[Bug 39471] Elminage Gothic freezes after clicking on "New Game"

[wine-bugs at winehq.org]

https://bugs.winehq.org/show_bug.cgi?id=39471

--- Comment #21 from fjfrackiewicz at gmail.com ---
(In reply to Andrew Eikum from comment #20)
> (In reply to fjfrackiewicz from comment #19)
> > (In reply to Andrew Eikum from comment #18)
> > > Are there two issues here? One where the intro movie freezes the game
> > > (comment #0), and a separate one where you got a crash after an hour of
> > > gametime (comment #1, comment #2)? I would expect the patch to fix the
> > > latter, but probably not the former.
> > 
> > It looks like it from where I stand. I think the "hour of game time" crash
> > might've been a random crash because I couldn't duplicate it. 
> > 
> > If I could remove the unnecessary comments, I would.
> 
> No worries. We'll keep this one about the intro movie bug.
> 
> I received a private email from someone else about the openal-1.15 bug a
> while ago, and their crash looks just like your crash. So I'm fairly
> confident that's what gave you that crash. I'll submit that patch very soon,
> I just need to give it a little more testing.
> 
> Anyway, back to the intro movie problem. The game repeatedly calls
> acmStreamOpen trying to convert some audio from WMA to PCM:
> 
> 0065:trace:msacm:acmStreamOpen ((nil), (nil), 0x506a8f0, 0x50a2ca0, (nil),
> 0, 0, 5)
> 0065:trace:msacm:acmStreamOpen src [wFormatTag=353, nChannels=2,
> nSamplesPerSec=44100, nAvgBytesPerSec=16005, nBlockAlign=2973,
> wBitsPerSample=16, cbSize=10]
> 0065:trace:msacm:acmStreamOpen dst [wFormatTag=1, nChannels=2,
> nSamplesPerSec=96000, nAvgBytesPerSec=192000, nBlockAlign=2,
> wBitsPerSample=8, cbSize=0]
> 0065:trace:msacm:acmStreamOpen => (512)
> 
> This continually fails because Wine's msacm32 can't decode WMA audio. I
> guess the game has a bug where it doesn't handle this failure gracefully,
> and eventually runs out of stack space, causing your freeze/crash.
> 
> So, either Wine's msacm32 should support converting WMA to PCM, or we're
> somehow misleading this program into thinking that it can do the conversion
> when it can't. Writing a test with the same acmStreamOpen call and running
> it on Windows would confirm what Windows's msacm32 can do, and tell us what
> Wine's should do.

I don't have access to a Windows machine at the moment. Is there anything I
could do on my end to help resolve this problem, provided I get access to a
Windows machine? Please keep in mind I probably can't write a test for
acmStreamopen so there's that limitation.

-- 
Do not reply to this email, post in Bugzilla using the
above URL to reply.
You are receiving this mail because:
You are watching all bug changes.