I have just done a completly CLEAN re-install of WindowsXp pro, updated through MS update, including SP3. Installed the latest Nvidia forceware (v169.21) for my GF 7600 GT, latest nForce 4 (v6.86) drivers for my motherboard, and latest drivers from creative labs for my Audigy2 Platinum eX. (v02.09.0016, [i]I think.[/i])

I then got a clean copy of the "warzone2100_r5090-debug.exe" nightly build and installed it. played through from the start of the campaign, without any cheats, loaded up the transporter in Alpha 3. and as soon as it left the screen... it crashed as usual.

As far as I can tell the warnings about missing files relatre to the FMV sequences that arn't supposed to be there anyway. and teh segfault just happens to occur when it tries to load the away map for Aplha 3, just after it tries-to-play/skips a FMV sequence.

With --debug=all the reports in the .RPT (WinXP) and .gdmp (linux) change to:
C:\Program Files\Warzone 2100\warzone2100.exe caused an Access

Violation at location 0347f762 Reading from location 06010000.

Log message 1: [scrAddMessage] Adding FLIGHT pViewData=0353D748
Log message 2: [driveInitVars] driveInitVars: Driving

And for linux

Dump caused by signal: SIGSEGV: Invalid memory reference: Address not mapped to object

Log message 1: [scrAddMessage] Adding FLIGHT pViewData=0x8ecb4e8
Log message 2: [driveInitVars] driveInitVars: Driving

The linux .gdmp then goes on (as linux actuall has a debugger):

#0 0xb7bb6751 in waitpid () from /usr/lib/debug/libc.so.6
No locals.
#1 0x0821f229 in posixExceptionHandler (signum=11, siginfo=0xbfb18e6c,
sigcontext=0xbfb18eec) at exceptionhandler.c:469
btBuffer = {0x821efdc, 0xb7f2b440, 0x0 <repeats 18 times>}
btSize = 2
pid = 2271
gdbPipe = {14, 15}
dumpFile = 12
signal = 0x82e4de0 "SIGSEGV: Invalid memory reference: Address not mapped to object"
allreadyRunning = 1
#2 <signal handler called>
No symbol table info available.
#3 0xb7f282f4 in ?? ()
No symbol table info available.

I think I've done as much as I can to rule out bad drivers or corrupt data problems. Any ideas? If not maybe one of the others will have an idea?

Attached a dxdiag report detailing full system hardware and driver details.

Attached updated crash report from winXP and updated save game just before I launched the transport.

Sorry for the long post.

(file #4330, file #4331)

If it were just the graphical corruption I get under wine or a crash under ONE OS then I would tend to agree.

However for me it crashes under WinXP (XP-pro-SP2) AND Linux (Ubuntu 8.04 for i386).
While my windows nvidia drivers may be slightly out of date, I checked the drivers in Ubuntu and they indeed latest version of Nvidia's Unix driver (169.12).

As for the savegames, I honestly can't remember why I included the 'Alpha 4 Start' save and re-reading my original posts maybe I wasn't too clear.

The Segfault doesn't occur immediately on loading the save 'Alpha 3 Start', I occurs when you launch the transport. (after first researching the 'Power Module' tech and upgrading a power generator.)

I have tried uninstalling and reinstalling WZ several times, and playing the campaign through from the beginning in case it was save/data corruption, and I'm not getting any other problems with my GeForce 7600 GT card in any other DirectX or OpenGL apps.

I'm currently having some speed and reliability issues with my internet connection so I cant update my windows drivers at the moment, but I did manage to update my svn checkout of trunk to r5101 and compile under linux, but I still got the segfault.

Did a few quick tests using r4875-debug natively under WinXP-pro-SP2 (ie not under wine) and it crashed utterly.

Out of about 20 runs, Most just crashed bringing up the "This program as encountered a serious error and needs to close, please inform M$ so we can fix it, blah blah blah."

One test run caused the system to hang completely, and one caused the system to reboot.

attached archived stderr.txt and .RPT file. stdout.txt was empty.

Guess Wine IS more stable that windows after all.

(file #4308)

Added an Archive containing screenshots of the graphical corruption under wine. It occurs in both fullscreen and windowed modes

(file #4301)