The first milestone in the Xouvert roadmap has been reached. The first version of the new replacement X-server features patches from different places around the net and a cleanup to the default X-tree.
First Release of Xouvert is Out
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Are there any distributions planning to use it? (Fedora, hopefully)
I hope to see continued refactoring of the XFree86 source tree with an aim towards modernization. The official tree has been stagnating somewhat since the release of XFree86 4, and many more modern features have failed to see inclusion.
Although I must say, I’m much more excited about KDrive (http://www.pps.jussieu.fr/~jch/software/kdrive.html) as the future of desktop X11. It would be wonderful to see a SNAP backend implemented, which would open the doors to mainstream usage. I’d help in the development myself, but unfortunately I’m buried under a number of other projects.
Maybe, until freedesktop.org’s uber-cool X comes out at the end of next year or early 2005. 😉
I hope that Xouvert has incorporated most of these important patches that XFree86 guys were not accepting for so long for no reason at all (e.g. code and fixes from ATi).
Gentoo will probably have it in portage anytime soon.
Actually, I’d look for BreakMyGentoo to have the xouvert ebuild. Of course, no matter what, I’m not hacker enough to do anything to X. I’ll mess with things that use X, but not X.
What does Xouvert add to existing Xfree86 servers? I assume there isn’t compositing, or shadows/transparency.
Also, Freedesktop in 2005? I thought they were planning a much sooner release than that.
Is it another X Server? I guess it´s not the X Server from freedesktop.
Xouvert: XFree86 fork with some code cleanups and addition of patches that the xfree86 guys were snobbing.
freedesktop.org’s X: Re-write of the core of their server (not a fork), rewrite of some of the extenstions, while reusing some xfree86 code mostly for some other extensions and drivers, but overall a new thing.
Xouvert would be interesting to serve as the “middle man” towards the migration to fdo’s X.
Clearer now?
if this will keep compatiblity with the nvidia drivers? If so, then I’ll give it a try.
http://www.gentoo.org/news/en/gwn/20031124-newsletter.xml#doc_chap4
Quote from last post @ topic (URL to forum in above URL):
“overlay-freedesktop-20031208.tar.bz2”
8 december 2003. Hmm, that’s today.
http://dev.gentoo.org/~spyderous/overlay-freedesktop/
Quote from README:
“– For freedesktop.org xserver (formerly kdrive)”
Slap me, that’s about freedesktop.org, not Xouvert. I mixed them up. Though freedesktop.org will deliver true transparancy support (my main interest).
Anyone know what the performance / memory footprint of Xouvert is like compared to XFree?
Is there a page showing how this Xouvert release is different from Xfree? I see some mail about what they are planning, but what was accomplished in this release?
I remember when this project was first announced that it was going to target desktop usage … does this mean Remote X is no-longer possible?
Jon Smirl (IP: —.ma.charter.com) – Posted on 2003-12-09 02:11:36
Is there a page showing how this Xouvert release is different from Xfree? I see some mail about what they are planning, but what was accomplished in this release?
True, I was looking for the “road map” to see what their eventual aim is and in what time frame they hope to achive it in.
Archie (IP: —.bredband.skanova.com) – Posted on 2003-12-09 01:59:17
Anyone know what the performance / memory footprint of Xouvert is like compared to XFree?
What is wrong right now? the memory consumption is no worse than Windows XP, the only difference is that you actually get a proper read of out of the memory usage rather than the fabricated “information” that is provided by information window bought up via ctrl-alt-delete. The issue is now made worse that Windows XP now no longer flushes out applications that have been loaded into memory. This *MAY* mean faster start up times for applications that are re-used, however, the netresult is a consumption-a-thon by the operating system in the vein hope of increasing application loading times.
Run Windows XP Professional and see how easy it is to push the memory usage to 84MB and for it to just sit there, even after all applications are closed off.
http://xouvert.org/docs/modularising-xfree86.html
It it kinda like a road map, he makes some very valid points and hopefully with a more open development process, we will have more developers willing to jump in and help out.
“What is wrong right now?”
The author never implied that there was anything wrong with the performance of XFree.
I am also curious as to the performance of Xouvert compared to XFree since it includes features that XFree did not include. Perhaps some reasons where regarding performance.
Do the new features of Xouvert outweight any performance hit they might incur over XFree? That is, I believe, what the author was asking.
The current Xouvert tree is almost identical to the current XFree86 tree. The change is a minor patch to allow it to install without interfering with a current XFree86 installation.
All of the work so far has been some wild speculation on the mailing lists, while Jonathon and a few others have been working to organize and tag the tree to put it into the arch repository.
Xouvert will not break your current X, and will not be incompatible with current XFree86. It’s mostly a place to get a stable release that includes patches excluded from XFree86 for various reasons. Xouvert plans for a release iteration roughly twice as fast as XFree86, meaning overall cleanups and patches going in and becomeing stable in 6 months instead of 1-2 years.
Xouvert also intends on making it easy to build X, and easy to hack on, in order to increase the number of competant X developers.
Ok so Xouvert is a fork of X, but is it compatible with X, will applications that run on X work on Xouvert too without modification?
Also, since this is a fork, it must be released under a license compatible with that of X and so the improvements will be available. Will Xfree86 merge the improvements from Xouvert? Is Xouvert just like Fedora for Redhat, a place where you can try new things you wouldn’t dare with the main product and than if they work well and get good feedback incorporate them into the main product?
Someone please explain this.
X is a standard. If it is not compatible, then it is not X anymore. So no, it won’t break applications.
Xouvert is a branch. Check the Xouvert FAQ for more information http://www.xouvert.org/faq.html#2.1
This release of xouvert is simply xfree 4.3,
So how come you call it xouvert release?
isn’t that sort of credit stealing?
I can’t wait to emerge this and ditch legacy Xfree. And yes, I like Xfree, but I want more, more releases, more often, more updates, more new features, more, more, more!
uhm won’t more features require protocol compatibility? i don’t think they are willing to step outside the lines here (unless of course the X protocol defines features not implemented by XF86…). I think the best we can hope for is cleaned up code with newer patches, a better developer community around thats more effecient. Not that I’m complaining but don’t expect 3D-desktops and the like just yet.
anyone try this out yet? any performance figures? do the nvidia drives still work (i assume they should since they are kernel modules not X patches bu t you never know)…
What does Xouvert mean? How does one pronounce it?
Come on, give http://www.xouvert.org a little visit to find your answer. It’s right there.
Ouvert is the french word for “open”. It’s pronounced ex ooh verr. Silent T, rolled r
Pronounce: http://www.xouvert.org/faq.html#1.2
Meaning: wild guess: it is just a name
Ok, It’s getting on my wick peopel asking all these Q’s about what enhancements the first Xouvert release provides and wanted to clear some stuff up.
1. This is a development release, which basically means, not for general usage, even though it is JUST Xfree 4.3.0 + patches “arched”.
Basically the intention of this release is so hacking can begin on all the nicieties.
2. this IS Xfree, for now, no added little bells + whistles, but things are being discussed on the ML, like adding PNG/JPG/SVG support to the server, cleaning up/changing the build system etc. (see the ML for more details)
3. As said before, this is Xfree, so as so, mem footprints are the same.
4. Expect to see innovation/progress in the next release. this is just a base to begin hacking/patching/experimenting.
I hope this cleared most Q’s up, if you have any more, feel free to ask, and i will try and answer. Just so people know, i’m not a dev of the project, but have been watching closly from a long time ago (about 1-2 months before the /.’ing).
cmf
Forgot Something…..
This is still compatible with X11, given it is an X11 implementation. So no worries of X/Y program not working.
Also, the driver system is still the same as Xfree’s, so no worries of Nvidia drivers not working!
Ok so Xouvert is a fork of X, but is it compatible with X, will applications that run on X work on Xouvert too without modification?
As it has been mentioned already, X is a standard. Xouvert is just a different implementation of that standard.
Ouvert is the french word for “open”. It’s pronounced ex ooh verr. Silent T, rolled r
Actually it’s pronounced “Zoo-vaire”
————————————————————–
This is good for *nix. Xfree has always been much slower at development than any other project. I don’t really have a problem with the current state of Xfree because I don’t really need it to do anything other than what it already does BUT innovation and faster development can’t be a bad thing. Xserver and Xouvert should speed up desktop development which is sorely needed for those who need/want all the bells and whistles similar to OSX’s Quartz and the proposed Longhorn display.
he issue is now made worse that Windows XP now no longer flushes out applications that have been loaded into memory. This *MAY* mean faster start up times for applications that are re-used, however, the netresult is a consumption-a-thon by the operating system in the vein hope of increasing application loading times.
As long as that memory is freed up when needed, what’s the problem ?
Free memory is wasted memory.