Ross Burton has written new entry on his personal blog with a good summary about X architecture these days. It’s a great reading if you just became lost between so many new extensions, different X servers and ways to hardware accelerated it.
Ross Burton has written new entry on his personal blog with a good summary about X architecture these days. It’s a great reading if you just became lost between so many new extensions, different X servers and ways to hardware accelerated it.
I wonder when such OpenGL extentions will become mainstream. So far I have only seen several hacks, it does not look like something usable for the everyday user will come out in the forseeable future.
Thanks for the link, Eugenia. =)
“Luminocity is pure crack. Pure, 100%, pristine crack, rolled into lines by naked Brazilian hand maidens.”
Now there’s a metaphor.
I’d imagine that as soon as drivers start working well with the new extensions it won’t take long until someone has some übercool idea for a neat composite manager and begins coding. At the very least we should start seeing some OS X clones pretty soon.
If I understood correctly from previous articles, by the end of this year we will already have this kind of technology on some GNU/Linux distributions, like Novell Linux Desktop 10, for example. GNOME 2.12, due in September, may also have something, but I’m not sure.
Yeah, I too vaguely remember Novell promised something like this in NLD 10, but I can’t seem to remember where I picked that up from.
No. If Novell wants to include some kind of Composition Manager, it’s their problem. It’s by now in a early experimental state. This technologies are going to be adopted slowly on the main desktops.
Gnome 2.12, due in September, will have Cairo as a backend if it uses GTK+ 2.8, so it may have high quality (antialiased) graphics for its widgets. But as a first version, I doubt it will have some kind of hardware acceleration even with Glitz as a backend of Cairo. Next version of GNOME (2.14 or 3.0 or whatever it may be, due in March 2006) will have Cairo as a mature technology and so it may start to use it to make some fancy graphics. Also, it may also include the first version of Luminocity as something experimental.
KDE has a composition manager in the actual 3.4 version, but it’s, again, highly experimental and unstable. Like GNOME, in the next version (3.5, due for sometime later this year)it won’t have a real stable composition manager and is going to use the same compmgr that’s using right now, so it will be (again) experimental.
In 2006 we will see KDE 4.0, based on QT4. It may have antialiased graphics like GNOME and some minor eye candy made in KDE, but once QT4 for Linux is mature enough, we will see some good fancy effects with (maybe) KDE 4.1 or KDE 4.2. For KDE 4.0 we will have a composition manager better than it’s today. Maybe at the same level as Luminocity, but I doubt it will be stable enough to everyday use.
Until Xorg matures to make Composite,GLX,etc stable and get support for the main manufacturers (that includes NVIDIA and ATI) we will not see something spectacular on the Linux desktop, and that’s not going to happen until very late 2006 or even 2007.
I saw this stuff (luminocity etc) demoed at the RedHat Summit last week because its going to go in Fedora soon.
MAN..OH MAN it blew my mind.
Its gonna put Linux desktop up there with OSX in the eye candy, smoothness, feel etc.
Getting support from ATI is worthless anyway. Much more likely to see composite support in the open source R300 driver than you ever are from ATI. I have a 9600XT in one of my boxes, and while their drivers are (finally) stable for me, feature set and speed are still pathetic – not even any 16-bit color modes, and switching to virtual terminals can cause a crash.
I’ve used some of these new X technologies on my other box (Geforce 6200), and it is already quite stable for me – there are problems when you start a 3D app, but that’s really not that hard to work around. Also, it’s much more important to get it working on Intel Integrated Graphics if you’re really talking marketshare, and the Luminocity guys have already demonstrated it working on that.
Where are links to videos of the latest and greatest related to xgl and luminocity, I can’t find anything.
Any Guadec links? Redhat Summit links?
After day 1, the guadec archive didn’t seem to get updated.
🙁
start talking to popular hardware maker and improve apt-get/red carpet approach for hardware support and software installation.
Here
http://gnomedesktop.org/node/2203
Are there any repositories for luminocity or xgl for Ubuntu?
I dont have the nerves and time to compile that stuff.
I wanna try this and not wait 5 years until it is called stable…
thx
.deb
start talking to popular hardware maker and improve apt-get/red carpet approach for hardware support and software installation.
What a stunning insight. You should write to some of them with your idea – I’m sure that none of them would have even considered it before.
Also, have you considered writing some articles for OSNews?
You could do one about how Beos would have a much greater market share if only more people used it. They lap that kind of stuff up round here.
At the moment some of the X guys are working on importing KAA into Xorg which should be a much better design for accelerating the things compositing managers need – at this point even though it’s not all going via GL and such it should be realistic to start enabling compositing on desktops by default and not have things be dog slow.
Are there any repositories for luminocity or xgl for Ubuntu?
I dont have the nerves and time to compile that stuff.
I wanna try this and not wait 5 years until it is called stable…
No. Luminocity requires compiling and risking borking your entire setup. I recently bought an old computer just to try out this stuff in Ubuntu (and not mess-up my main desktop). If I ever figure out how to do it, I’ll make a howto.
fine thx, good luck
Poofyhairguy:
http://www.ubuntulinux.org/wiki/LuminocityHowTo