • #64
          Originally posted by RealNC View Post
          Do you claim that on that 196MB machine you were running Firefox+Thunderbird+Acroread together, and then also run Wine and then give it a couple of weeks, and it all works fine with the only difference being that you run KDE3 there?

          Are you kidding me?
          Which part of the "I've got 1 Gb RAM" did you misunderstand? His backup server has 196MB RAM, not the machine running all this stuff.

          Comment


          • #65
            He was comparing the two machines :P Saying that KDE4 consumes all his memory because he runs all the other apps for 3 weeks. lol.

            Yes, KDE 4 has increased memory usage. But the comparison he gave there is just ludicrous

            Comment


            • #66
              Originally posted by RealNC
              He was comparing the two machines :P Saying that KDE4 consumes all his memory because he runs all the other apps for 3 weeks. lol.

              Yes, KDE 4 has increased memory usage. But the comparison he gave there is just ludicrous
              Only that I wasn't comparing anything...lol?

              I pointed out that I find it funny that when my machine still has about 200 Mb free--which is about the total memory of another old machine of mine--it slows down due to swapping. I only mentioned KDE 3.5 to note that 196 Mb are enough to run a full-blown DE without major problems. But you totally managed to interpret the situation in a much 'funny' way. I give you all the credit, though.

              Comment


              • #67
                Originally posted by yotambien View Post
                Only that I wasn't comparing anything...lol?

                I pointed out that I find it funny that when my machine still has about 200 Mb free--which is about the total memory of another old machine of mine--it slows down due to swapping.
                Oh, OK. I misunderstood your post.

                To help with your swap problems, change the swappiness of the kernel. The default is very high; it will start swapping even with plenty of RAM left. The default is "60". The range is 0 to 100. 0: never use swap, 100: always use swap even when there's no memory usage to speak of. It works a bit like a percentage, the higher it is, the sooner swap will be used. Some people reduce it to zero, but I don't recommend that since values under 20 can impact performance negatively. I use 30. Put this in /etc/sysctl.conf:

                vm.swappiness = 30

                After the next reboot, swap should only get used when memory is really about to run out.

                Comment


                • #68
                  I too think it has to do with that. Actually 30 is the value of swappiness I currently have, I guess I should experiment with it a bit more. The problem is that you read different things here and there and you don't know who to believe. According to Andrew Morton (if I get the guy right) users shouldn't bother with this: the system is doing what needs to be done; but according to what _I_ see, there's something sub-optimal going on.

                  Thanks for the tip, anyway.

                  ; )

                  Comment


                  • #69
                    I've been using Kubuntu for the last year or so. 9.10 was a fantastic release as far as Kubuntu goes, but honestly, it's the best so far. I've tried most of the other big ones (OpenSuse, Fedora, Arch, and Mandriva).

                    I really hated Opensuse because of it's graphical package management. Since I don't know what the package names are, I have to fire up Yast (I believe). It's so fricking slow; all the resources have to be updated. I tried to add some extra repositories for multimedia and updated KDE, but that causes tons of incompatible packages that I had to manually approve. I do have to say that 1-click install is very nice -- not that I could get it working.
                    I actually like Fedora quite a bit (I'm running it on my old P4 desktop). I'm just not that familiar with the console stuff, so it takes me longer to do stuff. Presto is great when you are updating KDE or for other large updates. SELinux has bitten me a few times.
                    Arch was a huge disappointment. The biggest problem that I have is how basic it is for beginners. If I'm a "beginner," then I'm just going to follow the guide and not have the expertise to make many modifications. The consequence is that I get a setup that is simply the beginners guide; why can't there just be a beginners spin?? After the second setup or so, it's not that bad, but still a huge waste of time. The thing that really put me off is that I couldn't get CPU scaling enabled, which nullifies all the benefit of running a lean distro on a laptop.
                    Kubuntu is certainly not perfect, and there's been lots of stuff that has been broken for a long time (Nepomuk), but it works well enough, and I know the underlying system.

                    As far as memory usage goes, I'm not too concerned. Here's what free reports on my system (KDE 4.4.1)

                    total used free shared buffers cached
                    Mem: 3074628 2022572 1052056 0 141712 1418784
                    -/+ buffers/cache: 462076 2612552
                    Swap: 0 0 0

                    System Monitor reports 0.45GB. I do have preload installed with aggressive settings, so that should be using quite a bit, but I'm not sure how that gets counted. I have 3GB, and I regularly get upset at how poorly Linux caches data. I have to second a poster earlier about the KDE libraries. It seems like Gnome Apps have to load a lot more data on startup; whereas, KDE has better common libraries , which causes more to be loaded on startup, but less with many apps running.

                    Quick side note about preload. By default it excludes your home directory from the cache. This usually doesn't matter too much, but Chrome and Firefox (along with all other apps, but browsers tend to have large profiles) store user profiles there. If preload can cache those profiles they open significantly faster. If you want to change this edit /etc/preload.conf and find the mapprefix line and add your home folder.

                    Comment


                    • #70
                      Originally posted by yotambien View Post
                      I too think it has to do with that. Actually 30 is the value of swappiness I currently have, I guess I should experiment with it a bit more. The problem is that you read different things here and there and you don't know who to believe. According to Andrew Morton (if I get the guy right) users shouldn't bother with this: the system is doing what needs to be done; but according to what _I_ see, there's something sub-optimal going on.

                      Thanks for the tip, anyway.

                      ; )
                      Morton is full of crap :P He wouldn't be if he was using normal applications on his "desktop" instead of Emacs and an xterm; those don't take insane amounts of time to come back from swap in his "swappiness=100" setup (Morton uses 100 as swappiness, proof that he doesn't really run real-life applications; xterm and Emacs aren't such.)

                      Comment

                      Working...
                      X