what is the problem? its only smaltalk
Originally Posted by bridgman
if that happens, that's great.
Originally Posted by Qaridarium
but if we look back, ATI took quite a while to support XServer 1.5
it cause havock for lots of distros that wanted to ship Xserver 1.5, but couldn't without breaking everyone with an ATI card
I'm gonna enable compositing in a few days! yay!
bridgman, is discussing this stuff really that bad? after all there are some book-long threads about the leaked 9.1 and 9.2 drivers (and no harm's done)
+1 about the open betas too (yeah, I know you're not taking votes )
brigeman don't like AMD Fanboys realy much :-)
Originally Posted by mits
look at my system thats an AMD-Fanboy system:
ass@ass:~$ infobash -v3
Host/Kernel/OS "ass" running Linux 2.6.28-6-generic i686 [ sidux 2008-04 Πόντος - kde-lite - (200812222321) ]
CPU Info (1) AMD Engineering Sample 1024 KB cache flags( sse3 nx lm svm ) clocked at [ 1000.000 MHz ]
(2) AMD Engineering Sample 1024 KB cache flags( sse3 nx lm svm ) clocked at [ 1000.000 MHz ]
(3) Dual-Core AMD Opteron 8218 1024 KB cache flags( sse3 nx lm svm ) clocked at [ 1000.000 MHz ]
(4) Dual-Core AMD Opteron 8218 1024 KB cache flags( sse3 nx lm svm ) clocked at [ 1000.000 MHz ]
Videocard ATI Device 954f X.Org 1.4.2 [ email@example.com ]
Network cards 2x Broadcom NetXtreme BCM5754 Gigabit Ethernet PCI Express
Processes 146 | Uptime 23:32 | Memory 633.0/3276.9MB | HDD ATA SAMSUNG SP2014N,ATA SAMSUNG HD103UJ Size 1200GB (88%used) | GLX Renderer ATI Radeon HD 4350 | GLX Version 2.1.8535 Release | Client Shell | Infobash v3.05
They're closed programs for a reason. I don't work at AMD so I don't know exactly what that reason is, but nevertheless it should be respected.
Originally Posted by mits
People will leak these things eventually. Happens all the time. Is there like some sort of "punishment mechanism" for these kinds of infractions? All I have seen so far is stern/angry letters from AMD reps. to some sites that host these leaks (Guru3D comes to mind), but if there is no real consequence of any sort this will keep happening.
Originally Posted by bridgman
Honestly, I'm pretty disappointed that it ever gets to the point where a punishment mechanism needs to be discussed. The reason I don't lie to or steal from people is not fear of punishment, it's that I know it's wrong.
I don't have a problem with people saying they don't believe in NDAs and refusing to sign them -- but if you willingly enter into an agreement not to disclose or distribute a driver then you shouldn't need the threat of punishment to make you keep your own commitment.
Not talking about Qaridarium here, but it seems pretty likely that someone who *did* sign an NDA is passing on the drivers anyways.
Sure, but perhaps it could be done a little more DISCREETLY ????
As far as "finding out when it's going to happen", I really don't think talking about leaked beta drivers is the answer. I know it's amusing to think that we ship a bunch of knowingly broken software but that is simply not the case. What has been happening over the last year or so is that we have been picking off the cases where software works on OUR systems but not on all of YOUR systems, and that takes time. The fact that a problem no longer occurs on one user's system is useful for us and promising for you, but you shouldn't take it as an indication that the same change will happen on your system.
We have had a clear policy of not discussing unreleased features or fixes for years, but I keep hearing "ATI promised this" and "ATI promised that" anyways. I don't normally challenge the claims, but when I do it usually turns out that the "promises" are actually just one well-intentioned user telling user B that something great is going to happen.
If a statement like that is made and then the fix/feature/whatever doesn't happen, the people waiting for it are a lot less happy than they would be if nothing had been said -- and they blame us no matter who made the statement. We don't want that.
Last edited by bridgman; 02-09-2009 at 09:27 PM.
ATI drivers may work ok with some kind of usage. For standard internet + office use only few people really get annoyed by a bit slower rending, of course those cry out very loud. An other usage would be watching videos - crashed Xserver when Xv output was used was not the exception, but happened really often, OpenGL output is usally more stable in that error case. As some distros enabled by default Compiz (even if the effects are so trimmed down that you hardly see a diff between Compiz and without) like Ubuntu a new set of errors went into the target of the fgrlx developers. They fixed most of those errors just for the Xserver used for Ubuntu and older Xservers now crash when composite is used (even if they claim to support em). The next thing seems to be KDE 4.x - I hope they don't forget KDE 3.5 users - there will be many as for example Debian 5.0 (lenny) will stay with it. One user group gets the fixes, the other group gets the bugs, thats the way how ATI fgrlx drivers work. Use the most common distro and the GUI the drivers are optimized then you can be happy with it, but if you don't when you definitely will begin to hate ATI - especially when you know somebody with Nvidia or own a 2nd pc with such a card. ATI does not care about alpha or beta versions of Ubuntu it seems - only on releases - so I expect they will not support Xserver 1.6 until they are more or less forced - the OSS initiative is good, but new cards are not fully supported by those drivers and if they want to sell new cards then there must be a working driver. Otherwise you could only buy 2nd card cards *g*.