Repeated crashing to Geoscape

[Napoleon XIX]

Fairly frequently (I'd estimate for at least 1/2 the battles), I suddenly crash out of the battlescape to the geoscape, to find everything as it was just before the mission started. Attempting to do the mission again has the same result.

Running 2.4 on Windows 7

This is the relevant section of the log (my name replaced with 'xxxx'):

2012/09/24 17:05:38 xxxx has taken control over team 1.
2012/09/24 17:06:46 Soldier entered the rescue zone
2012/09/24 17:06:46 Soldier left the rescue zone
2012/09/24 17:06:55 Team 1 ended round
2012/09/24 17:06:55 Team 7's round started!
2012/09/24 17:06:59 Team 7 ended round
2012/09/24 17:06:59 Team 0's round started!
2012/09/24 17:07:04 Team 0 ended round
2012/09/24 17:07:04 Team 1's round started!
2012/09/24 17:07:26 [STATS] xxxx (Gurneet Weissenbach) kills alien (Zrokk Tsk Hrekk) with Full-Auto of machinegun_ammo (entnum: 29)
2012/09/24 17:09:19 Team 1 ended round
2012/09/24 17:09:19 Team 7's round started!
2012/09/24 17:09:22 Team 7 ended round
2012/09/24 17:09:22 Team 0's round started!
2012/09/24 17:09:22 ********************
2012/09/24 17:09:22 ERROR: LE_DoEndPathMove: Actor movement is out of sync: 120:153:0 should be 111:144:0 (step 5 of 5) (team 7)
2012/09/24 17:09:22 ********************
2012/09/24 17:09:22 Shutdown server: Server crashed.
2012/09/24 17:09:22 Unload the game library

I can try to provide a savegame just before this happens, if necessary (though not this exact one) - I have got this in all the games I have tried - 2 full and 1 short. Having looked at past logs, I believe the bug is the same one. Please may you provide a fix - it prevents me from playing an otherwise truly excellent game.

[Quizer]

I have the same problem - during many battles, I get thrown out of the battlescape, and it's like the battle never happened. This crash always happens during the AI's turn, never during my own. Here is the log of my latest reproduction of this bug:

updateact2012/12/01 23:21:37 (player 0) It's team 1's turn!
2012/12/01 23:21:37 Quizer has taken control over team 1.
2012/12/01 23:21:46 Soldier entered the rescue zone
2012/12/01 23:21:48 Soldier entered the rescue zone
2012/12/01 23:21:53 Soldier entered the rescue zone
2012/12/01 23:21:55 Soldier entered the rescue zone
2012/12/01 23:21:56 Soldier entered the rescue zone
2012/12/01 23:22:01 Team 1 ended round
2012/12/01 23:22:01 Team 7's round started!
2012/12/01 23:22:02 [STATS] alien (Akrokk Tt Treknk) kills civilian (b c) with Snap Shot of plaspistol_ammo (entnum: 130)
2012/12/01 23:22:05 [STATS] alien (Uoorgh Zr Zus Soufz) kills civilian (b c) with Snap Shot of plasrifle_ammo (entnum: 128)
2012/12/01 23:22:07 R_AnimChange: No model given (walk)
2012/12/01 23:22:07 LET_StartPathMove: Could not change anim of le: 138, team: 7, pnum: 9
2012/12/01 23:22:07 ********************
2012/12/01 23:22:07 ERROR: LE_DoEndPathMove: Actor movement is out of sync: 157:128:0 should be 159:137:0 (step 2 of 2) (team 7)
2012/12/01 23:22:07 ********************
2012/12/01 23:22:07 Shutdown server: Server crashed.
2012/12/01 23:22:07 Unload the game library

It seems the problem has to do with alien movement - apparently the alien isn't where the game expects it to be. It always happens to aliens, not civilians (all my occurrences of this bug say 'team 7' in the error line). It seems to me that this often happens when the aliens are shooting around a lot, but since only deaths are logged and violence may happen off-screen (as during this particular trial run), it is impossible for me to confirm either way.

If I retry the mission enough times, I can usually get to the end without this happening within a couple of attempts, but it severely hampers my enjoyment of the game. I decided to get the stable release instead of a development version, but this isn't turning out to be particularly stable.

I'm pretty sure this problem has long since been fixed, and I would like to know if a patch exists, or, failing that, a workaround, or whether I'll have to get a more recent development version to avoid this problem. I tried looking over the bug list on the sourceforge page, but wasn't able to find this bug among the huge number listed there. I also wasn't sure what version the 'open' and 'closed' tags apply to - just because something is labeled as 'closed' and thus presumably fixed, doesn't mean this is the case for this release which is more than half a year old now.

Anyway, if someone reads this, please let us know if there is a patch or workaround for this bug.


thank you for your time


Quizer

[ShipIt]

The new SF is a holy s***. The bugreport states the problem was fixed.


I don´t think there will be a patch for 2.4. First, the switch from pure C to C++ makes it harder to patch 2.4 up. And it would need additional manpower to do so. But, we made a huge step with 2.5-dev and I hope we can release it in the near future.

[Quizer]

But there is nothing I can do to work around or try to avoid this bug?

I wish there was a 2.5 stable release or something where this is fixed, but I guess people focused more on the development rather than the viability of the current release...

I'm kind of wary of spending another gig of my precious limited bandwidth on one of the dev releases, not knowing how stable and/or balanced they are. Apart from this glaring problem, I'm quite happy with the 2.4 version. All the features that are implemented are really well-polished. I like the story aspects and the flow of the campaign, like how you have to research that breathing apparatus to even get to where you can make stun rods. The slower progression makes the struggle to keep up with the aliens technologically more real. Until I went back to try the old X-COM again, I hadn't even noticed how much I'd started taking all the improvements and little conveniences for granted, such as being able to rotate the battlescape. I hope this project keeps coming along.


Quizer

[ShipIt]

2.5-dev is quite stable already. I played nearly eight in game months (up to end of Dez 2084) for testing without major problems. There is a problem with multithreading, but this can be disabled. Beside that, 2.5-dev is already a huge step forward. No more endless grinding +ufocrash maps, the reworked weapon balancing, adjusted research times, wounds and weight system, the expanded campaign and a lot of other stuff, definitely worth to check it out.

[DarkRain]

...
The bugreport states the problem was fixed.
...

Interesting, I guess that means that bug #2297 could be closed (at least once 2.5 is out)