Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror
×
Debian Software Linux

Is Ubuntu Getting Slower? 544

An anonymous reader writes "Phoronix has a new article where they provide Ubuntu 7.04, 7.10, 8.04, and 8.10 benchmarks and had ran many tests. In that article, when using an Intel notebook they witness major slowdowns in different areas and ask the question, Is Ubuntu getting slower? From the article: 'A number of significant kernel changes had went on between these Ubuntu Linux releases including the Completely Fair Scheduler, the SLUB allocator, tickless kernel support, etc. We had also repeated many of these tests to confirm we were not experiencing a performance fluke or other issue (even though the Phoronix Test Suite carries out each test in a completely automated and repeatable fashion) but nothing had changed. Ubuntu 7.04 was certainly the Feisty Fawn for performance, but based upon these results perhaps it would be better to call Ubuntu 7.10 the Gooey Gibbon, 8.04 the Hungover Heron, and 8.10 the Idling Ibex.'"
This discussion has been archived. No new comments can be posted.

Is Ubuntu Getting Slower?

Comments Filter:
  • "That's quick" (Score:2, Informative)

    by G3ckoG33k ( 647276 ) on Monday October 27, 2008 @08:44AM (#25525515)

    "That's quick" was the phrase my girlfriend after an update of Debian Sid to include KDE 4.1 and OpenOffice.org 3.0 from Experimental. "Wish my slow machine at work was this quick".

    You don't have to guess what OS she is using there...

    Anyhow, once you replace 3.5.x with KDE 4.1 you will notice a difference. At least I did. (No, I didn't read the article first... Bad boy.)

  • Security Patching? (Score:5, Informative)

    by Ironsides ( 739422 ) on Monday October 27, 2008 @08:51AM (#25525585) Homepage Journal
    Ok, the article completely ignores this (as do most of the above posters it appears). Each version of Ubuntu tested seemed to have different kernel builds. How much of the slowdown is due to the kernel being patched for security and bugs and how much is due to the software that has been added?
  • Re:What hardware? (Score:5, Informative)

    by gbjbaanb ( 229885 ) on Monday October 27, 2008 @09:00AM (#25525689)

    When you test new software on old hardware of course it's going to be slower though.

    That's hardly a given, lots of software gets better as it ages - new features are added, but also performance tweaks get added.

    The problem is that software should be getting quicker on the same hardware, the alternative is bloaty apps that no-one wants to use. See Vista for the ultimate conclusion to that. You don;t want Ubuntu to end up the same, so its good that someone is pointing out performance issues. Hopefully the next release will have a few of these issues looked at and improved.

  • Yes, absolutely! (Score:2, Informative)

    by Phreakiture ( 547094 ) on Monday October 27, 2008 @09:01AM (#25525695) Homepage

    Yes, Ubuntu is getting slower, absolutely, without question on my part.

    My single biggest complaint against 8.04 was that it could not get out of its own way to play an MP3 on my somewhat modest hardware (Via MII-12000). It runs fine on my wife's machine, however (AMD Sempron on Via MoBo).

    Now, it is possible that the slowdown is only with 32-bit versions. My wife's machine is running the 64-bit version, and seems to run pretty well. In the mean time, I have reverted to Slackware, which has always been my refuge.

  • by Anonymous Coward on Monday October 27, 2008 @09:02AM (#25525701)

    They were using the Phoronix Test Suite. By saying they ran 1.4.0 Beta 1 it can be found all of the versions they were using. The Phoronix Test Suite syncs to specific versions and builds them from source no matter the OS.

  • Re:xubuntu (Score:5, Informative)

    by Mr.Ned ( 79679 ) on Monday October 27, 2008 @09:15AM (#25525823)

    Xubuntu's performance targeting appears limited to choice of desktop environment, which was a small component of what these benchmarks tested. The big performance increases the article talks about were in databases, compilers, encryption, memory access, and audio/video encoding/decoding, none of which really have much to do with the desktop environment.

  • by siride ( 974284 ) on Monday October 27, 2008 @09:50AM (#25526413)
    Cache misses if the code is used. If it's not, it incurs no penalty except for the piddly bit of space it uses on the harddrive.
  • by ahziem ( 661857 ) on Monday October 27, 2008 @09:50AM (#25526421) Homepage
    Here are the numbers

    It's called Wirth's Law

  • Xorg, mainly. (Score:4, Informative)

    by ciroknight ( 601098 ) on Monday October 27, 2008 @09:55AM (#25526499)
    Namely, graphics hardware. ATi graphics hardware and the FGLRX driver. FGLRX is known to have crappy 2D performance relative to its very strong 3D performance and the 2d performance that the open source driver excels at.

    Meanwhile, 2D performance on Intel's hardware is smoking everyone else's pipe.
  • Re:What hardware? (Score:3, Informative)

    by David Gerard ( 12369 ) <slashdot.davidgerard@co@uk> on Monday October 27, 2008 @10:04AM (#25526639) Homepage

    OS X regularly gets noticeably faster each release on the same hardware.

    Mind you, they did start from the horribly unoptimised dog called 10.0.

  • by Cthefuture ( 665326 ) on Monday October 27, 2008 @10:06AM (#25526671)

    Yes! In particular check the "ondemand" CPU scaler. That thing just doesn't work very well. It takes too long to trigger the higher clock speed and if you have multiple CPU's and/or are running lots of quick processes then the clock will constantly be shifting between speeds. This totally kills the performance.

    I turned off the CPU scaling on my Ubuntu workstation and I disable it on my laptop when I need maximum performance.

    This can be fixed with two changes to the ondemand profile. First it should bump all CPU cores to the higher speed no matter how many processes/threads are requiring performance. This is necessary because the kernel shifts threads between cores and you don't want to keep switching speeds on the various cores as it does this. Second is to add a delay before dropping the speed of a CPU. This allows time so that new threads/processes have full speed immediately.

  • by BlackCreek ( 1004083 ) on Monday October 27, 2008 @10:23AM (#25526927)
    Disk access has been slowing everything down.
    Variable elimination has been done, to varying extent, by multiple people here:
    https://bugs.launchpad.net/bugs/131094 [launchpad.net]
  • Most CPUs cant allow cores to run at independent speeds...
    On the other hand, AMD quad cores do, and i'm glad to have one core running full speed processing a single threaded program, and the other 3 cores as slow as possible to handle the background OS tasks..

  • by Khashishi ( 775369 ) on Monday October 27, 2008 @11:51AM (#25528233) Journal

    In other words, Ubuntus' not getting slower. The software that Ubuntu bundles is getting slower.

    Ubuntu is the software that it bundles.

  • Re:OH NOES!! (Score:2, Informative)

    by inaneframe ( 971456 ) on Monday October 27, 2008 @12:23PM (#25528855) Homepage
    Sounds like it has something to do with to whether an application takes advantage of multi-threaded enhancements or how it does.
  • by TheRaven64 ( 641858 ) on Monday October 27, 2008 @01:06PM (#25529673) Journal

    i seriously doubt anyone will say that OSX 10.0 runs slower on a 4 year old 1gb ram intel chip than OSX 1.5.whatever.

    Try it. Seriously. The 10.0 kernel had a significantly inferior VM subsystem (10.5 improved it a lot). 10.3 and 10.4 introduced more GPU-offloading in to the windowing system. Each version of OS X has been faster on the same hardware, although 10.4 and 10.5 have been more RAM-intensive.

  • by TheRaven64 ( 641858 ) on Monday October 27, 2008 @01:50PM (#25530455) Journal
    No, RAM intensive means that it needs more RAM. If you have 1GB of RAM then 10.5 will be faster than 10.0. If you have 256MB of RAM then 10.0 will be faster. Try doing a Spotlight search in 10.5 and again in 10.4. The version in 10.4 will be slower and the indexing will be more CPU-intensive while it runs. 10.5 actually does slightly better in constrained memory situations than 10.4, because of the rewritten VM subsystem. If you have enough RAM for both to run without swapping then the newer version will be faster. If you have enough RAM for only the older version to run without swapping then the older one will be faster. If you have enough RAM for neither to run without swapping then 10.5 will be faster if it's the newer one, otherwise the older one will be faster.

"If it ain't broke, don't fix it." - Bert Lantz

Working...