Page 4 of 5 FirstFirst ... 2345 LastLast
Results 31 to 40 of 42

Thread: Does Compiz Still Slow Down Your System?

  1. #31
    Join Date
    Jun 2009
    Posts
    2,932

    Default

    I don't know about Nouveau, I played using r600c. Radeon drivers are in a better state than nouveau.

  2. #32
    Join Date
    Jan 2011
    Location
    Ohio, USA
    Posts
    13

    Default Slow down?

    I've read this all over the place, including on other Linux forums. "Compiz slows down your system" or "You can't play 3D games with Compiz." Who's coming up with this crap? I have 3 Linux systems installed on this computer, all of which run Gnome with Compiz enabled. I never have performance problems and I can play any games I want, including 3D games and I never notice a performance lag. Maybe it's my system, I am running a 1GB Nvidia graphics card and I have a dual core and 4 GB of system RAM, but I really don't know what all of the commotion about Compiz is. On my system, it works and it works well.

    Now, if you want to talk performance lags, KWin is the one that gives my system fits (mostly because KDE still doesn't play nice with Nvidia.)

  3. #33
    Join Date
    Jan 2009
    Location
    Columbus, OH, USA
    Posts
    323

    Default

    Interesting. I'd also be kind of interested in seeing the window manager tests expanded. Fluxbox, TWM, ratpoison, E17, etc.

  4. #34
    Join Date
    Jun 2006
    Posts
    311

    Default

    Quote Originally Posted by SciFiDude79 View Post
    I've read this all over the place, including on other Linux forums. "Compiz slows down your system" or "You can't play 3D games with Compiz." Who's coming up with this crap? ...
    It's a meme, meme transcend time and space. It may have been true once, but isn't true in all cases now. Unfortunately, meme removal is extremely painful and extremely slow...

    But Michael's intent of the article was to dispell these issues.

    Note that most of the talk now focuses on kwin's impact. The cycle continues...

  5. #35
    Join Date
    Jun 2006
    Posts
    311

    Default

    Quote Originally Posted by Wyatt View Post
    Interesting. I'd also be kind of interested in seeing the window manager tests expanded. Fluxbox, TWM, ratpoison, E17, etc.
    With PTS3/OpenBenchmarking, it's trivial to do.

    Configure your system to match the system configuration under test highlighted in the article (compiz/metacity). Repeat the results for your particular environment by running the test as shown below.

    Code:
    ./phoronix-test-suite benchmark 1103109-IV-COMPIZTES51
    That should give you a set of results that you can then compare to ensure that you have the same behaviour as Michael. Once you've got that confidence, your next step is to switch WM, re-run, etc. Note that the second question is really asking what the distinguished aspect of the system under test. You'd put in your WM there.

    Michael has started putting OpenBenchmarking.org links at the end of each article to let people play with the data. If your data is interesting, there might be a Phoronix review of the results, or a blog post on OpenBenchmarking.org

  6. #36
    Join Date
    Feb 2010
    Posts
    519

    Default

    Quote Originally Posted by pingufunkybeat View Post
    They can handle much more complicated stuff than Quake3, though:

    Doom3
    Quake4
    Penumbra games
    Amnesia

    These are not open source, but that's an issue with open source games, and not open source drivers.
    Actually the code for the first Penumbra got open source as part of the first Humble Indie Bundle hitting the $1M mark (if I remember correctly... anyway )
    Assets aren't free-as-in-freedom though this is true for Urabn Terror too, right?

    So a benchmarking tool for Overture could be a project

  7. #37
    Join Date
    Apr 2009
    Posts
    35

    Default

    I Had made benchmarks with openarena, on my radeon RV620 chipset, opensource r600g driver.

    (Archlinux with lastest mesa-git)

    and with compiz enabled, it's always slower.

    around 55fps without compiz (metacity without compositingÓ and 40fps with compiz enabled.
    Same for all games i had tried (Hon,alienarena,doom3,....)

    So, i really don't understand how phoronix can have twice more fps with compiz enabled.... (or maybe compositing is enabled in metacity)

  8. #38
    Join Date
    Dec 2009
    Posts
    40

    Default

    The news should have been called "Does Compiz Slow Down Your Games?"

  9. #39
    Join Date
    Jun 2009
    Posts
    2,932

    Default

    Quote Originally Posted by SciFiDude79 View Post
    I've read this all over the place, including on other Linux forums. "Compiz slows down your system" or "You can't play 3D games with Compiz." Who's coming up with this crap? I have 3 Linux systems installed on this computer, all of which run Gnome with Compiz enabled. I never have performance problems and I can play any games I want, including 3D games and I never notice a performance lag. Maybe it's my system, I am running a 1GB Nvidia graphics card and I have a dual core and 4 GB of system RAM, but I really don't know what all of the commotion about Compiz is. On my system, it works and it works well.
    Nvidia blob does not suffer much from the extra blits required, as these are very optimised. Don't extrapolate from your system to everyobody.

  10. #40
    Join Date
    Jun 2009
    Posts
    2,932

    Default

    Quote Originally Posted by spirit View Post
    So, i really don't understand how phoronix can have twice more fps with compiz enabled.... (or maybe compositing is enabled in metacity)
    That's what I'm thinking. The issue is not Compiz itself, but the compositing environment, and if Metacity also composits (it has had this functionality since at least 2008), then it's no wonder that it's not faster.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •