Your Thoughts on WINE (Codeweaver products incl.)

[ninjatux]

Yesterday, I went on Codeweavers' website, the company that employs many WINE developers and facilitates its development. They offer several commercial versions of WINE for Linux and Mac. One of their products is Crossover Games. I bought the version for Mac OS X. They have an experimental build for PC-BSD that runs on FreeBSD, but it won't do me any good if the radeonhd driver doesn't support accelerated 3d.

I spent nearly all of yesterday testing the three games I play on Windows Vista, which are Team Fortress 2, Day of Defeat: Source, and Counter-Strike: Source. Out of these three, Counter-Strike: Source ran the best with good graphics quality. Day of Defeat: Source had equally good graphics quality, but occasionally crashed on the first click. Team Fortress 2 was disgusting. All settings for higher quality textures did not work. You basically could run it with the bare minimum. Being fed up with this, I cleaned up my system and kep Vista for gaming. The only good thing that came out of this is the one year of free updates from Codeweavers. So, if in the next 364 days, the Crossover Games product gets better, I will have access to it. Until then, Windows stays.

Going through all of this made me question WINE's goals. Is it really wise to continue the course? At least in the case of Codeweavers, I think they should really abandon their product lines and move toward a wrapper approach. An open source implementation of the Windows API is a very noble cause, but that's just the problem. Maintaining compatibility with hundreds of applications in the same version while adding compatibility for more and more features is too difficult, especially when you're working from scratch. I guess I'd like to see at least Codeweavers distribute specific installers for each application that bundles a version of WINE with patches that guarantees compatibility only with that application. They can sell each for $5 I don't know. Or WINE could allow you to maintain multiple copies of WINE and provide patchs sets for compatibility for specific applications.

Maybe I'm being naive here, but I don't think the current course that WINE and Codeweavers are taking is going to get us any farther than we have in terms of application availability and compatibility. I understand it's a work-in-progress, but a lot of the lower lever work has been done.

What are your thoughts on WINE?

[robbak]

An interesting comment on slashdot (yes, they do occur, on occasion) made this point: It might just make win32 just another Open Source API. With a little luck, Microsoft may just lose control over their API.

[TerryP]

I haven't looked at CodeWeavers games product yet, since even on WINEs app database the games I need were mostly paper weights for my needs (patched multiplayer+anti-cheat).


The last time I intentionally installed WINE was on a PC-BSD machine, at that time based on FreeBSD 6.0-Release. And I found WINE to be of little use for anything beyond running GVim or Emacs for Win32. Hopefully with the more recent leaps and bounds, WINE + FreeBSD actually counts for something.



What you suggest reminds me of what PC-BSD has been doing (which I could spit on), building PBI's with internal WINEs and a bundled set of software, e.g. IE+Flash or MSNMGR.



The problem I have with WINE, fast forward a few decades....


What if all/most mainstream applications for Linux and Windows are writtin using the Windows API via WINE and kludges for compatbility where they differ.



INSTEAD OF PORTING THE RAT *****ING PROGRAM PROPERLY !!!!


I believe word perfect once tried something similiar... :\

[DrJ]

Originally Posted by TerryP

The problem I have with WINE, fast forward a few decades....

What if all/most mainstream applications for Linux and Windows are writtin using the Windows API via WINE and kludges for compatbility where they differ.

INSTEAD OF PORTING THE RAT *****ING PROGRAM PROPERLY !!!!

This always comes up, but I think it is a red herring. Linux programs will be written for Linux, and that will not change.
Do you really think Gnome or KDE are suddenly going to switch to a Windows API? Really? And once the Linux version exists, it will be ported to BSD (with a few notable exceptions).

Commercial programs just don't run on BSD, as a rule, so if you want to run any of them, you need to use something like Wine or a VM. Yes, a few important ones run Linux with the Linuxator.

Practically, Wine can be useful for a few applications, but it will never be a full-fledged, bug-free implementation if history is any guide. The API it uses too is becoming out-of-date, so they will have to expand the whole thing for Vista.

I'm grateful Wine exists -- it can be useful. But it certainly is no panacea.

[TerryP]

Software for Linux should be written for Linux.


If WINE works as well as one would *hope* someday, why should commercial vendors write software for Linux and Windows when they can write Windows software with tweaks for WINE and avoid actually writing programs for Linux altogether?


Red herring maybe but I am a person who always looks at all of the possibilities I can see within the futures light.

[DrJ]

Originally Posted by TerryP

Red herring maybe but I am a person who always looks at all of the possibilities I can see within the futures light.

Maybe, but I just don't see it coming any time soon, and by that, I mean the next five years. Beyond that, who the heck knows?

Commercial vendors (which is all that this really is about) will write programs where they can sell them. The performance increase for Linux instead of Wine is worth it as long as there continues to be enough people who use it in their market segment. The only change I see is that the vendors will stop supporting Solaris/SPARC and move to Solaris/Intel. There are a surprising number of vendors in this boat.