KDE

NAVIGATION
CATEGORIES
REFERRENCE
LINKS
  • What will be the targets for the WM in KDE4?

    13 answers - 589 bytes - related search similar search Add To My Delicious Add To My Stumble Upon Add To My Google Mark Add To My Facebook Add To My Digg Add To My Reddit

    Hi guys:!
    I have been looking into all the AIGLX stuff that gnome can use, and I was
    interested in hacking KWin with AIGLX or Compiz or Both ;-). I know there
    have been some developments on this front, but I wanted to know if there is
    any document describing the targets for KWin4.
    Silly question:
    My ISP is putting me behind a proxy, and svn is not working properly. How
    can I get the kwin_composite branch? Is there anywhere where I can get it
    from?
    Thanks,
    Arturo.
    Kwin mailing list
    Kwin (AT) kde (DOT) org
  • No.1 | | 327 bytes | |

    Tuesday 08 August 2006 11:38, Arturo Garcia wrote:
    Silly question:

    My ISP is putting me behind a proxy, and svn is not working properly.
    How can I get the kwin_composite branch? Is there anywhere where I can get
    it from?
    Answered it myself.

    Kwin mailing list
    Kwin (AT) kde (DOT) org
  • No.2 | | 1168 bytes | |

    Tuesday 08 August 2006 12:38, Arturo Garcia wrote:
    Hi guys:!

    I have been looking into all the AIGLX stuff that gnome can use, and I
    was interested in hacking KWin with AIGLX or Compiz or Both ;-). I know
    there have been some developments on this front, but I wanted to know if
    there is any document describing the targets for KWin4.

    The targets for KDE4 for KWin are, to put it that way, whatever somebody
    implements for it, more or less. Specifically as far as the compositing stuff
    is concerned, I suggest reading the "Compositing manager" thread in the
    archives ().

    In a nutshell, the only existing development in KWin so far is the
    kwin_composite branch which has only XRender-based compositing, done by me.
    Since I'm pretty lame when it gets to graphics the XRender stuff is of
    questionable quality and I definitely can't do the GL version. I made
    a "call to arms" in my blog and there are several people people subscribed
    here as a result of that, but I'm not aware of any progress. So whoever feels
    like becoming the (lead) developer of the compositing stuff, just raise your
    hand
  • No.3 | | 1181 bytes | |

    Tuesday 08 August 2006 15:53, Lubos Lunak wrote:
    In a nutshell, the only existing development in KWin so far is the
    kwin_composite branch which has only XRender-based compositing, done by me.
    Is this already KDE4? I was thinking on hacking it so it is available for
    kde3, and then porting on top of that. Is it possible to run KWin_v4 under
    kde3, given qt4 is installed? I thought the WM would be a bit independent
    from the Desktop Environment?

    Since I'm pretty lame when it gets to graphics the XRender stuff is of
    questionable quality and I definitely can't do the GL version. I made
    a "call to arms" in my blog and there are several people people subscribed
    here as a result of that, but I'm not aware of any progress. So whoever
    feels like becoming the (lead) developer of the compositing stuff, just
    raise your hand
    Well, we can try thing I was thinking was, as a first step, forward
    the calls to some copy-paste-code-from-compiz, and then do the programming
    ourselves But this needs deep discussion. First would be to set it up.

    Arturo.

    Kwin mailing list
    Kwin (AT) kde (DOT) org
  • No.4 | | 348 bytes | |

    Tuesday 08 August 2006 19:54, C Pinheiro wrote:
    I'm a bit busy with some workjob issues, but I'm fiddling with GL
    when time allows it. I'm trying to do something barely usable here to
    start to post something meaningful.
    You mean on Kwin? Kwin_composite?

    Kwin mailing list
    Kwin (AT) kde (DOT) org
  • No.5 | | 2470 bytes | |

    Tuesday 08 August 2006 17:49, Arturo Garcia wrote:
    Tuesday 08 August 2006 15:53, Lubos Lunak wrote:
    In a nutshell, the only existing development in KWin so far is the
    kwin_composite branch which has only XRender-based compositing, done by
    me.

    Is this already KDE4? I was thinking on hacking it so it is available for
    kde3, and then porting on top of that. Is it possible to run KWin_v4 under
    kde3, given qt4 is installed? I thought the WM would be a bit independent
    from the Desktop Environment?

    Yes, it is KDE4, and yes, the is a bit independent from the desktop. Not from
    the libraries though.

    The reasons for using KDE4 are basically the expected timeframe (I somehow
    don't expect KWin will have ready-to-use compositing by the next month)
    and ehm well it's the next version of KWin :). Porting either way
    when it's ready shouldn't be overly difficult, definitely simpler than with
    most other KDE components, if it'd be still worth it by that time.

    For building, you just need to build everything up to kdebase (where you
    basically can build only kwin and skip the rest). And KWin should run fine
    alone just from failsafe session and I think it should run fine even with
    KDE3, however it's recommended to do KWin development in a separate X session
    as described in the HACKING file. Not just for this case but generally, trust
    me you don't want to use a broken WM with your normal session.

    Since I'm pretty lame when it gets to graphics the XRender stuff is of
    questionable quality and I definitely can't do the GL version. I made
    a "call to arms" in my blog and there are several people people
    subscribed here as a result of that, but I'm not aware of any progress.
    So whoever feels like becoming the (lead) developer of the compositing
    stuff, just raise your hand

    Well, we can try thing I was thinking was, as a first step, forward
    the calls to some copy-paste-code-from-compiz, and then do the programming
    ourselves But this needs deep discussion. First would be to set it up.

    Well, whatever. Being the KWin maintainer I'll of course try to help as
    possible and try to review and whatnot, but given what's said above I'd
    really prefer not to do the actual development (because, if nothing else, I
    have already enough things to do that I'm not awfully bad at).
  • No.6 | | 519 bytes | |

    Tuesday 08 August 2006 20:54, C Pinheiro wrote:
    Hi Lubos and all,
    --
    I'm a bit busy with some workjob issues, but I'm fiddling with GL
    when time allows it. I'm trying to do something barely usable here to
    start to post something meaningful.

    Perhaps it'd make sense to post here even things that are not really
    usable, just so that the design can be discussed or that basics like setting
    up GL (or whatever, I don't know) don't have to be reinvented each time.
  • No.7 | | 1695 bytes | |

    Wednesday 09 August 2006 13:08, Lubos Lunak wrote:
    Yes, it is KDE4, and yes, the is a bit independent from the desktop. Not
    from the libraries though.

    The reasons for using KDE4 are basically the expected timeframe (I somehow
    don't expect KWin will have ready-to-use compositing by the next month)
    and ehm well it's the next version of KWin :). Porting either way
    when it's ready shouldn't be overly difficult, definitely simpler than with
    most other KDE components, if it'd be still worth it by that time.
    Well, I am of the opinion that if we hack the current KDE3 Kwin that would be
    better as we can can test and stabilise quicker Kwin with composite & AIGLX.
    After that we can port quickly. That would ease development at least for me.
    I wouldn't like at this point to develop for KDE4 until kdelibs stabilise.
    I'm a newbie on KDE development. At least I can compile kwin_kde3 now.

    Well, whatever. Being the KWin maintainer I'll of course try to help as
    possible and try to review and whatnot, but given what's said above I'd
    really prefer not to do the actual development (because, if nothing else, I
    have already enough things to do that I'm not awfully bad at).
    I will try, as a first step, to backport anything useful from kwin_composite
    so we can see "something happening" on KWIN_KDE3 (i'll find out if it's
    possible). If it goes alright, then I will achieve two things: understand
    kwin_composite, and understand kwin. I'll come back to the list with my
    findings

    Thanks!

    Kwin mailing list
    Kwin (AT) kde (DOT) org
  • No.8 | | 1526 bytes | |

    Wednesday 09 August 2006 17:54, Arturo Garcia wrote:
    Wednesday 09 August 2006 13:08, Lubos Lunak wrote:
    Yes, it is KDE4, and yes, the is a bit independent from the desktop. Not
    from the libraries though.

    The reasons for using KDE4 are basically the expected timeframe (I
    somehow don't expect KWin will have ready-to-use compositing by the next
    month) and ehm well it's the next version of KWin :). Porting
    either way when it's ready shouldn't be overly difficult, definitely
    simpler than with most other KDE components, if it'd be still worth it by
    that time.

    Well, I am of the opinion that if we hack the current KDE3 Kwin that would
    be better as we can can test and stabilise quicker Kwin with composite &
    AIGLX. After that we can port quickly.

    I don't think so:
    - kdelibs are somewhat stable (as in, it usually compiles), moreover kwin
    mostly depends only on kdecore -even doing updates should be simple
    - building KDE4 is not _that_ big deal
    - as experience shows, porting is a pain, especially for big things like this
    is presumably bound to be
    - syncing one more branch will probably be pain as well
    - kwin in KDE4 has already some changes that make it diverge from the KDE3
    version and it's going to happen more
    - updates of the requirements for kwin_composite branch can be reduced if
    deemed necessary - I don't think there would be a big problem with developing
    it with one month old kdelibs
  • No.9 | | 920 bytes | |

    Wednesday 09 August 2006 14:13, Lubos Lunak wrote:
    Tuesday 08 August 2006 20:54, C Pinheiro wrote:
    Hi Lubos and all,
    --
    I'm a bit busy with some workjob issues, but I'm fiddling with GL
    when time allows it. I'm trying to do something barely usable here to
    start to post something meaningful.

    Perhaps it'd make sense to post here even things that are not really
    usable, just so that the design can be discussed or that basics like
    setting up GL (or whatever, I don't know) don't have to be reinvented
    each time.

    Actually, thinking of this, it might be a good idea to create at least a very
    crude scene_opengl.cpp that'd just compose window they way they are on the
    screen, without any effects at all (roughly the same way like
    scene_basic.cpp). Probably shouldn't be that difficult when "borrowing" code
    from glcompmgr/compiz.
  • No.10 | | 1361 bytes | |

    kenal ( 10 august 2006 5:55 pm) kirjutas Lubos Lunak:
    Wednesday 09 August 2006 14:13, Lubos Lunak wrote:
    Tuesday 08 August 2006 20:54, C Pinheiro wrote:
    Hi Lubos and all,
    --
    I'm a bit busy with some workjob issues, but I'm fiddling with GL
    when time allows it. I'm trying to do something barely usable here to
    start to post something meaningful.

    Perhaps it'd make sense to post here even things that are not really
    usable, just so that the design can be discussed or that basics like
    setting up GL (or whatever, I don't know) don't have to be reinvented
    each time.

    Actually, thinking of this, it might be a good idea to create at least a
    very crude scene_opengl.cpp that'd just compose window they way they are on
    the screen, without any effects at all (roughly the same way like
    scene_basic.cpp). Probably shouldn't be that difficult when "borrowing"
    code from glcompmgr/compiz.

    Hi

    I just wanted to say that I'm still here and still interested with helping,
    just very busy right now.
    If someone would code the opengl scene stuff (i.e. opengl and compositioning
    initialization), then in about 2 weeks I could help with effects and perhaps
    other UI stuff.

    Rivo

    Kwin mailing list
    Kwin (AT) kde (DOT) org
  • No.11 | | 1033 bytes | |

    Thursday 10 August 2006 16:55, Lubos Lunak wrote:
    Wednesday 09 August 2006 14:13, Lubos Lunak wrote:
    Tuesday 08 August 2006 20:54, C Pinheiro wrote:
    Hi Lubos and all,
    --
    I'm a bit busy with some workjob issues, but I'm fiddling with GL
    when time allows it. I'm trying to do something barely usable here to
    start to post something meaningful.

    Perhaps it'd make sense to post here even things that are not really
    usable, just so that the design can be discussed or that basics like
    setting up GL (or whatever, I don't know) don't have to be reinvented
    each time.

    Actually, thinking of this, it might be a good idea to create at least a
    very crude scene_opengl.cpp that'd just compose window they way they are on
    the screen, without any effects at all (roughly the same way like
    scene_basic.cpp). Probably shouldn't be that difficult when "borrowing"
    code from glcompmgr/compiz.

    Let me rephrase: Who's volunteering to do this?
  • No.12 | | 1059 bytes | |

    Thursday 10 August 2006 14:21, Lubos Lunak wrote:
    I don't think so:
    - kdelibs are somewhat stable (as in, it usually compiles), moreover kwin
    mostly depends only on kdecore -even doing updates should be simple
    - building KDE4 is not _that_ big deal
    - as experience shows, porting is a pain, especially for big things like
    this is presumably bound to be
    - syncing one more branch will probably be pain as well
    - kwin in KDE4 has already some changes that make it diverge from the KDE3
    version and it's going to happen more
    - updates of the requirements for kwin_composite branch can be reduced if
    deemed necessary - I don't think there would be a big problem with
    developing it with one month old kdelibs
    , fair enough! I was about to start today, but you have convinced me. I
    will setup KDE4 (I need to donwload the source, compile, etc, etc.) and work
    straight on the branched kwin_composite.

    Thanks! Speak to you all soon.

    Kwin mailing list
    Kwin (AT) kde (DOT) org
  • No.13 | | 737 bytes | |

    kenal ( 10 august 2006 9:33 pm) kirjutas Taupter:
    Em Quinta 10 Agosto 2006 14:21, Lubos Lunak escreveu:
    Actually, thinking of this, it might be a good idea to create at least
    a very crude scene_opengl.cpp that'd just compose window they way they
    are on the screen, without any effects at all (roughly the same way
    like scene_basic.cpp). Probably shouldn't be that difficult when
    "borrowing" code from glcompmgr/compiz.

    Let me rephrase: Who's volunteering to do this?

    Me.

    Hi all, I'm back now.
    Is anyone working on opengl scene code now? Looking at the svn, I don't see
    anything new there

    Rivo

    Kwin mailing list
    Kwin (AT) kde (DOT) org

Re: What will be the targets for the WM in KDE4?


max 4000 letters.
Your nickname that display:
In order to stop the spam: 3 + 3 =
QUESTION ON "KDE"

EMSDN.COM