It seems that you're using an outdated browser. Some things may not work as they should (or don't work at all).
We suggest you upgrade newer and better browser like: Chrome, Firefox, Internet Explorer or Opera

×
Community
General discussionThe "Judas™ does this run in Wine" thread v1.173(128 posts)(128 posts)
(128 posts)
Pages:
1
2
3
4
5
6
7
8
9
This is my favourite topic
Posted July 14, 2016
Please feel free to remove DEADBOLT from the list as we've just received the native Mac & Linux versions of the game and we should have them rolled out for you guys tomorrow :)
Posted July 15, 2016
DEADBOLT Mac & Linux versions added about 5 minutes ago :)
Posted July 23, 2016
avatar
rampancy: 40764.
This bug is fixed in Wine 1.9.15.
Posted August 06, 2016
avatar
JudasIscariot: ...
avatar
Gydion: So, Parkan: Iron Strategy or are you in the middle of a regression test?
Actually, I am in the middle of running multiple UE4 games to see what bugs pop out since a new patch was introduced in this bug report :)

As for regression testing, I seriously need an ELI5 guide on this because I can do the following commands:

git good wine 1.8
git bad wine 1.9

It shows me some random commit and then what? In short, I need regression testing for dummies and I am not afraid to admit it.
avatar
Gydion: So, Parkan: Iron Strategy or are you in the middle of a regression test?
avatar
rampancy: I've actually gotten it to run perfectly, except for the movies, which I can't quite fully figure out. I tried just about everything (installing Indeo Video, wmp9, CCCP) but nothing seems to work. The problem is that the movies lead to wonkiness upon startup (on my system at least), with the game window minimizing and maximizing suddenly. I got around all of that by disabling the movies (by renaming their files).

The only reason why I haven't made an official listing for it is that the only thing that seems to work is either CrossOver, or CrossOver's current fork of WINE run within Wineskin. Both Vanilla and staging 1.9.15 only give a black screen, or suffer from severe flickering/graphical corruption issues.
Have you tried amstream, devenum, and quartz for Parkan? What is the output in the terminal?
Post edited August 06, 2016 by JudasIscariot
Posted August 06, 2016
avatar
JudasIscariot: As for regression testing, I seriously need an ELI5 guide on this because I can do the following commands:

git good wine 1.8
git bad wine 1.9

It shows me some random commit and then what? In short, I need regression testing for dummies and I am not afraid to admit it.
avatar
immi101: start the bisecting with

git bisect start
git bisect good wine 1.8
git bisect bad wine 1.9

that will pick the commit exactly in the middle between 1.8 and 1.9
now compile, start the game and test if the regression is present
if yes type: git bisect bad << will pick a commit between this commit and 1.8
if not type: git bisect good << will pick a commit between this commit and 1.9

continue until you have it gives you the exact commit which produced the regression

there is a more detailed guide on the winehq wiki, it hink. Let me see if I can find it
//edit: https://wiki.winehq.org/Regression_Testing#A_Walkthrough
Alright but I would have to install the game with that version. I know that as long as I don't attempt to $make install this version of Wine, it won't interfere with my system Wine but how do I ensure I am able to install the game and then run it with that particular version of Wine?

Like I said, I need an actual ELI5 for this.
Posted August 06, 2016
avatar
JudasIscariot: Alright but I would have to install the game with that version. I know that as long as I don't attempt to $make install this version of Wine, it won't interfere with my system Wine but how do I ensure I am able to install the game and then run it with that particular version of Wine?
avatar
immi101: you can adjust the PATH variable to make sure that your self-built wine is picked up before the system wine.
run like this in the terminal:

PATH=/path/to/wine/build/dir:$PATH wine <cmd>

for convenience you can make yourself an alias:
open ~/.bashrc in a text editor
add at the end:
alias wine-git='PATH=/path/to/wine/build/dir:$PATH wine'

save & reload config with:
source ~/.bashrc (or open a new terminal)
then you can start your self-built wine with the command 'wine-git' and your system wine is still available under 'wine'

check and compare the output of wine --version <vs> wine-git --version to see if it is working

it gets a bit more complicated if you want to use winetricks with that though.

//edit:
replace /path/to/wine/build/dir with the actual path of course :p
OK, in any case I can always easily remove my system Wine and reinstall it if need be but the thing that gets me is the commits. They have long hash numbers so how am I supposed to tell git that a given commit is good/bad?

See, the problem I have is twofold:

1. The good/bad thing in git seems to pick a random commit so how can I avoid wasting my time compiling Wine over and over and over?

2. They have long hash numbers. How do I properly tag the seemingly random commits when they have hash numbers a mile long?

I wouldn't whine about this if there was a 5 step process like so:

1. Bisect
2. Git shows a commit
3. I guess compile at this point?
4. Tag commit as good/bad
5. Move on to the next one I guess

or something a little less arcane...
Posted August 07, 2016
avatar
JudasIscariot: I wouldn't whine about this if there was a 5 step process like so:

1. Bisect
2. Git shows a commit
3. I guess compile at this point?
4. Tag commit as good/bad
5. Move on to the next one I guess
avatar
Gydion: Whiner. :p That's actually what you essentially do. I would leave your system-wide Wine installed and simply run the Bisect Wine from the build directory (it's set up to "Just work"). Unless it's an install bug no need to reinstall with each test. Use the existing prefix. Regression test is all about compiling Wine over and over again. You can speed it up some.
Thanks :P I actually took the risk and tried it but I am not sure if I did it right, though. Although, thank goodness ccache exists :D Made the whole thing a lot easier :D

One thing that worries me is that when I ran a regression test between wine-1.8 and wine-1.9.0, every commit was bad, that is, the bug showed up every single time (context: I was running a regression test for Worms 2 since it works on 1.8.3 but not on Wine-devel).
Post edited August 07, 2016 by JudasIscariot
Posted August 08, 2016
So, it turns out that Worms 2 *will* work on Wine 1.9.16, one simply needs to have a symlink to an "nw2" directory in order to avoid the crash with the Visual C++ error.

The solution is in this comment by Bruno Jesus. The WINEDEBUG=+all output may have been a bit much but it helped find one culprit as to why the game appears to not work.
Posted August 08, 2016
avatar
JudasIscariot: So, it turns out that Worms 2 *will* work on Wine 1.9.16, one simply needs to have a symlink to an "nw2" directory in order to avoid the crash with the Visual C++ error.

The solution is in this comment by Bruno Jesus. The WINEDEBUG=+all output may have been a bit much but it helped find one culprit as to why the game appears to not work.
avatar
immi101: i vaguely recall that the mentioned land.dat should get modified if you set up your game with the frontend.exe.
More precisely it calls another binary (landgen.exe) which writes the chosen level into land.dat.

if you run the frontend with WINDEBUG=+process you should see landgen.exe getting invoked.

if the land.dat is not getting changed (as it appears to be), then you should always see the same level regardless of what you set in frontend.exe.

still leaves the question whether it fails to execute landgen.exe or if landgen.exe fails for some reason
So far I've been using the "Play quick game against the computer" option without delving into any settings whatsoever in the frontend.exe. Basically, I never picked a particular level to play.
Post edited August 08, 2016 by JudasIscariot
Posted August 09, 2016
avatar
JudasIscariot: One thing that worries me is that when I ran a regression test between wine-1.8 and wine-1.9.0, every commit was bad, that is, the bug showed up every single time (context: I was running a regression test for Worms 2 since it works on 1.8.3 but not on Wine-devel).
avatar
Gydion: That is rather suspicious. Did you test if wine-1.8 actually works for you? There have been plenty of commits between 1.8 and 1.8.3 that you would need to test it. With that said those very frst few related commits post wine-1.8 are not in wine-1.8.3. Do you know if the game uses any Microsoft Visual C++ Redistrib?
Gydion, after every compile with ccache I would run <long dir string>wine --version and I would see various git versions of Wine i.e wine-1.8-<git string>

Heck, I even downloaded Wine 1.8.3, compiled it, and it still crashed (again checked the version number) but it wasn't due to any commits, rather, it might be something to do with the game's CD check (when the crash happens there is an image of a spinning CD-ROM which looks like a CD check to me) and how we have circumvented it.

The following warning was logged with WINEDEBUG=+all:

523436 46574.943:0036:0037:warn:file:CreateFileW Unable to create file L"c:\\nw2\\data\\level\\Gulf\\Level.dir" (status c000003a)

Once a I made a symlink in that prefix's drive_c

$ln -s "GOG Games/Worms 2" nw2

the game worked in Wine 1.9.16 when running it with

$wine frontend.exe

So the issue at hand was that sometimes Wine is not very communicative when it comes to warning about errors such as those that come up with file operations made by an application.

As for a redistributables, I did not see any come up when I was installing the game.
Post edited August 09, 2016 by JudasIscariot
Posted August 09, 2016

next question: why is frontend.exe invoking landgen.exe incorrectly
avatar
immi101: this looks suspicious:

trace:file:GetShortPathNameA ".\\\\data\\water\\Yellow"
trace:file:GetShortPathNameW L".\\\\data\\water\\Yellow"
trace:file:GetShortPathNameW not found!

//edit: simple patch: https://bugs.winehq.org/attachment.cgi?id=55311
I'll be be testing Wine-git tonight as supposedly another issue I've had got fixed so I'll add the patch before compiling :)
Posted August 12, 2016
Just a quick report:

No Man's Sky: works in Wine in Windows Xp mode in 64-bit :D
Posted August 16, 2016
avatar
te_lanus: Do you think it's wrong to pirate games? *:
Isn't the piracy age of Blackbeard, William Kidd, Anne Bonny & Sir Henry Morgan already over?

Roughly how many PC games have you pirated in your life? *
do they really think one could remember. Heck my first NES console was a pirate console(Golden China) with pirate games on them, heck the NES console made it impossible for Nintendo to release their official NES here in South Africa.

In most flea markets I've been (in Johannesburg, Pretoria and Port Elizabeth) you can buy original boxed copies of games, and if the police isn't watching you might be able to buy the pirate copy for cheaper from the same store.
Are you in the correct thread?
Posted August 19, 2016
avatar
igrok: Sooo, has anyone managed to run War Wind successfully? It crashes for me when trying to play the intro, which is in avi format with ffcvid codec. I renamed the folder with the movies and the game runs, but I wonder whether there is a way to make the in-game movies playable.
You might need a winetrick or three. Can you post the terminal output in a pastebin somewhere?
Posted August 19, 2016
avatar
JudasIscariot: You might need a winetrick or three. Can you post the terminal output in a pastebin somewhere?
avatar
igrok: Here is the pastebin link http://pastebin.com/5ggB8Y1d
It seems there is a problem with cinepak codec similar to a very old winehq report.
Hmmm, your version mentions wine-1.8. Any chance you can test this on Wine 1.8.3 or Wine 1.9.16?
Pages:
1
2
3
4
5
6
7
8
9
This is my favourite topic
Community
General discussionThe "Judas™ does this run in Wine" thread v1.173(128 posts)(128 posts)
(128 posts)