It has been one year and four days since X.Org 7.3 was released and a number of months since X.Org 7.4 was supposed to be released, but today X.Org 7.4 is scheduled to finally make it out the door! This release is shipping quite late and with a slimmed down set of features, but in this article we have more details on what this release holds in store for the Linux desktop community and why it may be a short-lived release.
Unfortunately the most exciting technologies that were supposed to be released for Xorg 7.4 were cut from the final release. This is a real disappointment. DRI2, glucose, and randr 1.3 were going to be real visible improvements but we are left with little improvement in the end users eyes with the actual release of 7.4. I have been personally waiting for DRI2 because I have an Intel chip and I use Compiz. EXA perfomance sucks and XAA cannot use XV output without a special plugin for Compiz and a patch for mplayer and then it only works with mplayer. Glucose was going to fill in the gaps between AIGLX and XGL but there hasn’t even been any mention of it lately.
Then you’ll probably be pleased to know that intel wants to release a special release this year of xserver 1.6 (xorg 7.4 comes with xserver 1.5). That release will have the neat intel features coming up uxa,gem and dri2.
Also as a bonus if Xinput2 is ready in time it will be included, allowing for multipointer X (which is already merged in the main branch).
But generally it is quite annoying that Xorg releases keep getting worse (delays AND feature cuts). It would be nice if linux based companies invested more in X.
May be a problem with finding quality software engineers. I know Novell/Suse are looking for people and having a tough time finding applicants to work on file systems. May not be the companies at fault, just lack of qualified people.
I think the problem has more to do with waiting for the companies with closed drivers to catch up. Even with all the advances that got cut from 7.4, the nvidia legacy drivers still don’t support the ABI changes to the xserver-1.5 and their non-legacy driver barely makes the cut. And this issue has been known for quite a while. Not sure about drivers elsewhere…
I just got a new motherboard. The graphics chipset came out about 3 months ago. And I’m running the latest development version of my preferred distro, which has Xorg 7.4 and the very latest point release of the video drivers. Not only does 3d not work… neither does 2d. In fact, even during the text based installation of the OS the screen is strangely wrapped vertically and flickers at about 10 Hz. The vesa driver gets me a sort of usable screen, but not at the correct aspect ratio. The evil proprietary company that can’t keep up with our open source speed of development and released this chipset without making sure that it had working drivers for Linux? Intel. It’s a G43 chipset with X4500 video. I’m having to use my Nvidia 7200SE with nvidia’s drivers, which performs flawlessly, until it all gets sorted out.
My gripe is that even when the companies do everything we ask: Release good docs… supply the development infrastructure… assign engineers to assist with development… our FOSS video drivers are still crap for some reason. In 2008 we should at least be able to get text mode right!
Edited 2008-09-10 17:43 UTC
Its a shame your so full of crap. I’ve been wanting a G45 for like forever. As far as I am aware support for x4500 stuff has been out before the chipset launch see the linux intel site for the news. What has muddied the water however is is this GEM/TTM where TTM support has been stripped before GEM has been implemented if your going what eh!? your probably running a stable kernel and openarena a lot faster. Which means if you are cutting edge your running slower with the “Failed to initialize TTM buffer manager. Falling back to classic.” Of course you have never had much luck
with hardware I guess its because you can’t say that laptop that for forced at gunpoint on you doesn’t work anymore, although I am pleased that your employers are so generous.
Like I say its a shame I would love to know if x4500 chip esp the more serious G45 really is the 1.7 times faster that it is advertised or whether it is marketing hype. Seriously you are such a lying bitch.
As for Nvidia, shit you lying lying toerag. I can’t think of another way of putting it. I’ll tell you that f–king chipset holds back every damn release of every distro ever, and leaves security holes for years. The reality is on a proprietary system provides a enough graphical grunt with an integrated solution to run fear at minimal setting for about ~A‘^Alb35 on a providing a cheap proprietary system but it f–ks ibex in the ass.
Seriously sort it out your hating so I can get some real information.
Yes. It’s out but it’s quite broken. The X server segfaults with a backtrace going into even 2D.
(Irrelevant drivel trimmed.)
Whether it is memory management or not, it immediately segfaults.
1. It’s a desktop.
2. It’s my desktop.
3. I’ve semi-retired. Not that my old employer ever would have bought anyone a laptop.
The marketing hype is actually 300% as fast as 3100.
I Just left this quote in to highlight what a skilled debater you are, Cyclops. Not your best work, though. I thought your “slimy little toady” post was quite outstanding:
http://www.osnews.com/thread?272915
Anyway, like I say, the Intel FOSS driver segfaults immediately when X first starts, and my Nvidia with the proprietary drivers performs perfectly. It’s the FOSS driver holding the distro release back and not NVidia’s driver.
Yeah, right. I’ll work on my “hating”. Surely Cyclops harbors none in his heart. :-0
Edited 2008-09-10 21:00 UTC
now you don’t want to be posting old posts, now do you.
BTW, someone actually did get an X4500HD board running for a period in July. OpenArena got about 30 fps at 1024×768 and undisclosed quality settings.
Edited 2008-09-10 21:20 UTC
If you were serious and I suspect your not. I would be astonished. I get 40+ at 1600×1200 now without TTM on a X3500. Your just a player, it sounds like a lie, so it probably is one.
Believe it.
http://tinyurl.com/6qpmtl
http://tinyurl.com/58ejvd
And there is no way that your X3x00 is getting 40 fps in OpenArena at 1600×1200. So that’s just a filthy rotten lie you noxious, egg-laying, hermaphroditic, smarmy, ablutophobic, and generally unpleasant person!
tell me where to post a screenshot and of which level.
00:00.0 Host bridge: Intel Corporation 965 G1 Memory Controller Hub (rev 03)
00:02.0 VGA compatible controller: Intel Corporation 965 G1 Integrated Graphics Controller (rev 03)
00:02.1 Display controller: Intel Corporation Device 2983 (rev 03)
00:1a.0 USB Controller: Intel Corporation USB UHCI Controller #4 (rev 02)
00:1a.1 USB Controller: Intel Corporation USB UHCI Controller #5 (rev 02)
00:1a.2 USB Controller: Intel Corporation USB UHCI Controller #6 (rev 02)
00:1a.7 USB Controller: Intel Corporation USB2 EHCI Controller #2 (rev 02)
00:1b.0 Audio device: Intel Corporation HD Audio Controller (rev 02)
00:1c.0 PCI bridge: Intel Corporation PCI Express Port 1 (rev 02)
00:1c.1 PCI bridge: Intel Corporation PCI Express Port 2 (rev 02)
00:1c.4 PCI bridge: Intel Corporation PCI Express Port 5 (rev 02)
00:1d.0 USB Controller: Intel Corporation USB UHCI Controller #1 (rev 02)
00:1d.1 USB Controller: Intel Corporation USB UHCI Controller #2 (rev 02)
00:1d.2 USB Controller: Intel Corporation USB UHCI Controller #3 (rev 02)
00:1d.7 USB Controller: Intel Corporation USB2 EHCI Controller #1 (rev 02)
00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev 92)
00:1f.0 ISA bridge: Intel Corporation LPC Interface Controller (rev 02)
00:1f.2 SATA controller: Intel Corporation 6 port SATA AHCI Controller (rev 02)
00:1f.3 SMBus: Intel Corporation SMBus Controller (rev 02)
01:00.0 IDE interface: JMicron Technologies, Inc. JMB368 IDE controller
02:00.0 Multimedia video controller: Micronas Semiconductor Holding AG Device 0720
04:03.0 FireWire (IEEE 1394): VIA Technologies, Inc. IEEE 1394 Host Controller (rev c0)
any more setails of what I’m running I will happily provide
http://img165.imageshack.us/my.php?image=shot0001tgaoi1.jpg
http://pastebin.com/d661451ba
Oh I’m sorry did I stutter
oh and my glxinfo with the TTM error
http://pastebin.com/d741b012f
Linux localhost 2.6.27-rc5y #5 SMP PREEMPT Wed Sep 10 19:09:58 GMT 2008 x86_64 Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz GenuineIntel GNU/Linux
You see the bottom line sbergman27 I don’t need to lie about anything. the ~40 with a genuine 1.7 speed increase now would give allow me vsync on my monitor. I suspect GEM will give me more as well. You might not like what I say but then you wouldn’t.
Cyclops, you don’t fool me for a minute. You just transferred that screen shot over to your Vista box and touched up the FPS number in PhotoShop.
LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL LOL
Burn
Seriously, Cyclops. I owe you an apology. I’ve already PM’d you more details. But late last night I got my X4500 going not only in 2D but also in 3D. (Note: Using the DVI port was causing the segfault. VGA port works.) There are still some issues. But OpenArena at “Very High Quality” and 1680×1050 does about 32 fps in a typical timedemo of the first level. (Although they do drop to 16 FPS for an instant here and there.) “Normal” quality and 1440×900 plays great and looks fantastic other than some mouse and keyboard delay weirdness.
So although you surprised me with your 40 FPS claim for the X3500 (or any IGP for that matter), now that I have seen the X4500 in action, it seems quite believable. And I must say, this changes my view of IGPs.
-Steve
Thats what I thought, Jesus, time and time again you look a retard and you keep coming back for more. What is wrong with you. I suspect you wont be posting links to this comment either
LOL… you better watch out. You’re treading on thin ice. Soon enough Cyclops will label you as a dreaded “Vista User.”
LOL is funny how I mention the word subterfuge you come running. many use very wear that badge with pride in their nick.
Of course I normally post under one of my many other accounts.
The problem is, these things take a while, so while the driver has been out, your distro (I’m guessing ubuntu) hasn’t integrated it yet.
I agree that there should be a better infrastructure to deliver opensource graphics card updates, but don’t blame others for a problem that is basically “distro didn’t update it yet”.
Also, the 7200SE is kinda old, so if intel made discrete graphics cards, your argument would sound something like “It’s a G43 chipset with X4500 video. I’m having to use my G965 with intel’s drivers, which performs flawlessly, until it all gets sorted out.”
I remember when a friend bought a brand new 8800GT and there was no nvidia driver for like 1 or 2 months.
The bottom line is: updates take a while to get to most distros. Don’t blame intel (they’ve been doing an excellent work) for that.
Replying to self is kinda lame, but just wanted to add that back in July, phoronix.com was already reviewing a X4500HD under linux:
http://www.phoronix.com/scan.php?page=article&item=intel_x4500hd&nu…
and if you read the first paragraph you even have another link:
http://www.phoronix.com/scan.php?page=news_item&px=NjUzMw
about support being commited in mid-june!
They also mention what I mentioned: you have to compile some stuff from sources, because it’s not packaged yet. But the code is there.
Did you even read my post all the way through? I explicitly stated that the development version of the distro, which I am using, has the latest stable code from upstream. The problem is that months after the release of the boards with the chipset, Intel’s Linux driver is *broken*. I even compiled it myself from the upstream code. Same segfault. So don’t be so quick to blame the user.
Turns out, it does work, with some fiddling, as long as you don’t try to do anything fancy like use the DVI port instead of the analog VGA port. How many Windows users today, do you think, got home with their new MB only to find out that plugging a cable into the DVI port makes the graphics driver immediately crash?
Part of my irritation comes from disappointment at going so far out of my way to buy Intel instead of something else and then having the worst graphics experience I’ve had in years. Intel has certainly not lived up to expectations in this case.
But also, part of my reason for bringing all this up is that, as you say, moving to an alpha version of one’s distro just to get working SATA and IGP *is* a suboptimal situation to say the least.
And by the way, 5 days of fiddling *did* get me some pretty impressive performance for an IGP. I like fiddling. So overall its been worth it and I am pleased with my purchase. But still, not every potential G43 purchaser/Linux user would be pleased with Intel or Linux at this point.
G965 performs flawlessly? You must be kidding. I got my laptop for 5 months already and can’t even play Warcraft III on it. Opened a bug report at freedesktop 4 months ago and it was confirmed by Intel developer.
Today, the bug is still not fixed. I blame Intel.
It would be nice if linux based companies invested more in X.
That’s exactly the problem. Development is so fast these days that whole technology get dropped or totally revamped before anyone can stabilize them enough to do a release (see DRI2).
Well GPUs are evolving pretty fast so it’s not a surprise that X has a hard time catching up.
Note this isn’t necessarily the full story though: Linux sound’s system is still unstable even though the hardware mostly stopped evolving many years ago.
I’ve never had stability problems with sound in Linux. In fact, most of the complaints I’ve heard about ALSA tend to be about design decisions (the fact that it doesn’t use /dev files except as a part of OSS emulation) rather than its actual stability.
Well, you’re lucky but PulseAudio is very recent and some users had issue with it (and before PA there were other): so like I said the sound system is still evolving even though the hardware has mostly stopped evolving years ago, which is an indication that there is a problem.
Then you’ll probably be pleased to know that intel wants to release a special release this year of xserver 1.6 (xorg 7.4 comes with xserver 1.5). That release will have the neat intel features coming up uxa,gem and dri2.
I hope this is the case but I don’t exactly trust Xorg release schedules. I wouldn’t be surprised if we have to wait 6 months to a year until the next version is released, or we will get another intermediate release that lacks many of the features that were promised for xorg 7.4.
I second that. I am getting sick of all the empty promises of the xorg team. For the last 3 releases they have promised that certain features will be included and then when the time came the features were delayed … WTF only MS is allowed to do that
True, it’s so bad I hear they are considering offering a 30% discount on this release.
A typical enough response. But a stagnant X could easily cost FOSS 30% of our potential user base. I’m not sure Xorg is advancing any faster than XFree86 did under Lord Dawes.
You got to be kidding.
Wicked! I agree with this sentiment 100%.
If people are so upset that xorg has dropped some features, then maybe those people should volunteer. I’m sure that xorg needs beta testers, documentation writers, developers, christ donate some cash. Obviously they need developers the most, but every little bit helps.
Development is never free, even FOSS development.
The way you tell it makes it sound as though finished features were cut for no good reason. More accurately, these were unfinished features, buggy features, or features that were finished but dependent on unfinished or buggy features. They were left out because they weren’t ready for prime time. Period. If the xorg / xserver folks are guilty of anything here, it is perhaps being a bit too optimistic last year! The fact is these guys have done a stupendous job updating and modernizing X for every day use. I can’t wait to implement xorg 7.4 and xserver 1.5 on my Debian system. Great job guys!!!
Great job? How is an incredibly late, feature-cut release great? If MS did something like this people would be howling at the moon. I appreciate the work that has been done but something has to change. The Xorg team has been over promising and under delivering. We need either more realistic time frames or better management. I don’t think we have to cheer the developers just because they managed to ship something they deemed “stable”, while offering practially none of the benefits touted even 6 months ago nevermind the fact that it was already at least 6 months late then.
If you don’t like it, help it. You didn’t help, so you don’t have the right to whine about it (well you do but it doesn’t matter really what you say).
This is OSS, not a closed source company. If you PAYED for it then you have “some” right to whine, but not for a free OSS software you don’t.
All projects out there need more people who GIVE BACK, something you obviously never heard about.
If you don’t like it, help it. You didn’t help, so you don’t have the right to whine about it (well you do but it doesn’t matter really what you say).
This is OSS, not a closed source company. If you PAYED for it then you have “some” right to whine, but not for a free OSS software you don’t.
All projects out there need more people who GIVE BACK, something you obviously never heard about.
BS. How can you argue that I don’t give back when you have NO idea? I do give back, but that’s not even the point. Criticism isn’t whining. Maybe you enjoy being misled but I don’t. There is no reason for either OSS or proprietary software to promise so much and then under deliver several months late regulary. Something is broken. X releases have been problematic forever and someone has to get it straightened out. Projects like GNOME don’t seem to have these kinds of problems. GNOME generally has a roadmap that they follow very closely. If something isn’t finished in time they postpone the release, but it is never for very long because they haven’t promised more than they can deliver. This is acceptable. Xorg releases are not.
It’s not like I’m bitching about every little bug that I enounter, demanding that my problems be taken care of first. I just want to know “Where’s the beef?”. This release went way over schedule because of several new features being added that then weren’t added at all and it still took a year! I don’t see how anyone can be happy with that even X maintainers. It’s a huge disappointment and anyone who says otherwise is in denial.
This is where I step in to give my opinion, no matter if anyone is interested in hearing it or not.
X releases have been problematic forever and someone has to get it straightened out. Projects like GNOME don’t seem to have these kinds of problems. GNOME generally has a roadmap that they follow very closely. If something isn’t finished in time they postpone the release, but it is never for very long because they haven’t promised more than they can deliver.
Indeed. Both GNOME and KDE are HUGE projects consisting of dozens of different packages but both projects are always on schedule and both seem to only promise what they know they can accomplish. But they have a different mindset from the beginning: promise what you know that can be kept, and don’t start working on new features before the old ones are implemented. X developers seemingly rather start working on new features, discarding the promised ones, and delivering neither.
I am not saying they don’t do good job. I couldn’t do even nearly as good. But they need a roadmap and they need to stick to it. Jumping all over the place creates mess like this we’ve just got in our hands…
If your goal is to encourage them to do a better job, then whining and complaining is possibly the worst way to achieve that. Everybody likes compliments, nobody likes people who whine and complain. And yet what do people do? They whine and complain when there are problems but say nothing when things are going well.
Think about this.
Don’t even get me started on “constructive criticism”. Most of the criticisms that I’ve seen on OSAlert and Slashdot are far from “constructive”. They’re usually whines and complaints that the developer can’t do anything with, even though seeing all that stuff it still affects them mentally in a negative way. It’s surprising how many people think they’re being “constructive” when in reality they’re not.
If I’m an X.org developer it would be much easier for me to stop giving you free code and free support and to move on to something else. That would put you in an even worse position than before because now you’ve lost a person who does stuff for you for free.
Edited 2008-09-11 15:06 UTC
Well if you can’t take criticism you probably shouldn’t be in charge of a major project like X. We’re not talking about grammar school children here, we’re talking about adults. Coddling people’s egos isn’t going to get you anywhere. If you accept that status quo then obviously nothing is going to change and we’ll continue to get ridiculously late and stripped releases. I never attacked X developers personally and I wouldn’t because there is no need to. I simply criticised their release roadmap/scheduling. I think it’s obvious that it deserves criticism.
Correct, but the basic desire for compliments and rejection for criticism still exists in many, if not most, adults. Would your boss stand your criticism? Maybe he would, but there’s a pretty high chance he wouldn’t, no matter how constructive you’re trying to be. Maybe you would think that he’s being childish, and you would be right, but that doesn’t take away the very fact that you would still be screwed. You aren’t going to get very far by continuously ignoring this basic human property.
Research Abraham Lincoln’s life style. He realized this and refrained from criticism as much as possible. Instead of criticizing, judging, condemning, he tried to understand the other party. As as a result people listened to him.
That said, what you and I say here doesn’t matter. Mark Shuttleworth just injected more money into improving FOSS, e.g. by hiring usability experts and programmers. Instead of criticizing, judging, condemning, he chose to take action. This is what improves FOSS, not the complaint of some random guy on OSAlert. At the end of the day, actions still speak louder than words.
Edited 2008-09-12 14:58 UTC
Your agruments remind me of the PTA when I was in grammar school. When I was in sixth grade they decided that every student should be declared “student of the month” at least once so they created all kinds of different categories for “student of the month” and made sure everyone got elected at least once. Do you know what happened? It didn’t mean anything anymore. You didn’t care if you got it because you knew it was a joke and wasn’t a reward for your work. There is no reason to strive for anything anymore if it is just given to you without merit.
People excel when challenged. They flounder when the status quo is accepted. It’s not only unhelpful but completely dishonest to tell grown adults that they are doing a “great” job when they so obviously aren’t. I know I said it before but I will reiterate, I’m not complaining about the actual work being done, I’m complaining about the work not being done. I just upgraded to 7.4 and I am pleasantly surprised at some of the benefits (EXA is working even without GEM!) but that doesn’t mean everything is fine and dandy. The release schedule needs to be worked on and it’s obvious that management is atrocious. It needed to be said. How else are you supposed to know if you’re actually doing a good job or not? Isn’t that important?
Don’t get me wrong, criticism certainly is useful. But one musn’t get overboard and focus only on the bad things.
Maybe because X.org is *not* MS?
Look at Chrome vs IE8. Both use multiple processes. People treat Chrome like the second coming of Christ, while non rejoice about IE8. And apparently, nobody thinks that this unequal treatment of MS is strange.
So the fact that X.org is not MS is a valid reason for praising them.
And how much were you paying for X.org again? Nothing, I suppose. Why don’t you lend a hand, either by contributing code yourself, or by donating money to the developers, or by hiring someone to contribute? Yes yadda yadda open source developers are teh suck and elitist. That doesn’t mean that they have to work for you for free, especially when they’re already doing their best. If you’ve paid money for X.org then go complain to your vendor and tell them to assign more people to X.org development.
Yes yadda yadda open source developers are jerks. Nonsense. Nothing’s going to change just because you’re complaining. It doesn’t get work done. *People who write code* get work done. You can complain all about the wolves that are threatening your livestock, but it isn’t going to help you until you do something about it.
Edited 2008-09-11 14:59 UTC
The standard response you see when you critise FOSS.
Fact is: FOSS is free. That doesn’t exempt it from critism. Please understand this.
Uhm no. You didn’t read the rest of my comment.
It doesn’t matter how much you whine about “standard response”. Whining doesn’t get you anywhere. America isn’t built by people who whine. Democracy isn’t formed by people who whine. Software isn’t made by people who whine. Problems aren’t fixed by people who whine.
It’s by people who do.
It has got nothing to do with being exempt from criticism or not. Whining isn’t going to solve it, regardless of whether it’s exempt from criticism. You aren’t solving the problem.
I’ve seen this happening over and over in many places. People complain, whine. And a year later, nothing changed and the problem still exist. Why? Because whining and complaining is all that people have done – nobody actually stepped forward and fixed the problem. People who say “Typical standard FOSS response” aren’t getting it. I’m criticizing you for not doing anything useful, not for criticizing.
Edited 2008-09-11 18:00 UTC
Well, maybe they’re not getting it, but you don’t seem to get something else: most people just want something that works. They couldn’t care less of the development process of their software, they just want to use their computer. It’s not because the source is available that you have the time, the will and/or the technical expertise to fix it.
It will only get worse as more people are adopting free software. If the developers can’t handle it, perhaps they should stop bragging their environment as a suitable alternative for everyone to proprietary environments, where these companies do cope with whining because it’s their bread and butter.
What it really comes down to is this inconsistent, and unfortunately common, viewpoint:
“We want and fully expect world domination. And if they don’t come flocking to us then, well, they’re just stupid. And if anyone criticizes our products or methods then, well, they’re just lazy whiners.”
Edited 2008-09-12 07:18 UTC
And just who expects full world domination? I quote (http://slashdot.org/comments.pl?sid=655187&cid=24723577) from a member of the open source community:
Nope, doesn’t anything like the zealotous/elitist image that you’re trying to draw.
Which means that Mr. John Bailey, Slashdot reader, does not exhibit the popular, inconsistent view. Nothing more and nothing less. I never claimed that you could not find some reasonable people in the community if you looked. You’ve presented a straw man argument.
I guess if I quoted *two* people who *do* exhibit the view, I’d win the argument, huh?
I suppose 1 example isn’t proof. On the other hand, can you prove that the “inconsistent view” is widespread? What reason do you have to think that you’re the one who’s right? What makes you think that you’re not just incorrectly generalizing?
You love the word “straw man” so much. But the fact is, neither you nor I can prove or disprove our claims. Your use of the word “straw man” is hypocritical at best.
What about the developers who are trying their damn best to please users like you? Do they deserve to be flamed like this too?
What about the people who sincerely care about making software as good as possible for their users? Do they deserve to be labeled as “fanatics”/”zealots”/”elitist” by people like you?
Edited 2008-09-12 09:24 UTC
Its a phrase rather than a word. But what can I say? That logical fallacy is common, and I point it out when I see it being (mis)used.
You seem to be the one who wants to “prove” something. I find it sufficient to make my observations and let people come to their own conclusions.
I don’t believe that I have addressed them. They tend to have more realistic attitudes than some of the general fans. If the shoe fits wear it, I suppose.
I’m actually an advocate, a systems integrator, an administrator, and an end user. All four of my hats are concerned about Xorg.
People like me? Odd that you would condemn “labeling” and then use that phrase in the same sentence.
You might want to re-read that John Bailey post you linked to earlier and absorb some of his exhibited serenity.
-Steve
Fine. Suppose that you really are being sincere. If there are developers out there who sincerely care about their users, and who sincerely want Linux on the desktop to succeed, then what are they supposed to do, according to you?
I have done my share. I’ve written software and contributed code to make things better. I’ve listened carefully to my users. I put a great deal of care into software usability – I acknowledge that I’m not the best UI designer, but nevertheless I’m still doing my best by listening to user feedback. Heck, I’ve spent so much time on pleasing users that others are saying I’m doing way too much free work. It saddens me that when I read websites, people still talk about “elitist/scumbag/zealotous open source developers”, as if all my work has been in vain and none of it has been appreciated. Moreover, after having spent years into pleasing users, users seem to have become more demanding than before, demanding even more free work.
Tell me, what am I supposed to do in such a situation? What can any developer do in such a situation? What would you do?
I’ve already given my opinion on that. If the shoe fits wear it. If it doesn’t then don’t. I have not really addressed developers specifically, but more the general infantry. And you are really the one talking about elitism. But since you bring it up, there are, indeed, arrogant and elitist developers out there who end up doing more harm than good. David Dawes? J~Af^APrg Schilling? Jeff Johnson?
So on the one hand, you say that developers shouldn’t concern themselves with these matters and do whatever works for them, i.e. telling them not to care, but on the other hand you still want them to do that you want, i.e. to do more their best to give you a better X.org? Aren’t you contradicting yourself?
Oh yes, I’m sooo elitist and evil for listening to my users, up to points where others tell me to stop working for people to free.
Apparently having the opinion that instead of whining, criticizing and condemning, one should be helping (which includes myself), is considered “elitism”.
Edited 2008-09-12 15:01 UTC
Hongli,
Now you are just talking nonsense. Or perhaps you are confusing what I have said with what others have said. When you are ready to have a discussion rather than play “free association” let me know. Until then, I don’t really see any point in continuing.
-Steve
Edited 2008-09-12 16:56 UTC
Yes most people want something that just works. Most people also want to be rich. Do they get it by whining or do they get it by working hard? It doesn’t matter what you say about people “not caring”. It still doesn’t change anything. I can “not care” about how cars work but that doesn’t mean I’m allowed to drive in one until I have a drivers license. They can “not care” all they want, that doesn’t mean their demands are reasonable.
Heck, I can even say that I don’t care who you are or whether you are a programmer, and that I expect you to deliver a full desktop environment to me within 6 months, for free. Will you listen to me if I said that? If not, then how can you possibly expect others to listen to you? Do unto others as you would have them do unto you.
Why are you complaining to the free workers? Complain to the paid vendors, i.e. RedHat, Canonical, etc. from whom you have bought a support contract. They are the commercial companies here. If you want to complain like a customer, complain to the vendors.
You take open source for granted. You expect more and more free work from open source developers, and if anything isn’t to your liking, you flame them. Why are you wasting your time doing all this? Is it because you don’t want to pay for a proprietary system? Heck, even most actual users aren’t that ungrateful. They usually say something along the lines of
“hey could you implement this and that? that would be awesome. but I understand that you have your own life and that you’re probably busy, so if you can’t or can’t do it immediately then that’s cool”
And surprise – developers usually listen to these people. For some reason, you, and many other OSAlerters and Slashdotters, feel the need to act ungrateful, and when developers don’t listen to you, you blame them. Hello, what happened to simple social skills?
I’m not arguing about the merits of open source. I’m talking about your attitude to people.
Edited 2008-09-12 08:03 UTC
What? Not only I’m using proprietary systems, but I have collaborated with open-source developers. Aren’t you a little bit quick to judge?
Yeah, that’s a really good question. Why the hell are you flaming me? I’m just reporting how it is and you flame me like some medieval witch.
I’m going to say it again: most people don’t want to get involved in the development process of the software they are using, they just want to use it. Asking them to “do something to fix the problem” instead of whining and complaining is unrealistic. It happens that I have the expertise, the time and the will to contribute, but I am fully aware that most people don’t. No more, no less.
Hey, I’m not saying that whining is the way to go, but it’s something that happens and have to live with it, unless you want to stay in an ivory tower.
Fine, I’m sorry. That said, there are a lot of people out there who are like that. The conversation usually goes like this:
Person A> feature X is missing
Developer> sorry, haven’t got the resources to do it right now. but if you hire me I’ll give this job more priority
Person A> WTF? (logs off and goes looking for a torrent of $2000 commercial package)
How’s the weather back in 1998, FooBarWidget? I hear that back then all the FOSS devs were working for free in their spare time, instead of posting their patches from intel.com, redhat.com, ibm.com, novell.com, and fortune100.com addresses, and collecting paychecks for it. The “you’re all just a bunch of freeloader parasites sponging off of altruistic developers working for free” argument went out with the 90s. Guess what? X.org isn’t autopackage.
Edited 2008-09-12 18:13 UTC
You would be right if all, or at least most of them are actually collecting paychecks. How come the guy who posted the xserver announcement doesn’t have an address from a fortune 100 company? http://lists.freedesktop.org/archives/xorg/2008-September/038187.ht…
I see many more developers who aren’t working on the project professionally. Maybe even today, a significant portion of the developers aren’t getting paid for working on open source?
And yet open source developers are often criticized for scratching their own itch. How is that possible if most or all open source developers these days are paid? Apparently most open source developers are only paid when it’s convenient to claim that?
I haven’t worked on autopackage for years, so that has got nothing to with it.
Edited 2008-09-12 18:52 UTC
Adam Jackson works for Red Hat.
X.org, Linux, Gnome, KDE, OpenOffice.org, Mozilla etc are all mostly funded at this point. People contribute patches and whatnot, but actual core developers that are not collecting a paycheck tend to end up getting hired by somebody. There is a huge difference between the flagship apps, and smaller one-offs.
at the rate status changes were happening.
The 2d nv driver gets more of the latest card support, it runs really nice with EXA on.
Runs nice if you dont need 3d, way better than the nvidia driver. I’m running it with KDE4.1/Xrender desktop effects, no laggyness unlike even the new beta nvidia driver.
FYI
Option “AccelMethod” “EXA”
Option “MigrationHeuristic” “greedy”
you don’t get xv acceleration either… No 3d is fine for me but I can’t use it without xv.
With nv, exa acceleration is still done in software as the driver is 2d only (no 3d engine code) and some xrender operations require 3d hardware (if using hardware acceleration)…
You conspicuously fail to compare which Nvidia driver within KDE 4.1 and I’m sure that’s for a reason seeing as the latest Beta addresses KDE’s requests.
Not sure what is the point of this sentence.
However, it seems clear that SlackerJack is running the 2D nv driver within KDE 4.1, and that Slackerjack found that the new beta Nvidia driver exhibits some “laggyness” under KDE 4.1, while the nv driver does not.
The only area I experience sluggishness is with Plasma.
Yet on my 8600GTS this helps out:
Option “BackingStore” “true”
KMail itself has bugs unrelated to X that make it still not ready for prime-time.
I’ll take Thunderbird of that until perhaps Mailody shows me something worth using over it.
Konqueror scrolling rates are not a problem. Page rendering is dog ass slow.
4GB Ram on a Pentium 940 has OSAlert painfully slow to render, yet on Opera 9.5.2 it works smooth [I’m betting this Qt3 vs. Qt4 within KDE4.1.1 has a lot to do with it still needing to be flushed out.]
Not to be outdone, Iceweasel from Sid crashes quite often, even from a clean user config and install ever since 3.x has been released.
With all the new features comes tons of performance regressions and exceptions not accounted for in prior branches.
I’ve been using the 177.70 Beta driver, yes it’s better but scrolling with Various KDE4/Qt4 apps is still laggy and slow, resizing frames and scrolling in kmail is very laggy.
Amarok2, Quassel are laggy but not with the nv driver with them settings, it’s actually a pleasure to use.
Hope this version fixes a problem that I am experiencing with Xorg 7.3, where it always respawns and crashes during shutdown with colourful patterns on the screen.
I’m glad to hear it’s not just me! I’ve been blighted with this since xf86-video-i810-2.*. Really hope it gets sorted, as I dread shutting down now
Also, xorg seems to have lost momentum now. Every release something gets cutted and we are left with something sub-par.
Even to this date, compositing is not working as good as on proprietary OS.
Anyway, the server is what the companies earn their money on so no wonder nobody is investing in the client.
I have only two things to say(It turned out to be three) about this release the first is as someone who has loathed Phoronix for like forever, just check my history on how often I criticize it. I still dread any article that contains a benchmark, for the love of God I dread those things, but their X.Org coverage although perhaps overly negative, in the light of many large projects overrunning their deadlines(Gnome/Linux being a bit of an exception) has been excellent.
I made the right choice in choosing a then cutting edge X3500 integrated solution as it was a major step back from the although aging chipset X800 card, and many days I have regretted my choice, but my intel drivers have been little movers and shakers, although I’m more than a little concerned about GEM. I suspect I will have to step back from running a cutting edge solution for whats looking like 3 months, but I will be getting the nextgen motherboard once they are at commodity prices.
I think its a shame that this release has turned out like it has. Reading the article the whole thing reads like a stopgap solution for the 8.10 distributions. Which has clearly meant that some of the new technology has had to be pushed back at least 3 months to what seems to be the real release. It just shows you what, albeit a bit of a dick(or perhaps because) Linus skills go further than a little bit of coding, and what a difference a man like him in the driving seat makes to a large scale project. X’s development seems to progressing at quite the pace, but getting a timely cohesive release seems to be beyond their grasp. I’ve often proposed that the FOSS community would benefit from managers; layers; marketing people from the “community” taking a greater interest, or IBM etc stepping up to the plate, but it seems to attract comedians with iconic wit I hope thats not a damning indication.
Although seriously this release alleviates a barrier to Linux greater than Microsoft; Desktop Gaming; Adobe and Thom the xorg.conf. Praise Red Hat
The whole purpose of the opensource world is to ‘scratch and itch’ so the inevitable result, however, is that unsexy (but crucial) parts of the stack are neglected by their lack of sex appeal. Xorg is an example of just that – where people’s interests sit at the top and visible layer rather than the unsexy work that sits underneath. There are some projects, like the Linux kernel, which have obtained a level of ‘attractiveness’ but the general rule of thumb – the more unsexy it is, the less developers. That can be seen in GTK+/GLib/Pango, Xorg and numerous other low level components.
It is unfortunate that companies aren’t putting the man power they need to fix these issues – and these issues aren’t new, this has been going on for a while now – a Sun engineer in one of his blogs talked about the sliding quality and quantity of contributors to the project. Unfortunately, the solution is going to cost Linux distributors/Sun/HP/IBM/etc more money – and given the nature of doing the least amount possible, I doubt there will be any sizeable contributions from them anytime soon.
Of course, this is an improvement over the Free Software world, which is mostly “scratch and sniff”.
I was going to agree with you your same point has been put by Linus as others in a “positive light” see any article regarding contribution to the kernel. Albeit you must remember your precious OS X is built on Open source.
As far as amateur[sic] FOSS programming goes, you can see where it really shines and thats small tools and add-ons that make up a whole. IMO are where it really shines which is why I’m really excited about Openoffice.org 3.0 and Firefox 3.0 is simply fantastic.
Who said I was criticising OpenSource? nice to see that when you disagree with someone, you and your many sock puppets silence critics. How many accounts do you have here? 2, 3 or 12?
The problem with the ‘scratch and itch’ philosophy is that if something is critical but doesn’t provide a sufficient itch to developers, it will be ignored in favour of a more interest itch.
I don’t know how the hell you came to the conclusion that I ‘hated’ open source – but then again cyclops, you’re a know hyperbolist and liar who is willing to twist and distort other peoples posts to suit your own twisted ends.
Who made any value judgement over amateur? you do realise that amateur is not a pejorative but merely meaning, “this person is not paid for the work done”. As for OpenOffice.org 3.0 and Firefox, may I suggest you look at who are the main contributors to these projects.
Edited 2008-09-10 21:59 UTC
Yes, but the pejorative tinting of amateur is what comes to most peoples’ minds when its used to describe one’s work.
To me, I have to hear more to make a judgement. If someone says, “this is a great application from a local amateur developer” means to me that this guy is able to make a great piece of software regardless of the fact that he won’t make any money off it. To me, an amateur who makes a high quality piece of software will receive more ‘wow’ from me than a paid developer who makes an application of similar quality. Maybe I’m just a romantic who cheers for the underdog.
I understand what you mean – some people make the leap to the conclusion that anything that isn’t done by ‘paid’ experts is obviously of questionable quality (even if all evidence demonstrates to the contrary).
I’m still giggling at that number of accounts, you need to report that abuse. As for my twisting and turning, I’m sorry thats package and parcel here. Its regular that a discussion here can be put in both a positive and a negative light, from DRM to proprietary to Development Models, Companies producing software and hardware in synergy etc etc often these are approached with a back or white approach when the answer is more complex and very very grey, and almost everything is about pros vs cons. Thats why I regularly object to subterfuge of the “I run Linux but Vista is great variety” or “Someone got a x4500 running Openarena but it only ran at 30 fps at 1024×800” simply because it nobody benefits. You have yourself commented on my use of the the word “Smackdown”.
I put [sic] after amateur its actually a reference to Linus post regarding the development of the kernel clearly one you did not get.
I also made a reference to OpenOffice 3.0 because its the version where add-ons come into their own, and Firefox for its plethora of add-ons, but again its a reference you did not get.
Glucose, AFAIK, isn’t needed anymore with the memory manager(s).
It was supposed to provide Xrender acceleration when AIGLX was running, by implementing it in OpenGL(using glitz?) and sending indirectly to X server which would run as part of its rendering context (so glucose is “indirectly” rendered). Same thing was done with Xgl, using glitz library to translate XRender calls to OpenGL.
Now, the new EXA-TTM(GEM,UXA?) plays nicely with memory manager which can now take offline pixmaps rendered by accelerated EXA Xrender and composite them with the window manager running on AIGLX. Same as with Glucose, if not even simpler and without GL indirection.
it is funny that nvidia which has been doing this as they have their own memory manager and probably have an architecture similar in function to uxa to accelerate render. Yet, using xgl is still faster than nvidia’s render acceleration, especially as with the use of a compositing manager, the gpu is accelerating pure opengl only (not render+opengl)
Only the most recent beta drivers seem to be approaching the speed of xgl (using kde4).
I was quite amazed when the Fedora folks decided to release Fedora 9 with a beta X server (and no solid idea when it would go final). It’s *still* in beta and, what’s worse, is that now that the X.org devs have finalised 1.5, Fedora are in still in update lockdown from the August security incident (I see rawhide updates are still available – Fedora 10 Alpha happily downloads them for instance).
ATI and Nvidia refuse to support any pre-releases of X.org, which basically means that Fedora 9 has been without easily-installable ATI/Nvidia 3D (most Fedora users would do this via the livna repos, which have no F9 RPMs for ATI/Nvidia) in its entire 4 months to date. A shocking state of affairs, although amusingly my Intel laptop is happily running F9 3D thanks to its integrated Intel chip (open source drivers show the way!).
On my Dell Vostro 400 with an ATI 2600 XT, the standard 2D X driver (“radeon”) in FC9 is ridiculously flickery and unusable, which was the final straw for me and I stayed on FC8 as my primary desktop (the last Fedora release to properly support 3D for ATI/Nvidia – F10 Alpha doesn’t either, BTW).
For better or worse, Fedora is a very strong supporter of OpenSource, and I believe will neither postpone/hold up the release of open source software due to the lag in closed source drivers, or bundle said closed source drivers.
I personally don’t have a problem with this, and is the main reason why I only get intel based graphics cards for any “work horse” system that I want kept up to date.
There is a *lot* of stuff going on right now. The problem is that X is kind of in a middle zone where the old cruft is being removed or refactored, and the new technologies are as yet unproven, or at least unstable. New semantics have to be worked out and the answers aren’t always clear cut. But I think once these issues get ironed out, we will start to see a lot of visible progress, instead of stuff under the hood.
as much as i love open source i switched back to windows xp because i don’t have too worry about the new xy version of what ever and the breakage that comes along with it. and seriously all that composite stuff is usless. how about light stable and fast instead of glitz and glam.