Wine 1.3.23 Hooks Into More Direct3D 9.0 Functionality
Phoronix: Wine 1.3.23 Hooks Into More Direct3D 9.0 Functionality
Wine 1.3.23 has been released this afternoon with more Direct3D/DirectX 9.0 functions being implemented by this popular free software project...
Looks like this release also breaks a lot of things in both DX 9 and DX 8. Alpha Prime crashes, Project Nomads (which is a DX 8 game) has severe graphical glitches. It would probably be a good idea to skip it.
Hmm... Explorer shell; as in, desktop, icons, (and, maybe) taskbar? Interesting....
PUBLICLY available DX93D specification?
Better idea is to file a bug informing about regression.
Originally Posted by Reloaded211
Agreed, 1.3.23 essentially broke every game I bothered to check. I'm back to 1.3.22 for the next two weeks. I'll probably temporarily install 1.3.23 again just to file some bug reports in case that hasn't been done yet before I get rid of it altogether (it's not like 'sudo make install' takes a long time).
Originally Posted by Reloaded211
Counter Strike Source is one hell of a lot better for me now. Getting a solid 70fps on Ubuntu 11.04 - classic. Is set to dx8.1 mode however. Otherwise very happy, only issue I have now is the odd crash when using the config menu..... Wonder how Portal 2 works?
Wine = buggy as hell = unreliable
I have a great idea for Wine: do not release another version of Wine until all regressions are fixed.
You have a stable branch in which bugs get fixed. Those fixes get passed along to the dev branch. In the dev branch, you work on fixing regressions as well as bugs from new code adds, like the Linux kernel does. You don't make the dev branch be a stable release until ALL THE REGRESSIONS ARE FIXED. Otherwise, your users cannot depend at all on one stable branch being better than the last one, because they aren't.
I shouldn't have to go get version 1.0 of Wine because 1.2 is horribly broken for certain games. 1.2 is supposed to be better than 1.0. Stable branch + version increment = better, period.
Wine seriously needs to get that straightened around. Perhaps they need to develop some tests to put a dev branch through. Find several open source or shareware games or graphics benchmarks (like PTS or whatever) and have several testers run them all.
In fact, PTS could make a test series specifically for Wine regression testing. It would help get everyone on the same page and would encourage testing if that is the missing component needed to get truly stable Wine releases. Once you have some standard basic tests in place, then you can also listen to other sources for any other regressions, but you have to start somewhere.
Wine project has to keep adding features to not stay behind windows api development even more..... to workaround regressions, there are many tools to create a different wine prefix for each application you need, with the wine version that works..... Wine developers already do a hell of a job, you can too make some effort to tune your system.... the tools are there, you don't have to build one from scratch.
So you want to make Wine users have to create multiple Wine prefixes and deal with all that BS just to properly run Windows programs instead of Wine running the programs properly to begin with? I think I speak for all Wine users when I say no thanks.
Originally Posted by Bitiquinho
The solution is not workarounds, the solution is to fix all regressions and bugs in Wine.
Who cares if the Windows API keeps expanding? Focus on getting the code that exists stable and eliminating regressions. If Wine did that, maybe users would actually start getting excited about new versions of Wine instead of "oh, yay, another update that broke all these existing programs from working."
Tags for this Thread