The most important thing in my book is stability. And the proprietary drivers for ATI never delivered that and were focused mostly on performance above all else.
Personally I think that stability matters. And the open source drivers can deliver that. They can, with the help of involved users, fix bugs fater, support hardware longer, and provide a high quality out-of-the-box experience for Linux.. which the proprietary ATI drivers never could.
So ATI/AMD should be hailed and congratulated for what they have done.
Let them worry about the latest and greatest and getting the last 10% of the performance that some folks need... I want decent performing, reliable, stable, and zero effort 3D acceleration; which is what the OSS drivers can deliver.
Don't underestimate the high quality of experience that zero effort can deliver. Seriously. Having things 'just work' is WONDERFUL and makes it much easier for everybody... from the developers to the end users.
When shit 'just works'... Linux is _SLICK_. It's a terrific operating system and surpasses even OS X in many ways in terms of usability. But we can't get there relying on proprietary drivers... they just can't deliver what is required.
lol anyways this is how things are:
AMD is dropping support for cards where there isnt yet a valid alternative. In other words they are kindly inviting users that own such cards(example me) either to wait (that jesus comes back on earth) or simply go and buy a new cards.
I'm sick of all this. Everyone should.
I'de like to know the day my ATI card will do something it never did on Linux: WORK.
WINE only Fullfeatured on ATI/AMD carts with OpenGL3!!!!
not realy for DX10 for DX9 becourse bevore OpenGL3 comes out wine only uses Nvidia exstansions to bring DX9 to work!
so in Fakt! you are on Nvidia for WINE or use an OpenGL3 Cart from AMD/ATI!
and R300-R500 only has OpenGL2.1!!!! with NO Nvidia extansions!
Get real. Most people wouldn't invest a penny in Linux if they were AMD, you need a _really_ long vision to believe that that investment should pay off sooner or (much) later. If I were AMD and wanted to _make good money_, I wouldn't give a rat's ass about Linux.
What you get on Linux is per user _much_ more than what a normal Windows user gets. The fact the quality on Windows is still better is just because Linux has a _very_ small share of the market.
becourse the definition of the market share is wrong.
2 definitions one of them says money the OS with the most mony ern is the winner.. or the second definition piece the OS with the most
quantity is the winner.
In the first definition linux Can't win! becourse Linux is Free and const nothing!
In the second definition linux can't win becourse only Computer with OS pre installed are Cound to the statistic!
And linux user don't by pre installed hartware they Buy Software-Less hartware and install linux!
so in my Business and frend circle Linux Grows up to 50% Real market share!
i know a lot people they Buy Windows AND MACOS and they let RUN LINUX not windows or macos! (i have windows licenes and macos licenses to! )
You know, my original reaction was rather kneejerk. I would have even posted it, had I not fallen asleep. But stepping back, I find two things: first off, I need more information before I can draw a solid conclusion on all of this. Second, and possibly more importantly, I am absolutely appalled at some of the willful ignorance I am seeing in the replies: This is part of the "sense of entitlement" that people talk about when they paint a negative picture of Linux users!
"I paid money so you should grovel at my feet!" That's the sort of attitude I'm seeing in some of this. Absurd. It's ridiculous even before you consider that comparing the price of even a hundred-dozen cards to what ATi (and now AMD) pays supporting our OS reveals a multiple-orders-of-magnitude disparity. You know, even if you don't write code; even if you don't bother to submit bug reports at all, you should at least be aware. Being a Linux user doesn't conjure you a license to ignore everything that's happening outside of a very myopic scope; it gives you an impetus to join your community of fellow users to solve complex problems. You know, that thing we're supposed to be good at? I'm honestly having trouble comprehending this concept that there are users claiming to have hardware that hasn't worked for years. Users that are posting on the forum of a site actually dedicated to news about the current state of their drivers and alternatives not knowing about the drivers...this is all a big joke, right?
I have a 9700 Pro. All-in-wonder. Great card. Now, I could have waited five years for a "driver that works," but instead, I chose to read. I chose to join my distro IRC channel. I chose to root out reference to my problem and hone my troubleshooting skills. In the end, I found the issue that I was having, corrected it (in userspace, even!), and had a driver that worked. fglrx worked. Later, on my laptop, fglrx stopped working on a new kernel. Rather than complain, I knew, having kept myself abreast of driver development, that R300 support is great in radeon.ko. Fantastic. My open driver works. This isn't handing out solutions on a platter. Nor is it being the insular geek cabal from their moms' basements. This is a user being proactive about solving a user's problems by communicating well.
So yes, if you think that, by being a single paying customer out of the multitude of paying customers, you're somehow entitled to a sense of outrage when AMD exercises good business sense in a terrible economy while still supporting your hardware at great cost to them, you are indeed whining and likewise every bit a part of the problem. Why not cool your head a little?
That said, Mr. Bridgman, I wonder if you could possibly be the supplier of the information I seek? Here is the situation as I have experienced it currently:
The problem, unless the next release is a bloody miracle, is that fglrx tends to depend on parts of the kernel that are long obsolete and due for removal. Even 2.6.30 will most likely break fglrx in a new and novel way. And now, with older hardware, there's no update in sight. What's more, fglrx really doesn't have an outstanding track record with "just work"-ing for many even after you finally get it installed. My own experience, and those that I correspond with, indicates as much.
"fglrx on one side and nvidia on the other...Larrabee can't get here fast enough; to hell with the cries of 'monopoly!'" was how I actually ended my original post.
However, now I must instead ask this question of you, Mr. Bridgman and associates:
Will this action of discontinuing support for older hardware result in an fglrx that is much smaller, substantially more stable, and better able to cope with the kernel release cycle?
So what, when they won't compile against newer kernels or work with newer X servers? Do you expect people to run (say) Fedora 9 forever? Or the 2.6.27 kernel? Do you understand that Fedora 8 is already dead in that no new security fixes/updates are being released for it?Quote:
The old drivers will continue to work.
Huh? Windows? What's that? I run LINUX, and am not remotely interested in what happens in this "Windows World" of yours.Quote:
Or will the windows driver stiop working with the release of 9.5?
Yes they do; drivers that are incompatible with your OS are useless, and compatibility is something that must be actively maintained. fglrx doesn't even work with Fedora 10 out of the box - so frankly I must call "BS" on your post.Quote:
No! Nobody forces you to update the drivers.