X.Org 6.8.2 is Out

ertz writes "The X.Org Foundation today announced the fourth release of the X Window System since the formation of the Foundation in January of 2004. The new X.Org release, called X Window System Version 11, Release 6.8.2 (X11R6.8.2) builds on the work of X.org X11R6.8.0 and X11R6.8.1 released in 2004. X11R6.8.2 combines the latest developments from many people and companies working with the X Window System and an open X.Org Foundation Release Team. All Official X.Org Releases are available for download from the ftp site and at mirror-sites world-wide."

So is Xfree86 dead?

[glrotate]

Is it being actively maintained or developed?

Re:So is Xfree86 dead?

[Quattro Vezina]

I believe it's still being developed, but most distributions are no longer packaging it.

Re:So is Xfree86 dead?

[RealAlaskan]

Is it being actively maintained or developed?

Well, if nothing has changed since the fork, the answer is probably: ``Not really.'' Wasn't the glacial pace and control-freak policies of Xfree the reason for the fork in the first place?

Re:So is Xfree86 dead?

[nofx_3]

Wasn't the glacial pace and control-freak policies of Xfree the reason for the fork in the first place?

Actually IIRC much of the reason for the fork was due to a license change that many groups/people thought was too restrictive and incompatible with the popular OSS licencies (GPL/BSD/APACHE etc...)

-kaplanfx

Re:So is Xfree86 dead?

[RealAlaskan]

Actually IIRC much of the reason for the fork was due to a license change that many groups/people thought was too restrictive and incompatible with the popular OSS licencies (GPL/BSD/APACHE etc...)

I remember that, and I agree it was the straw that broke the camel's back, but I also recall that there had been long-standing, wide-spread dissatisfaction with the pace of development and the access to the process.

I was exaggerating when I said that Xfree isn't being developed; it still seems to be lumbering along at about the same old pace. I think that the pace at which x.org is moving will have nearly as much to do with its success as the new, improved (actually, same old?) license.

Re:So is Xfree86 dead?

[eatjello]

not dead, just forked. xfree86 is still in active development, but is currently under a feature freeze so they can concentrate on cleaning up the code (something xfree very much needs).

Re:So is Xfree86 dead?

[iabervon]

It is being maintained and developed as actively as it ever was, which is to say that there's a first release candidate for the first version after the split ready now. Whether that counts as "active" or not depends on your point of view.

Re:So is Xfree86 dead?

[Just Some Guy]

X.org was largely formed because the people who most actively wanted to contribute to XFree86 weren't allowed to. Honestly, I don't think that there's really any left who could or would want to actively continue XFree86. On the other hand, X.org is being sponsored by HP, Sun, and IBM, and has representatives from SUSE, HP, Sun, and Red Hat on its board of directors.

No, the real question is whether anyone qualified to continue X development is still sticking with XFree86. From what I can see, the answer is no.

Re:So is Xfree86 dead?

[eschasi]

One could get contributions by demonstrating that there was an actual need. Thus far, it appears to be wishful thinking on the part of the developers. IMHO it would be much more effective approach for HP, Sun, et al to approach the card and chip vendors and twist their arms to release better specs and drivers under a mutually acceptable open source license. The end result would be much better cards for the $ than what we're likely to get from OGP.

Re:So is Xfree86 dead?

[Chandon Seldon]

In some alternate universe where the only use for accelerated 3d was gaming, your post would make tons of sense.

Here in the real world, hardware accelerated 3d is an important capibility for everything from CAD to basic 2d desktop rendering.

The requirement for 3d hardware acceleration for general usage applications is becoming more and more widespread. Already features that were only avaiable in high-end 3d cards in 1995 are now required to get a reasonable user experiance out of both Windows XP and Mac OS X - I wouldn't assume that modern Open Source desktop environments won't use the same techniques to keep up.

Re:So is Xfree86 dead?

[Citizen of Earth]

Is it being actively maintained or developed?

If a tree falls in the forest and nobody hears it...?

Debian?

[Anonymous Coward]

Anyone in the know know why Debian is sticking to a fork of the old XFree code, and not moving to x.org like other distros?

Re:Debian?

[bogaboga]

Debian is very conservative in upgrades. I understand that it is why Debian is very stable too. They (Debian) wait for the early adopters (Mandrake et al...) to see and iron out the bugs. Why are you anonymous?

Re:Debian?

[stefanlasiewski]

Debian is very conservative in upgrades. I understand that it is why Debian is very stable too.

But isn't that why we have Stable, Testing and Unstable? "Stable" should be conservative in upgrades, Testing and Unstable are for incorporating new software into the future products?

Maybe they are just way more convservative then I realized :)

Debian/unstable

[phorm]

Yes, I'm very dissappointed in Debian/unstable for this. Certainly many other packages are available in unstable, up to CVS and bleeding-edge upgrades. But no X.org.

I've had some nasty things happen with package dependencies breaking in unstable, so I'm fairly sure they're not holding off because of that.

Re:Debian/unstable

[plumpy]

The problem is that right now Debian is preparing for a new 'stable' release (Sarge), so there aren't any huge new upgrades like X allowed to go in.

Once sarge releases, things will hopefully go back to the craziness we all love.

You can get a lot of newer packages out of the new(er) 'experimental' repository, but X.org isn't in there.

Re:Debian/unstable

[Kethinov]

You can compile X.org in Debian just fine. I've done it. Not that it really matters. The hardware accelerated features that X.org is developing are still too new and unstable to be usable at any reasonable level. I'm really not going to care one way or another if Debian gets X.org until GNOME and KDE start doing useful things with cario, glitz, damage, and whatnot. Granted, work on this has indeed started, but until it is usable at at least a beta level, Debian need not waste their time rushing to adopt the X fork. They will. Give them time. The rush work that needs to be done is not at the distro level but at the development level. I want to see GNOME and KDE using nifty hardware accelerated effects. Real transparency built into the terminal and an Expose clone. Then distros can start worrying about packaging X.org.

Re:Debian?

[node 3]

Anyone in the know know why Debian is sticking to a fork of the old XFree code, and not moving to x.org like other distros?

Because it's not sticking to a fork of the old XFree86 code, and it is moving to X.Org, like other distros.

Re:Debian?

[stefanlasiewski]

Because it's not sticking to a fork of the old XFree86 code, and it is moving to X.Org, like other distros.

I think many of us are wondering when x.org will be added to Debian unstable. It's nowhere in the package lists for stable, testing or unstable [debian.org].

I can't find any recent news regarding X.org for Debian [google.com] on Debian.org, or in the FAQ [debian.org]. I see some discussion and debate about 7 months ago, but many things have changed since then.

Re:Debian?

[FreeUser]

Debian will move to Xorg the day Duke Nukem Forever is released.

I see you've been playing with the Duke Nukem beta as well.

The experimental version of Debian that will be running xorg REQUIRES Duke Nukem Forever to boot. It runs as a client process, accessible by shooting the bouncer at the alien-infested oyster bar (from the opening screen take 2 lefts, a right, a left, and straight on three blocks. It'll be on your right.) Killing the bouncer will fork off a process that boostraps the new Debian, from which point you can fire up xorg either using startx or by defaulting to runlevel 5.

Re:Debian?

[Torham]

See the FAQ:

http://necrotic.deadbeast.net/xsf/XFree86/trunk/de bian/local/FAQ.xhtml#debianplans [deadbeast.net]

Re:Debian?

[abrotman]

http://necrotic.deadbeast.net/xsf/XFree86/trunk/de bian/local/FAQ.xhtml#debianplans

This explains some of what is going on with debian. Sarge(when released) will not have X.org packages.

The relationship between unstable and testing

[steveha]

I always hear about Sarge, but what about the unstable branch of Debian ? I still don't get why Sarge would require unstable to be frozen like that - and in fact, it isn't the case for most of the other packages.

It's very simple.

Debian has three major branches: stable, testing, and unstable.

The stable branch is treated very carefully. It will get security patches, but otherwise will not be changed. It's a "frozen" release. Most Debian users will run the stable branch on their servers.

The testing and unstable branches work together and are closely related. The unstable branch is where new packages are checked in. Once the new package has been in unstable for a while and is working out well, it will be auto-migrated into the testing branch.

And this is the answer to your question: Debian cannot update the unstable branch to X.org without cutting off the testing branch from further updates, or risking that X.org packages might get migrated into sarge by the scripts that update testing. Why would the Debian guys make more work for themselves by doing this?

All three branches have "code names". The unstable branch is code-named "sid", always. The testing branch is currently code-named "sarge". When sarge is "released", what will happen? First, the current stable branch (code-named "woody") will be retired from the main servers. Second, the servers will be updated to have the sarge packages listed as the stable branch. Third, a new code name will be chosen for the next release, and the testing branch will be named with that code name. (At that exact moment, I guess the testing branch will be identical to the stable branch, but that won't be true for long.) Finally, all the checkins that were held back, waiting for the release of sarge, will start to flood into unstable; this is when you can expect to see X.org in unstable.

Actually there is a fourth branch of Debian: experimental. You will really see X.org show up in experimental before it even shows up in unstable. Once people have good success with the packages in experimental, the packages will be checked in to unstable. (Just because it is called "unstable" doesn't mean that Debian is completely careless.)

The problem is that the expected date release of Sarge was pushed back over and over.

This is just Debian for you. Debian is a loose coalition of volunteers, and their sole goal is to put out a distribution that will be rock solid. They ship "when it's done", not according to some schedule.

Note that there is any reason you cannot use sarge now. Why wait? It's already very stable. I used to use unstable on my desktops, and that was stable enough for me; testing should be even more stable.

The Debian X Strike Force was IMHO quite slow at reacting to the upcoming of XOrg.

The X Strike Force is not a large team, it has a lot of work to do, and what you think of it doesn't really change anything. If you join the X Strike Force and help them get their work done, then I will listen attentively to your opinions, and until then, I'll gently suggest you not complain of their slowness.

If you want to combine the Debian goodness with the X.org exciting new flavor, I have two suggestions for you.

First, you can read the discussion here [debian.net] about how to compile your own X.org from the CVS, and set that up on your Debian system. It works so well there is "no need for packages", according to that discussion.

Also, if you would like everything that is good about Debian but with faster release cycles, you ought to look into Ubuntu [ubuntu-linux.org]. Ubuntu is committed to a new version every six months, and their next release (due to release in April 2005) already has X.org checked in. I'm using that to type this message. It's definitely not as stable as the released version of Ubuntu from October 2004 but I can deal with it and I like th

What about even beyond experimental

[kjj]

I really thing there need to be a fifth branch that even goes way beyond experimental. A tree where just about anything will show up literally a day or two after its release. It could be called the god branch or something.

If you really want X.org in debian:

[gehrehmee]

1. Add:
   Package: *
   Pin: release a=hoary
   Pin-Priority: 50
   
   to /etc/apt/preferences.
2. Add:
   # Ubuntu Hoary
   deb http://archive.ubuntu.com/ubuntu/ hoary main restricted universe multiverse
   to /etc/apt/source.list.
3. Run:
   apt-get update
   apt-get install grep-dctrl
   cat status | grep-dctrl xorg -F Source -s Package | perl -e 'while (<>) { print; print "Pin: release a=hoary\nPin-Priority: 1000\n\n" }' >> /etc/apt/preferences
4. Move the paragraph from step 1 to the end of the preferences file. (not sure if this is actually neccesary)
5. Run: apt-get upgrade
6. Hope it all works well. Did I mention I don't take responsibility for any breakage this causes? :)

When all is said and done, these steps tell apt where to get ubuntu packages from, then tell it not to install any them, then tells them to make an exception for the Xorg packages, treating them just like they were regular debian packages.

You'll also have the option of installing any software in ubuntu that's not in debian yet, and all of the potential breakage that implies.

Ati Drivers

[espergreen]

I wonder if Ati users will have to wait another 6 months to get 6.8.2 support.

Re:Ati Drivers

[MarcQuadra]

I wonder when Linux users will stop buying hardware that doesn't have published interfaces.

I also wonder when people with ATI card that are actually supported will realize it. My RADEON 9200 and 7500 get full 3D acceleration without the closed drivers.

Re:Ati Drivers

[arose]

Not full, I had to get a GeForce to replace my RADEON 7000 because the maximum point size was 1 and that made Wings 3D very difficult. Notify me if they ever fix that.

Re:Ati Drivers

[Anonymous Coward]

>> I wonder when Linux users will stop buying hardware that doesn't have published interfaces.>>

I wonder when some Linux users will stop being so arrogant. Many people come to Linux AFTER they have purchased an ATI card with a desktop or notebook.

"Switch to Linux it's better."
"Okay. Reformat hard drive, install, configure. Hey, i can't get my ATI card to work."
"You are so stupid. Why didn't you buy a card that works with Linux?"

Re:Ati Drivers

[orasio]

Ubuntu can use your card.
You can try the live version.

About 3d, you need ATI to release a driver.
They did publish specs in the past, so there were drivers up to 9200, but then they stopped.

The problem you talk about is not so. There are some 3d video cards, and some winmodems that are not supported, but most hardware is.
Proprietary drivers have problems of their own, too.
I have some hardware that does run on the Linux kernel, and doesn't on newer versions of win (aimslabs tv tuner, samsung webcam) because

Re:Ati Drivers

[xoboots]

" If there aren't open source drivers that support hardware then we will never be independent of microsoft."

How do you mean? If there are open specifications for driver iterfaces won't that accomplish the same goal? Think of it from the manufacturer's point of view. They already have to conform their drivers to Microsoft's interfaces. You don't just knock off drivers over night, either -- its a big job that is easy to get wrong. So they spend their time making sure that things work on Windows because that'

Re:Ati Drivers

[GlassHeart]

anyone who seriously wants to switch, should not bitch about it after the fact.

What exactly constitutes "seriously wants to switch"? Why should a newbie commit to Linux before his old video card can even work? What you're asking for is bordering on religion.

In this world, you tell somebody that Linux is better, they believe you, try it out, and then make a commitment.

They should just cut their losses and buy an NVIDIA card.

No, in many cases "cutting their losses" means ejecting the live Linux CD, an

Re:Ati Drivers

[jusdisgi]

Absolutely. Even my IGP mobile athlon works perfectly (w/ accel) in 6.8.1 without screwing with anything or getting special drivers.

Which card instead?

[tepples]

I wonder when Linux users will stop buying hardware that doesn't have published interfaces.

As soon as you tell us what to buy instead. Other than NVIDIA and ATI, neither of which publishes a full register level spec, which video chipsets are available as consumer level video cards sold in Best Buy stores or as part of a notebook computer? Or do you expect us all to buy X11 thin clients instead of video cards?

Re:Which card instead?

[LourensV]

Help is on the way. TechSource [techsource.com], which makes video cards for air traffic control and medical applications has started work on a desktop/workstation card that will be able to accelerate enough of X.org and OpenGL to be usable. It won't blow your socks off in gaming, but for many users that's not important anyway. More importantly, the specs will be fully open. More information on OpenGraphics.org [opengraphics.org].

Oh and yes, it's been mentioned on Slashdot [slashdot.org] before.

Join us and help out!

Re:Ati Drivers

[xoboots]

"I wonder when Linux users will stop buying hardware that doesn't have published interfaces.
"

Can you suggest an affordable, modern, consumer grade performance video card that meets this criteria? No you can't because there aren't any and you know it.

Mostly stability

[Anonymous Coward]

It looks like its moslty a stability update than features:
The X11R6.8.2 release is intended to be a stable bug fix release ("Maintenance update") for the X11R6.8.0 and X11R6.8.1 X11 releases of the Xorg Foundation, containing bug fixes, security updates and a small set of new features, which include the following:

* ATI R100 video driver
* ATI "radeon" video driver
* ATI Rage128 video driver
* CYGWIN infrastructure update
* DMX Library updates
* Intel i810 video driver
* libXpm security update (CAN-2004-0914)
* Mesa (OpenGL) update to release 6.2
* Fixes to the pseudocolor emulation layer (currently only used by the Neomagic driver.)
* "nv" (Nvidia) video driver
* Postscript print driver
* Xprint infrastructure update

Re:Mostly stability

[ajs]

So here's the important question: has the "nv" driver gained access to enough of the modern NVidia cards that I can stop using the binary-only driver to play Neverwinter Nights?

Re:Mostly stability

[DarkSarin]

Not bloody likely.

Seriously, the nv drivers suck at 3D largely because nvidia does not open their specs sufficiently. I, along with everyone else, would like them to, but personally, I think that's a ways out. They release a good driver though, and I am happy with it.

Re:Mostly stability

[ZephyrXero]

We need open hardware [kerneltrap.org]

Re:Mostly stability

[Grishnakh]

Huh? Who needs fabs? Most companies designing chips don't actually manufacture them; they send them to contract manufacturers who do. With open hardware, you'd have to pay to have a batch of chips made, but it's far less than it'd cost to build a fab.

Ever design circuit boards? I have. I tried making some by hand with rub-off stuff from Radio Shack and ferric chloride, and it was a PITA. Then I found www.custompcb.com in Malaysia which sent me two 4x5.5" boards for $26, including shipping. $26 is a lot cheaper than it would have cost me to set up my own photolithography and etching lab.

Debian

[ArsonSmith]

I wish they'd release Sarge already so that Xorg will go into unstable.

Re:Debian

[Anonymous Coward]

Get Ubuntu instead. It's essentially a cleaned up version of Debian Sid and it includes X.org.

Re:Debian

[InfiniteWisdom]

Instead of wishing, go here [debian.org] and help it happen by fixing as many of the 93 release-critical bugs as you can. Once that number is low enough, Sarge will be released.

Re:Debian

[mabinogi]

As opposed to Windows where all you can do is wait.

Change log?

[Anonymous Coward]

This announcement means nothing without a changelog

version numbers

[Quasar1999]

Okay, what's with the crazy version numbers? Can we not have some universal version numbering system... where if more than say 10% of the API is updated then make it a major number change... I mean... how long has it been at version 6? Since 2001???

Version 6

[hummassa]

the X windowing system is at Version 11 release 6 (X11R6) since the early 1990's.

Re:version numbers

[iabervon]

May 16, 1994, actually. But they haven't removed anything from the API.

Re:version numbers

[temojen]

It implements the 6th revision of the 11th X-windowing system specification. Hence it's X11R6, but Version 11 of the X-Windowing system is the one that's a standard, so they can drop the 11. The result: revision 6. The 8.2 is the implementation version.

Re:version numbers

[commodoresloat]

Maybe he's thinking of 1995.

Marijuana's a hell of a drug.

YAY!

[zoloto]

This is actually a little faster on my 500mhz intel system! And who said code get's slower as it ages?

Re:YAY!

[Anonymous Coward]

Not me. Every new version of OSX that has come out has made my old Mac run faster.

Torrent?

[darkmeridian]

Can someone set up a torrent at www.mininova.org? It is an open-tracker and well-populated.

Someone should have done this before we slashdotted their server.

Re:Torrent?

[TheSurfer]

As you wish: click [mininova.org] :)

NetBSD

[Coneasfast]

when will netbsd switch to xorg for its official X.

i know they have no problem with the new XFree86 license, but there are other reasons. Xorg is the new de facto standard. it has more features, cleaner code, and the best xfree86 developers have moved to xorg. xfree86 will soon be obsolete, it's time they switch.

what's holding them back? they can still keep xfree86 on as an alternative too.

Re:NetBSD

[agent dero]

`cd /usr/pkgsrc/meta-pkgs/xorg; setenv X11_TYPE xorg; make install clean`

It's been in pkgsrc for a while, and works fine :)

Re:NetBSD

[agent dero]

The reason it's most likely not in base yet, is because X.org has problems on other problems. The older version of X in the base system has been hacked up a great deal to really make sure it'll work on as many platforms as possible.

If it ain't broke....

The attitude seems to be that, if you want the newer features of X.org, it's not a problem to upgrade, otherwise, the older X is 'good enough'

Re:NetBSD

[Coneasfast]

The older version of X in the base system has been hacked up a great deal to really make sure it'll work on as many platforms as possible.

hmm, i must ask, have these 'hacks' that are required been reported/merged with xorg tree? has there been any collaboration efforts? or has it been modified for quite a while that it would take a lot of work to merge back?

Just a guess...

[temojen]

NetBSD's big claim to fame is portability.... It takes a while to fully test a large package on that many architectures and implementations (Chances are it takes a really long time to compile on a VAX 11/780, for example).

Re:NetBSD

[Mr.Ned]

Just for the record, xorg has been in pkgsrc for quite a while. It's easy to use - simply add 'X11_TYPE=xorg' to /etc/mk.conf and then pkg_add or 'make' new applications. You can remove /usr/X11 at your leisure.

Mirror

[TorrentNinja]

here is a mirror http://mirrordot.org/stories/63d72f637743d71a6cfd0 1e6d95be129/index.html [mirrordot.org]

Changelog

[BinLadenMyHero]

Dear Taco,

Please post a link to a summary of changes [x.org] when anouncing the release of a new version of any software.

Re:Changelog

[Anonymous Coward]

X.org is getting hammered by /. so here is the summary of changes:

2. Summary of new features in X11R6.8.2

This is a sampling of the new features in X11R6.8.2. A more complete list of changes can be found in the ChangeLog file that is part of the X source tree.

* Freetype was updated to version 2.1.8. But installing freetype from X distributions would often or usually result in the replacement or use of "stale" versions of freetype. On Linux, FreeBSD, Solaris 10, and SCO5, therefore, X11R6.8.2 will by default use the version of freetype2 that is installed on the system. If your system doesn't come with an installed freetype2 and you wish to use the version supplied with this distribution, please add: #define HasFreetype2 NO to config/cf/host.def.
* The XTT font module and FreeType1 have been retired because FreeType2 subsumes their functionality
* Retire old PS Type1 font rasterizer (except for CID font usage) since the FreeType2 rasterizer now handles PS Type1 (*.pfa, *.pfb)
* Render implementation fixes
* Updated x86emu and resynced with upstream at Scitech
* Updated SiS driver
* Updated Nvidia driver (opensource version)
* Render acceleration for ATI's R100 and R200-series cards
* Substantial speedups in the software implementation of the render extensions when compiled with gcc 3.4 on the i386 architecture.
* Infrastructure for rotation support in drivers
* New Trapezoid specification for the Render extension
o Respecify Render to include only 'normal' traps
o Allow backward compatibility but internally covert to new format
* Software mouse cursor is now based on the Damage extension
* A new keyboard driver is enabled by default. The old driver is disabled unless explicitly compiled in by defining the macro |USE_DEPRECATED_KEYBOARD_DRIVER|.
* All extensions (except Xserver-specific extensions "DMX" and "XpExtension") can now be enabled/disabled from the configuration file and from the command line.
* Mac OS X updates:
o Support dynamic screen configuration changes in rootless mode
o Added option to always use Mac command key equivalents
o Interpret scroll wheel mouse events correctly when shift is held down
o Added trivial Xinput support
o Fixed launch of X clients from Finder with a space in their path
o Fixed some GLX rendering problems on Mac OS X 10.2 and earlier
* Updated xterm version

2.1. New X extensions

X11R6.8.2 includes four new extensions:

XFixes

The XFixes extension is a collection of improvements for deficiencies in the core protocol, including

* Notification when a selection changes
* Server-side region objects
* Allow clients to track the cursor image

Damage

The Damage extension allows a client to be notified whenever something is drawn to a window. This feature is useful for VNC servers, for screen magnifiers, and for clients using the Composite extension to update the screen.
Composite (experimental)

The Composite extension allows a client to request that all drawing to window is redirected to off-screen buffer. Though the Damage extension the client, called a 'compositing manager', can know which areas of a window is modified and render the windows on screen. By making use of the drawing requests from both the core protocol and the RENDER extension, the compositing manager can create special effects, such as translucennt windows.

The Composite extension is considered experimental in X11R6.8.2 and is turned off by default.
XEvIE (X Event Interception Extension) (experimental)

XEvIE is an extension to intercept core keyboard and pointing device input It allows consumation, modification or synthesis of input events before these are sent to their final destination (i.e., interested clients). This feature is required by the GNOME accessibility project.

The XEvIE extension is considered experimental

All I want

[suso]

is for the nvidia (from nvidia) driver to not lockup under Gentoo. Unfortunately, I guess nobody has found the reason for that happening yet.

Re:All I want

[arkhan_jg]

Had a similar problem myself.

Update to a moderately recent version of portage, (2.0.51 iirc)

and add
>=media-video/nvidia-kernel-1.0.6629
>=medi a -video/nvidia-glx-1.0.6629

to your /etc/portage/package.mask file, then "emerge nvidia-kernel nvidia-glx"

That should drop you down to the 6111 driver, which is far more stable than the piece of crap 6629. Reboot, or kill your window manager, rmmod nvidia, and modprobe the new one and things should be a lot more stable.

I've gone from locking up X two or thre

Try disabling SBA and FW

[linuxkrn]

I have a supermicro (7505 chipset) dual Xeon board that supports both SBA and FW. However, when I try to enable SBA/FW I get lockups. I fixed this by disabling them in the modules config. /etc/modules.d/nvidia options nvidia NVreg_EnableAGPSBA=0 NVreg_EnableAGPFW=0 Then run modules-update and reload the nvidia module. You may have to reboot for it to take effect. One thing to note is the new drivers DIDN'T default to off as the older ones had. So some people are seeing fixes just because they downgrad

Re:Mine is perfect

[suso]

Its not just me though (as this thread shows), if you look through the gentoo forums for this problem you will see that some people have the problem and others don't. People who have the same nvidia drivers will have differing results. Its a very strange and hard to track down problem.

X.org

[Sunspire]

Forget about the X.org website, it's worthless. If you want to see what's changed in 6.8.2, turn to the release notes over at Freedesktop.org [freedesktop.org].

duh

[BinLadenMyHero]

It's exactly the same [x.org] in X.org, as I pointed before [slashdot.org].

PCI-Express and X86-64 fixes

[Anonymous Coward]

I bought parts for a new PC which arrived on Monday, and while setting up SuSE 9.2 I discovered a slight ... incompatibility. If you're using a PCI-Express NVidia card on x86-64, things may prove somewhat problematic with X.org 6.8.1. In my case, the graphical installer simply wouldn't run, and after installing via VNC I couldn't get the proprietary NVidia driver to work without serious corruption of the mouse pointer, missing text and so on.

I had been concerned that I'd have to switch off stuff like dual-head, hardware acceleration etc., but it turns out it's a (now fixed) bug in X.org regarding PIC-Express and 64-bit Linux on AMD processors. I was downloading some semi-official 6.8.2 packages just before seeing this...

So, if you're having problems with X on a spangly new system, I hope this helps. ;-)

Re:PCI-Express and X86-64 fixes

[Ford Prefect]

... and, it works! Now I can escape from the login-forgetting w3m...

Currently I've only got single-head going, but everything's gloriously accelerated, including OpenGL. Even the open nv driver now works for the basics. Hurrah!

If you want the SuSE 9.2 RPMs for 6.8.2 RC3, read this [suse.com]. You'll need to manually copy a SaX2-produced XF86Config to xorg.conf unless you upgrade SaX2 as well - I haven't found pre-compiled x86-64 RPMs for that yet, so I think I'm going to have to compile some myself.

Ah, life on the

Re:PCI-Express and X86-64 fixes

[njchick]

The free "nv" driver is single head only, but you can choose which head to use by using the CrtcNumber option. To support dual head, the driver should call xf86SetEntitySharable(), which it doesn't do yet, perhaps for a good reason.

New freetype

[m50d]

Now maybe some of the "Linux fonts are awful" trolls will stop.

Re:New freetype

[sloanster]

Now maybe some of the "Linux fonts are awful" trolls will stop.

No, they will never stop. Even though linux fonts have been totally sweet for some time now, the fudsters will forever gripe about the 1993-era slackware distro that they once saw, as if it were still relevant.

Re:New freetype

[Eneff]

Mepis for one.

(I swapped them out for the Microsoft fonts, which I prefer.)

Section Linux?

[devphaeton]

I guess the BSDs still don't count on /.

Re:Section Linux?

[Quattro Vezina]

No, it's just quid pro quo for the KDE 3.4 Beta 2 article being posted under the BSD section.

Why is this under "Linux"?

[MondoMor]

This applies to a broad range of OSes. It has very little to do with Linux directly.

Re:Why is this under "Linux"?

[Jason Hood]

This applies to a broad range of OSes.

Very true.


It has very little to do with Linux directly.

False, linux is by far the largest user platform for xorg.

XFree

[LittleLebowskiUrbanA]

What happened to those guys? David Dawes' crusade finally just peter out? Does anyone else still use XFree?

Xgl

[Doc Ruby]

How about Xgl [nat.org], the port of X to OpenGL HW/SW?

Re:Xgl

[Anonymous Coward]

Gotta love a post about people missing the point that completely misses the point.

Re:Xgl

[Doc Ruby]

Er, you've missed the point. The point is that his X server is calling OpenGL, for all its rendering. So the HW can do all kinds of special effects, like piping scaled windows around for better representation of related contexts. Quibbling about beta features like dropshadow differentiation is really just sour grapes. I worked at Apple for a while; I know how tempting it is to complain when someone else furthers a technique Apple pioneered, or even just pioneered in promoting. If Apple were publishing GPL OpenGL X versions that run on other OS'es than OSX, there might be something to complain about. But not really - then we'd be happy to have some competition to keep things moving forward.

Re:Xgl

[Doc Ruby]

No, you haven't read the description of Xgl to which I linked. Friedman wrote

"Want live, running thumbnailed versions of iconified windows? Done. Want your six virtual desktops to be the six faces of a cube that spins, with lighting?"

That's a lot more than eyecandy, a lot more than dropshadows. That's the beginning of a newly interactive media desktop. The rest of the OpenGL features for X are left as an exercise for the imagination for the next several years.

Linux?

[fm6]

This should have been put in the BSD section as well, for obvious reasons. I'd add the Unix section, except there isn't one. Come to think of it, wouldn't it make a lot of sense to have a single section for Linux and Unix (including BSD)? The distinction between Linux and Unix is more legal than technical.

About the whole X Window server

[mnmn]

I've used Windows, MacOSX, various commercial unices, linux and bsd, and have come to a conclusion, that the X server's client-server design compromises the latency of usage.

I thought this was a driver issue, for example, on the same machine, opening, moving and resizing windows is very snappy on any window system beside X, be it MSWindows (yeah I know crappy, insecure, bloaty etc, but snappy), BeOS or OSX. Even the X11 on Darwin isnt quite as snappy as it should be being a GUI system.

In the case of BeOS,

Re:About the whole X Window server

[ratboy666]

On the current crop of uni-processor systems, X latency may be an issue. Not a big one, though... Seems worst with OS that is not "HT" aware running on a Xeon.

But, multi-CPU is the "wave of the future". On such systems, X will render on its own processor, and will seem to perform BETTER than the alternatives. X Windows "sweet spot" has always been either (1) at least 2 CPUs or (2) a dedicated X terminal with high-speed link.

As to "X-lite", the networking is already bypassed if running on the same box.

Rat

"Client server is slow" myth dispelled, once more

[DrWhizBang]

With all due respect, you don't really seem to know much on this subject. As I understand it, it is not the client-server model, nor the inefficiency of the X protocol that is at issue here. It is rather a the stagnancy of the x toolkit (which could be blamed on the xfree86 organization, if one likes to point fingers) that has caused this.

As has been stated in another thread, X11R6 was first released in 1994. No significant changes were made to its drawing libraries before the addition or the render extension (with anti-aliased fonts) by Keith Packard in the 2001/2002 timeframe. In 1994, things that we take for granted like true-colour displays. Windows 95 had not been released - Windows 3.1 was mostly seen in 256 colours!

As more graphical applications (e.g. web browsers, image viewers) became the norm, and 32-bit colour became common, application writers sought solutions that would allow them the functionality they needed. GTK+ and QT became toolkits that supplied the features that X lacked, at the cost or having to perform client side rendering. This pushes more and more pixels with higher bit-depths through the X protocol to the server. Some solutions were devised for special cases like OpenGL (GLX) and video (Xvideo), but X's core display system did without updates.

Since the clients now had to push lots of bits through the X protocol to the server, 2D graphics displayed the latency that you describe, even on really fast hardware. In a way, the Render extension seems to have pushed this over the edge since software fallbacks required (esp. for text) made rendering crawl.

The solution that the X.org guys have come up with is this: reduce the reading and writing over the X "pipe". There are a few methods that they are using. First is the XFixes extension. This extension supplies some additional functions that were missing in the core protocol - like the ability to address a region. Once this was in place, the Damage extension could be created, which allows the client and server to pass less information back in forth because they can now identify when a region has been damaged and needs to be redrawn.

The next piece is Composite and the composite manager. Composite allows the server to draw windows into an offscreen region so that the composite manager can redraw them on the screen. By doing this, the composite manager can use the hardware acceleration in the video card to do smooth opaque moves, and additionaly special effects. Theoretically, a composite manager could be written to use OpenGL, which would be really smooth. I can testify, however, that using Composite and xcompmgr on my PC at home is smooth as glass. 32 bit colour, drop shadows, and all the niceties...

The next step will be Cairo, Glitz, and XGL. I am anxiously waiting for a release of this stuff, because it is way cool.

The window manager is the problem

[spitzak]

Client/server is not the problem. In many cases it can be much faster because it encourages batching together hundreds or thousands of requests into a single context switch.

The problem with X is much simpler, but nobody wants to hear it. The problem is the design of having a seperate process that is the "window manager".

Anybody who has used X for many years will know that the problems with moving and resizing windows have remained pretty much constant despite the fact that the machines themselves have inc

Xdmx Sounds VERY Cool

[eno2001]

It allows multihead over the network. This might just be the solution I've been looking for to make all the systems in my house display the same instance of an app on the screen at the same on multiple machines. Imagine running am IM app only once, but seeing it on any machine you happen to be working on. Multicast apps! :)

Good news for IBM X40 owners

[asv108]

i810 driver updates * Dualhead support (i830+) * i915 support * New 3D driver (i830+) * i810 driver is now supported for AMD64 This should fix the current hackish method for using video out.

I am waiting for the *next* release..

[wowbagger]

I am waiting on the *next* release, which, as I understand it, will have the GATOS drivers for the various ATI All-in-wonder cards merged in by default.

What that means is that, out of the box, the ATI Radeon 7500 All-in-wonder will have accelerated 3D, video capture, and TV tuner support.

The current CVS image has these built in, but as I understood it, they did not want to merge that into this release.

Of course, for the newer cards (R300 and later) the 3D will still require the ATI FGLRX driver, which do

I'd like to see two pointers, two focuses

[DunbarTheInept]

What with USB working the way it does, where you can chain off as many devices as you feel like, and computers being fast enough to handle all of them at once, it seems to be like it should be possible to do the following:

Three Users, user zero, one, and two, are sitting in a conference room using a giant screen projector as the monitor, attached to a laptop someone brought. There are three different keyboards and three different mice attached to the laptop as USB devices. Some might even be IR so they are being used from across the room.

User zero picks up keyboard 0 and mouse 0, uses mouse zero to click on a terminal window and focus it, then uses keyboard 0 to type into it.

Meanwhile User one sits at keyboard 1 and mouse 1 to demonstrate something on the web using a browser window.

Meanwhile User two, using keyboard 2 and mouse 2, is making a diagram in openoffice.

Essentailly, there are three different "input contexts" each one consisting of one mouse and one keyboard, and each has its own mouse pointer, and it's own keyboard focus, and the X server is interleaving thier input events together and dispatching them to the appropriate applications.

The place where I would have found such a thing useful was a roleplaying game where I had a lot of visual aids on computer, one of which was a map with little tokens players could move to represent themselves on the map (each token was a layer in Gimp) It would have been handy to have public mice for them and my private mouse for me to use on the private GM screen (the laptop's own screen).

But, it doesn't seem to be possible without writing it myself.... ....hmmm..... I wonder how one goes about learning the X input system....

Re:X.org, openoffice.org

[TheRaven64]

OpenOffice.org is used because OpenOffice it trademarked (by someone other than OpenOffice.org). X.org is probably used because X by itself is often used as an abbreviation for the X Windowing System, rather than a specific implementation.

Re:X.org, openoffice.org

[temojen]

Because X.org comes from the X organization and if they just called it X, people might confuse it with other implementations of the X11R6 specification.

Re:Windows driven Linux

[creysoft]

Windows does not drive video card development. Games drive video card development, and they only drive it one direction: forward. The bigger problem is that PCs have such a wide variety of video cards, ranging from high-end to low-end, external to built-in, and they all differ in various ways. The way you do something on one card may not be the same way you do it on another.

Apple doesn't have this problem, because Quartz Extreme supports a finite set of graphics cards, and Apple computers all ship with com

Re:Windows driven Linux

[TheRaven64]

X does network transparency at too low a level. For general windowed applications, better performance can be achieved by running the controls on the client side and only rendering controls not present on the client system. NeWS did something like this - the controls were all interpreted and ran on the client display. I believe Windows remote desktop does this, and similar functionality is planned for GNUstep, using their distributed objects implementation.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Arbitrary resolution and refresh rate

[erikharrison]

I'll answer this, just FYI, though you probably won't actually see it.

The problem is the refresh rate of your Video Card. Since X.org has little in the way of card specs, the result is that we have to do stuff like ask the BIOS what modes it can set the card to. The BIOS rarely has all the resolutions and refresh rates that the card can handle, but since we don't know which registers to twiddle and bits to flip, we have to ask the BIOS to do all the work. We simply cannot make the card do any other resolut