You are not logged in.
are you happy for new Xorg?
Offline
Absolutely no.
The support for vesa is broken (ok, it works, but at a speed that is not comparable with the performances of the computers of the last decade).
I decided to install the buggy ATI drivers, they work decently, but if I log out to KDM, instead of shutting the computer down, the computer freezes.
It was a big mistake to buy an ATI graphic card...
Offline
are you happy for new Xorg?
Yes, I am. My ATI Radeon 9000 works with Mesa drivers faster than before. I don't use ATI drivers long time ago.
Offline
I'm satisfied. I don't have any problems with new Xorg ,but i dont see any improvements in speed ,stability ,etc 7.1 just work like the 7.0 did.
I use Nvidia card with their binary drivers.
Offline
I'm satisfied. I don't have any problems with new Xorg ,but i dont see any improvements in speed ,stability ,etc 7.1 just work like the 7.0 did.
I use Nvidia card with their binary drivers.
Same here. Everything seems exactly like it did before. It works, so I'm happy.
Offline
are you happy for new Xorg?
Yes, it didn't break anything. I have GeForce 4 on my server and use nv driver.
to live is to die
Offline
Yes I am relly happy, my video card is now supported
Have you tried to turn it off and on again?
Offline
No visible differences with the 7.0 version, so I'm happy.
Offline
No visible differences with the 7.0 version, so I'm happy.
yep, same here. Hopefully nvidia will bring out the 9xxx driver which will contain the texture_from_pixmap extension to support aiglx :>
Offline
no. same experience as Ugo, and still no DVI support for matrox G cards. reverted to 7.0.0
Offline
yes, but I still get a XKB error on startup.
KISS = "It can scarcely be denied that the supreme goal of all theory is to make the irreducible basic elements as simple and as few as possible without having to surrender the adequate representation of a single datum of experience." - Albert Einstein
Offline
Yes, as things now feel more smooth when scrolling or dragging windows. I'm using a GeForce FX with the proprietary nvidia drivers.
Offline
ATI fglrx user, and I'm satisfied - no obvious problems here.
Offline
Xorg 7.1 has made a huge leap forward, not in speed or stability as much as in the feature set. Most of you may not have realized the great changes in xorg-server cos you don't use them. Like DRI/GLX accelerated screen rotation using libshadow in xorg-server.
Rotation was a nono until xorg-server 1.0.99 on most cards other than Nvidia. Don't know wether ATI supported this with the older version but anyway, it was a pretty hacky experience on i810 and siliconmotion. Now it works flawlessly.
Also the free ATI drivers for r300+ got a great boost for stability. Especially on ppc I can experience an improvement. My suspend-to-ram never locked up my machine but screwed up my screen until every pixel was refreshed properly by displayinjg new stuff. This among many other annoying little things have been fixed.
7.0 was the first modular release, now with 7.1 lots of bugs caused by the split have been wiped out. Be happy about it, even if it's not much of a different experience for the end user. Simply remember back when Xorg still was a monolithic beast, no release for more than half a year. Only custom patches that cause other things to break eventually. ![]()
I recognize that while theory and practice are, in theory, the same, they are, in practice, different. -Mark Mitchell
Offline
I presume you're talking about the recent upgrade to Xorg 7.1.1. If you are, I gotta say I'm not liking this new version at all. I'm pretty sure that something got broke in the mga driver this time around.
I'm having nothing but problems with mga driver and am forced to revert to the vesa driver to get by for now. The vesa driver's performance is less than ideal and rather resource heavy. Of course it's going to be. Am also stuck to a verticle refresh rate of 60Hz which isn't easy on the eyes and brain due to the flicker.
What happenned in my case, is that after doing and update with pacman -Syu, it upgraded Xorg. Only problem is that after update of Xorg to the newest version, X came up to a black screen upon next restart of X.
I think I know what is going on though. For some reason it's not setting the proper horizontal sync and verticle refresh. It's like it's totally ignoring my settings in the xorg.conf and trying to set them to something that is outside of what my monitor can handle. I spent all day yesterday experimenting with the xorg.conf to try to get it to work with the mga driver. I've tried everything from commenting out any hsync and vrefresh in case they were confusing Xorg's probing of them, to forcing conservative settings, to using Option "NoDDC". Nothing seems to work. The only way I can get the Xserver to come up without a black screen is to use a no frills driver like vesa.
I've been looking around the net and seems like all sorts of different matrox models are having problems with the newer version of Xorg and mga driver.
I know there is always going to be minor bugs, but seems nowadays too many Linux projects are released with major bugs. I tend to think that the issues that I and other matrox card owners have been experiencing could be constituted as major bugs. I don't really think a driver trying ignoring setting in the xorg.conf and trying to overdrive the monitor should be considered a minor one. Afterall, if you don't have a reliable display, then you really don't have squat.
Offline
kth5,
I understand what you're saying and I like seeing Xorg incorporating more modern features and support also. But when things are broken to the point that alot of users can't get the more basic and necessary functionality to work reliably, all those nice extra features become kind of moot point. Right now I would just be happy to be able to get the mga driver working. If I could get the other stuff working, that would of course be even greater.
Offline
kth5,
I understand what you're saying and I like seeing Xorg incorporating more modern features and support also. But when things are broken to the point that alot of users can't get the more basic and necessary functionality to work reliably, all those nice extra features become kind of moot point. Right now I would just be happy to be able to get the mga driver working. If I could get the other stuff working, that would of course be even greater.
I can't really second this on many computers I run. They are different architectures, just to name some:
iBook / Rage128 / r128
x86 / Nvidia GF 6200 / nvidia
powermac g3 / Radeon 9250 / radeon
x86 / Via Apollo CLE266 / via
x86 / Geode NX1 / nsc or vesa
x86 / Intel 855GME / i810
These machines all run stable, every one except for the Geode supports DRI well and stable. In all kinds of configurations. Xdmcp, XvMC (on i810 and nvidia) and rotation (except ATI) works perfectly. I have yet to find an issue.
As you can see, I don't have a gfx card that needs to be driven by mga. So I can't tell, but they did an upgrade on mga by 0.2 so I guess this could happen. Maybe it's after all just an issue with your specific chipset, as it always seems to be. Broken VBE's and BIOS software on motherboards cause most of the problems, Xorg drivers cannot handle properly as it seems. Like i810 and i945+ with non 4:3 setups...
[EDIT]
Do you run a flat panel or CRT? On DVI or VGA outputs?
[/EDIT]
I recognize that while theory and practice are, in theory, the same, they are, in practice, different. -Mark Mitchell
Offline
most of the problems seem to be from the crappy proprietary driver support offered for linux by the manufacturers. they've got to close source everything, like anyone is going to be interested in ripping off DVI code for a 10 year old matrox card.
they're getting much better, but one fatal error is all it takes, and then you're out of luck with no display.
it's kind of weird that vesa wouldn't work right, though. isn't that like the most "basic" driver for when all else fails that's pretty much supposed to work on everything? :?:
Offline
Xorg 7.1 has made a huge leap forward, not in speed or stability as much as in the feature set. Most of you may not have realized the great changes in xorg-server cos you don't use them. Like DRI/GLX accelerated screen rotation using libshadow in xorg-server.
Rotation was a nono until xorg-server 1.0.99 on most cards other than Nvidia. Don't know wether ATI supported this with the older version but anyway, it was a pretty hacky experience on i810 and siliconmotion. Now it works flawlessly.
sounds interesting. how do you use that stuff now?
Offline
Slackhack,
Not only is VESA basic, but it's a standard that's been around for ages. It's a standard that's probably older than Linux. Of course there have been revisions of it over time to accomidate newer functionality (i.e. there were different versions of vbe). But AFAIK, it's a very open standard.
The idea with VESA is that graphic cards support it (most do in one for another). It provides some basic 2D functionality. I believe where the problem stems is that many video card manufacturers have taken liberties in how much or how little they suport it (but I don't see this as being the explaination for the problems with the vesa driver in current xorg. I'll get to that later). I've noticed that from one card to another they all seem to have their own share of quirks when it comes to vesa support (I mean vesa outside of the scope of just Linux here). Like back in the day, many DOS games utilized and often required a video card to be vesa compliant. Back then we had SDD (Sci-tech's display doctor) to fill the void where cards poorly supported and implemented vesa in their bios. SDD was essentially a video bios that you load up into system memory. Remember TSRs? SDD was a TSR program.
You're pretty much right in the assessment that vesa is sort of the fallback standard these days. If a card can't utilized vesa for crap, then a person is stuck with super ugly vga as a last resort.
Matrox, has always been pretty good about supporting vesa. Alot better than most. When we use the mga driver for example, this is a driver written more specifically for certain mga graphic chipsets, whereas vesa is more of a generic means by which to utilize them. Moreso a generic means to utilize any vesa compliant video card. Vesa will lack alot of the extra functionality of course, but still I don't see how xorg could drop the ball on vesa. Usually the blame falls on the video card manufacturers and them being so closed about things. I don't see how this can be a good excuse when it comes vesa since it's an open standard.
Kth5,
I'm using a CRT connected through VGA.
Offline
In all kinds of configurations. Xdmcp, XvMC (on i810 and nvidia) and rotation (except ATI) works perfectly. I have yet to find an issue.
I thought XvMC was working on i810 and i815 but not newer Intel chips. Has this changed?
kth5 wrote:Rotation was a nono until xorg-server 1.0.99 on most cards other than Nvidia. Don't know wether ATI supported this with the older version but anyway, it was a pretty hacky experience on i810 and siliconmotion. Now it works flawlessly.
sounds interesting. how do you use that stuff now?
Try 'xrandr -o inverted' (and 'xrandr -o normal' to return to standard orientation).
Offline
7.1 is nice, but seems rather subject to annoying bugs. The feeling I get is that the FD.O people wanted to get it out as soon as possible.
(Then again, there are bugs in the Arch packages that I never saw with Basilburn's PKGBUILDs, so I'm not sure it's the server that's completely at fault.)
Offline
kth5 wrote:In all kinds of configurations. Xdmcp, XvMC (on i810 and nvidia) and rotation (except ATI) works perfectly. I have yet to find an issue.
I thought XvMC was working on i810 and i815 but not newer Intel chips. Has this changed?
I've been able to try this on 855GME and 850GM chipsets. Works fine here and they previously seemed to be using the i830 extensions to the driver.
Try 'xrandr -o inverted' (and 'xrandr -o normal' to return to standard orientation).
Yup, I usually put that in bashrc for my applications. It should also work with xorg.conf in the device section:
Option "Rotation" "CCW" # 90 degrees Counterclock wise
# Option "Rotation" "CW" # 90 degrees Clock wiseI recognize that while theory and practice are, in theory, the same, they are, in practice, different. -Mark Mitchell
Offline
kth5 wrote:baze wrote:kth5 wrote:Rotation was a nono until xorg-server 1.0.99 on most cards other than Nvidia. Don't know wether ATI supported this with the older version but anyway, it was a pretty hacky experience on i810 and siliconmotion. Now it works flawlessly.
sounds interesting. how do you use that stuff now?
Try 'xrandr -o inverted' (and 'xrandr -o normal' to return to standard orientation).
is it supposed to work while x is already running and are there any changes that need to be done to the xorg.conf?
when i try xrandr -o inverted, i getX Error of failed request: BadMatch (invalid parameter attributes) Major opcode of failed request: 156 (RANDR) Minor opcode of failed request: 2 (RRSetScreenConfig) Serial number of failed request: 12 Current serial number in output stream: 12
Offline
is it supposed to work while x is already running and are there any changes that need to be done to the xorg.conf?
It worked out of the box for me, no changes were needed in xorg.conf. Try running xrandr without arguments, it should tell which orientations are available:
$ xrandr
SZ: Pixels Physical Refresh
*0 1024 x 768 ( 306mm x 229mm ) *56
1 800 x 600 ( 306mm x 229mm ) 60
2 640 x 480 ( 306mm x 229mm ) 60
Current rotation - normal
Current reflection - none
Rotations possible - normal left inverted right
Reflections possible - noneOffline