PROJECT: Playtest Thread III

[Protra3211]

The playtest 10/23/05 I have not had any crashes yet. Played about 300 turns . But did have the problem with the units vanishing at times. Still it only happend 3 or 4 times.
Ctp2 can take alot of ram if your playing many ai civs with large map. My old computor only had 128 megs of ram and by the time you reached modern age in Ctp2 the computor could bog down which can cause problems. My new PC zips along just fine.
Just a side note: I think its good that E makes a build (thank you E) that we can try out to see if it should be included in the next main code updates for a new playtest

[Ekmek]

Wow, its good to see life here

First on my builds: My builds are basically "patches" (loosely) to the latest playtest. To use them you install the latest playtest 10-23-05 and then you DL my build and copy over it.

Because I dont know enough to debug but have had moderate success in adding things I put them in a build for me to test out.

I have kept the"purists" like Tamerlin in mind and stuck to having my features an option by picking flags in the database, so if there are none of the flags they shouldnt interfere with the normal game.

I have been posting my builds for download mainly because there is a long time between "official" playtests. I remember #2 having bugs and fixes were posted to the code but a exe was not compiled so the community was stuck with #2. My builds allow for the latest of the revision (when I keep up) and that way we can test some of the debugging.

With that I think the vanishing unit problem seen in 10-23 was fixed and I included in my build and 10-23 with my build on it looks real stable from the times I played it.

Fugi thanks for the input and bringing some life in here (it gets quiet time to time). Please try the 10-23 playtest and with my build on it and let US know if there are anymore bugs thanks!

[Fugi the Great]

Thanks E, will do. Downloading now.

[Tamerlin]

Originally posted by Maquiladora
I cant see the logic in complaining when you find a bug in the playtest, thats all.

Read again Maq, I am not complaining at all.

Happy New Year to everybody...

[Tamerlin]

Originally posted by E
I have kept the"purists" like Tamerlin in mind and stuck to having my features an option by picking flags in the database, so if there are none of the flags they shouldnt interfere with the normal game.

Thank you E !

[Maquiladora]

Originally posted by Tamerlin

Read again Maq, I am not complaining at all.

Actually i was only referring to Fugis comment, complaining of finding bugs in the playtest, which was bizarre at the time.

[Fugi the Great]

Posting bugs from first game with E's source code since this is thread is about troubleshooting.

Game was stable up to turn 275, got a little tired so I saved the game and exited the program. Instantly wallpaper popped up but at the wrong screen resolution, still at the game resolution. Opened up Task Manager and noticed that CTP.exe was still running, using 99% of CPU and 317,796 K of memory. Shutdown CTP from Task Manager and everything went back to normal. So that problem is still there.

Reload savegave the next day, play to 1395AD - turn 339, launched an attack on an enemy city and got this after all cannon fired:

[Fugi the Great]

After I took the city, which was a cake walk especially when you have 5 full armies bombarding the city, I moved the 4 and half armies out of the area to attack the next city and left 6 defenders, 3 Infantry and 3 Cannon. From the first picture you can clearly see that I have clicked on the city and you can see the units inside of it, 6 total, but it only shows the number 3 and it shows that it is highlighted on the infantry unit. The second picture was when I clicked on the cannon unit, the highlight for the city disappeared and the cannon are invisible. So the game sort-of still has the invisible unit problem, or a variation of it. The next city I bombarded look the same way as my previous post, but it seemed that the doubling effect didn't start until the cannon did the automatic bombard the next turn after the initial bombardment. (anything to do with the automatic bombard code?) With all these strange display problems, it makes me wonder if maybe there is something wrong with the way this game got linked to Direct X. (does it need to be?) There must be something different that was done between Playtest 1 and the following versions since I have no problems at all with that one. Though it is a little strange that it worked fine up to this point, but this about the time where the last few playtests start acting up for me, course this is the first city I attacked after my newly invented cannon.

Don't know if this has anything to do with it but the blue player (middle player) - Korea is taking a extremely long time to take his turn. The music starts as soon as his turn is done, the last half of the other AI player take their turn, I take my turn, the first half take their turn, his turn is up, the song ends normally, waits a two minutes (no animation is going on at this time as well - all trade routes stop), he finally is done and next song starts. Almost looks frozen during those two minutes, can not click on anything during that time. Other than that, it is a little stange that the AI keeps sending in un- excorted Settler units into my territory (we are at war). The Carthage, the one that I am attacking in these pictures had 3 cities left, but it still wasted his time sending in settler units that I keep enslaving and he also sends in slaver units of his own that just end up dead, wouldn't have made more sense to build military units instead and try to retake those cities that I took from him? And I thought that this game was supposed to get rid of need for ICS, but Carthage had 6 other cities I took from him earlier that were only 4 spaces away from the next city. Will try more tonight.

[J Bytheway]

You say that the problems started when you started attacking with cannon. Had you used any other bombarding units before that? If not, then I suspect that the changes to the bombard code are at fault. We already know taht there's something wrong with them, because of the order doubling effect.

[Fugi the Great]

Sorry for not replying sooner, but no, I had some catapult earlier in the game, but never used them.

currently on turn 516, game is still allowing me to play, unlike the last few playtests that would have thrown errors that would not be get past the 400th turn. The Blue player is now part of the empire, so now the orange player is taking the longest time, up to 11 minutes to take its turn. The AI normally only took 30 seconds to take their turn during other playtests. I don't think this has anything to do with the player, there must be something going on in the game. I started playing with 22 empires, 2 more were added by revolts of other empires. The blue player was the one taking up all the time until his country was added, now the dark orange player is doing the same thing. Orange is one player position away from the blue player. Is there anything that happens different when the middle player is up or ended his turn?

Whoever is modifying the optimize sliders, you might still have a problem there. The first picture attached is right after I pushed the Optimize Specialist button. I never usually use it, but I thought I would see what it would do. It assigned 7 people as farmers, on the second picture is shows that those 7 farmers are not adding any food to the city, so there is no need for them, so I added them back to the pool and the city is now more efficient.

[Fugi the Great]

Not sure what is going on here. For some reason one of the artillery units was acting a little weird and would hide the number of units in a square, put it in a city and fortified it but the image of it would not disappear, so then I pulled it out and added it to the army and you can see the result, also the top 12 in that stack is blinking. Hope that helps.

[Ekmek]

I updated my build to 539 its just some slight modifications to the stuff I added, sneak attck works better (you can attack sneakattack defenders now without war), good that have productionpercent flag now work, its something I want to add to OIl to make it a decisive resource in late games, andI fixed tenew combat modifiers I wasnt addressing the float in the code.

[Ekmek]

Rev540 build has been uploaded and in the signature!

I had a pretty good night of coding and got 5 things I really wanted implemented.

The significant changes are:

-Settlers can now add to a city!

New Flags are:
-CantPillage flag for tile improvements now works (finally)
-NonLethalBombard implemented; now with this flag bombard units dont destroy units (but weakens them)
-CanBombardTiles implemented; with this flag, Tile Improvements can now be bombarded and destroyed
-CollateralTileDamage implemeted; if unit has this flag it will kill units and the the tile imps they are on (dont give it NLB though)


I figure I'll have to update my EMOD to incorporate the changes...

[Fugi the Great]

Guess I'll have to give that one a try, the game I currently playing is only on turn 583, CTD on turn 581, reload autosave - okay, then another CTD on 583, reload okay, play turn and save, then exit program. That game was getting a little unstable, AI turns were over 18 minutes, kind of hard to want to play a game when you only can play a hand full of turns at a sitting, I'm just glad that I had something to do on my other computer that needed doing at the same time, so it wasn't a complete waste of time waiting for my turn.

[Ekmek]

not sure what the long turns are for, maybe the AI pathing? I should also note that my playtest builds are not optimized. I cant really explain what that means or if it is the reason for the slowness, I've just been told many times that because of the compiler I'm using my builds aren't optimized.

[Fugi the Great]

Rev540
CTD turn 28 - reload autosave - okay
CTD turn 44 - reload autosave - okay
CTD turn 135 - reload autosave - okay
CTD same spot, could not continue game.

Well, so much for that one.

crash.txt:

0x004cb6b0 [?GetSpecialForcesSupportModifier@MilitaryReadiness @@QBENJ@Z + 0x0]
0x00484034 [?GetDescription@ArmyEstablishEmbassyOrderEvent@@EA EXPADJ@Z + 0x14]
0x00501cb3 [?SaveUnits@MapFile@@QAE_NPAU_iobuf@@@Z + 0xb3]
0x004fe5a0 [?gameinit_Initialize@@YAJJJAAVCivArchive@@@Z + 0x12f0]
0x004fe30e [?gameinit_Initialize@@YAJJJAAVCivArchive@@@Z + 0x105e]
0x00494041 [?NotifyAdvance@CityData@@QAEXJ@Z + 0xa1]
0x0049d7e2 [?SetEnabled@CriticalMessagesPrefs@@QAEXPBD_N@Z + 0x172]
0x005ba9df [?CheckRequiredFields@RiskRecord@@QAEXPAVDBLexer@@@ Z + 0x3cf]
0x005ba976 [?CheckRequiredFields@RiskRecord@@QAEXPAVDBLexer@@@ Z + 0x366]
0x005bb5da [?Parse@SpecialAttackInfoRecord@@QAEJPAVDBLexer@@@Z + 0x3a]
0x005b9294 [?Serialize@ResourceRecord@@QAEXAAVCivArchive@@@Z + 0x204]
0x005bb438 [?Serialize@SpecialAttackInfoRecord@@QAEXAAVCivArch ive@@@Z + 0x78]
0x005bb3db [?Serialize@SpecialAttackInfoRecord@@QAEXAAVCivArch ive@@@Z + 0x1b]
0x0040f081 [?GetNextPosition@Anim@@QAEJJ@Z + 0x11]
0x00407dd0 [??0CivApp@@QAE@XZ + 0x10]
0x00407847 [?AddSearchPath@?$aui_Resource@Vaui_Image@@@@QAE?AW 4AUI_ERRCODE@@PAD@Z + 0xc7]
0x008a790c [?g_mapTokenData@@3PAUTokenData@@A + 0x9e5c]
0x77e814c7 [__onexitbegin + 0x773b5fef]

[Ekmek]

hmmmm,

I'm dont read the crash text too well o hopefully J bytheay, fromafar, or martin G or someone can spot the problem and put in a fix.

My guess would be a pathing problem, but I dont know why it happened so late in the game. Any chance you can tell what techlevel you are at or the player where the crash is? maybe its a late game order that is causing the crash.

Thanks for testing this, sorry about the problems

[Fugi the Great]

I know very little about the error codes used in this program, but the line:

0x00407847 [?AddSearchPath@?$aui_Resource@Vaui_Image@@@@QAE?AW 4AUI_ERRCODE@@PAD@Z + 0xc7]

sort of looks like it can't find a picture for a resource that you may have added. Am I even close?

well anyway, tried the cheat mode to see if anything was amiss by removing the fog of war, the turn completed okay, so I reloaded the autosave again and got the game up to turn 139 where it crashed again.

Crash.txt:
0x004cb6b0 [?GetSpecialForcesSupportModifier@MilitaryReadiness @@QBENJ@Z + 0x0]
0x00484034 [?GetDescription@ArmyEstablishEmbassyOrderEvent@@EA EXPADJ@Z + 0x14]
0x00501cb3 [?SaveUnits@MapFile@@QAE_NPAU_iobuf@@@Z + 0xb3]
0x004fe5a0 [?gameinit_Initialize@@YAJJJAAVCivArchive@@@Z + 0x12f0]
0x004fe30e [?gameinit_Initialize@@YAJJJAAVCivArchive@@@Z + 0x105e]
0x0048ec6b [?CalculateTradeRoutes@CityData@@QAEJ_N@Z + 0xcb]
0x004954ac [?GetValue@CityData@@QBEJXZ + 0x14c]
0x004900e3 [?CheckRiot@CityData@@QAEXXZ + 0x133]
0x0048fbaf [?BeginTurn@CityData@@QAEJXZ + 0x12f]
0x0048f1f6 [?SupportBuildings@CityData@@QAEJ_N@Z + 0x106]
0x0049c19f [?ReleaseCityName@CivilisationData@@QAEXJ@Z + 0xf]
0x005ba9df [?CheckRequiredFields@RiskRecord@@QAEXPAVDBLexer@@@ Z + 0x3cf]
0x005ba976 [?CheckRequiredFields@RiskRecord@@QAEXPAVDBLexer@@@ Z + 0x366]
0x005bb5da [?Parse@SpecialAttackInfoRecord@@QAEJPAVDBLexer@@@Z + 0x3a]
0x005b9294 [?Serialize@ResourceRecord@@QAEXAAVCivArchive@@@Z + 0x204]
0x005bb438 [?Serialize@SpecialAttackInfoRecord@@QAEXAAVCivArch ive@@@Z + 0x78]
0x005bb3db [?Serialize@SpecialAttackInfoRecord@@QAEXAAVCivArch ive@@@Z + 0x1b]
0x0040f081 [?GetNextPosition@Anim@@QAEJJ@Z + 0x11]
0x00407dd0 [??0CivApp@@QAE@XZ + 0x10]
0x00407847 [?AddSearchPath@?$aui_Resource@Vaui_Image@@@@QAE?AW 4AUI_ERRCODE@@PAD@Z + 0xc7]
0x008a790c [?g_mapTokenData@@3PAUTokenData@@A + 0x9e5c]
0x77e814c7 [__onexitbegin + 0x773b5fef]


Well anyway, tried the game on the German side since it seemed they were the ones bombing the game out, and as soon as I ended my turn - bam CTD. Tried it as the French (the next player) and it finished the turn with no problem. Went back to the German player the next turn - bam CTD. Hope that helps. Game is saved, so if anyone needs it for troubleshooting...

Aren't cities supposed to be spread out a little further than this? (see below) How are they supposed to grow? Those two cities up in the mountains aren't going to be very useful, no growth and nothing but specialist farmers just to keep the population where it is, no one left to build other stuff with. The French had four cities practically on top of each other - two spaces between cities so they have no chance to expand.

[Ekmek]

Fugi,
Thanks for helping on this. I'm assuming you only added the Rev540 build and not my other EMOD right? Also if it is just the Rev540 did you copy everything in the zip over? There are some additional files there (but nothing I added)

Until someone better comes along any chance you can check out the Germans and see what they are buildinig in their cities? That might be the problem. I remember (a long time ago) I was having a build problem like that too but I think it was a mod of mine (IIRC I think the problem was building an oil refinery for some reason). It was around the industrial age it happened, is that where you are at? Are there any unfinished tile improvements? or wonders?

feel free to post your save game and I'll try to give it a shot.


As for the AI building, sometimes they build close depending on the situation for operational depth IIRC when they were discussing how the AI will build its empire. The mountain cities are a problemit looks like, but the AI will treat them as forts I think. However in my EMOD I did make where the AI can only build on grass, plains, and deserts and it settled pretty good.

[Fugi the Great]

You assume correct. And just to make sure that all needed files were moved over I moved all the files, not just the executable that I did last time (since that was the only new file), but apparently something is different since I now get a whole mess of Out of Bound errors that weren't there before when hit the turn button to see if that was the problem.
As for what the Germans were building: 3 catapults, 1 Granary, 2 Hoplites, 4 settlers, and one slaver. The slaver is the only one that would be completed next turn, or the turn that would screw up, and a settler out near the edge of the empire moves back toward the safety of the empire when seeing a neighboring military unit. The only unfinished tiles that I can see are in the neighbor to the north, probably a road since it is on a swamp.

Well anyway, started a different game, playing as Etruscans, player one this time, 22 empires, huge map, doughnut world. Everything was fine until turn CTD on turn 58. Reloaded from autosave okay and played okay until turn 68 with another CTD - no matter how many times I reloaded the autosave, the game refused to go further.

crash.txt:

0x004cb6b0 [?GetSpecialForcesSupportModifier@MilitaryReadiness @@QBENJ@Z + 0x0]
0x00484034 [?GetDescription@ArmyEstablishEmbassyOrderEvent@@EA EXPADJ@Z + 0x14]
0x00501cb3 [?SaveUnits@MapFile@@QAE_NPAU_iobuf@@@Z + 0xb3]
0x004fe5a0 [?gameinit_Initialize@@YAJJJAAVCivArchive@@@Z + 0x12f0]
0x004fe30e [?gameinit_Initialize@@YAJJJAAVCivArchive@@@Z + 0x105e]
0x00494041 [?NotifyAdvance@CityData@@QAEXJ@Z + 0xa1]
0x0049d7e2 [?SetEnabled@CriticalMessagesPrefs@@QAEXPBD_N@Z + 0x172]
0x005ba9df [?CheckRequiredFields@RiskRecord@@QAEXPAVDBLexer@@@ Z + 0x3cf]
0x005ba976 [?CheckRequiredFields@RiskRecord@@QAEXPAVDBLexer@@@ Z + 0x366]
0x005bb5da [?Parse@SpecialAttackInfoRecord@@QAEJPAVDBLexer@@@Z + 0x3a]
0x005b9294 [?Serialize@ResourceRecord@@QAEXAAVCivArchive@@@Z + 0x204]
0x005bb438 [?Serialize@SpecialAttackInfoRecord@@QAEXAAVCivArch ive@@@Z + 0x78]
0x005bb3db [?Serialize@SpecialAttackInfoRecord@@QAEXAAVCivArch ive@@@Z + 0x1b]
0x0040f081 [?GetNextPosition@Anim@@QAEJJ@Z + 0x11]
0x00407dd0 [??0CivApp@@QAE@XZ + 0x10]
0x00407847 [?AddSearchPath@?$aui_Resource@Vaui_Image@@@@QAE?AW 4AUI_ERRCODE@@PAD@Z + 0xc7]
0x008a790c [?g_mapTokenData@@3PAUTokenData@@A + 0x9e5c]
0x77e814c7 [__onexitbegin + 0x773b5fef]

Oh well, back to Playtest 1.

Game save from first game attached:

[Ekmek]

I compiled a rev540 Debug version and the last bunch is this:

    c3debug.cpp@320 : Assertion (!targets.empty()) Failed in File:F:\SVN-Code\trunk\ctp2_code\ai\mapanalysis\SettleMap.cpp, Line:228
    c3debug.cpp@323 : Stack Trace: '  0x00457720  [(unknown) + 0x0]
  0x00452174  [(unknown) + 0x0]
  0x00b5af1e  [(unknown) + 0x0]
  0x00b73550  [(unknown) + 0x0]
  0x00b715a5  [(unknown) + 0x0]
  0x006206e1  [(unknown) + 0x0]
  0x006ed906  [(unknown) + 0x0]
  0x006ed831  [(unknown) + 0x0]
  0x006ef8d5  [(unknown) + 0x0]
  0x006eb3b2  [(unknown) + 0x0]
  0x006ef4a5  [(unknown) + 0x0]
  0x006ef3de  [(unknown) + 0x0]
  0x006ef2d9  [(unknown) + 0x0]
  0x006ef0c6  [(unknown) + 0x0]
  0x004b3071  [(unknown) + 0x0]
  0x004a8f44  [(unknown) + 0x0]
  0x004a660a  [(unknown) + 0x0]
  0x0046d39f  [(unknown) + 0x0]
  0x0046d8de  [(unknown) + 0x0]
  0x00461c2e  [(unknown) + 0x0]
  0x00461233  [(unknown) + 0x0]
  0x00b88fc3  [(unknown) + 0x0]
  0x7c816d4f  [(unknown) + 0x0]

the last repeating stuff is the stack, but it does appear the mapanalysis is the problem.

The line in question has something to do with settle target, Can someone with better programming skills help?

[Ekmek]

Fugi,

I messed around with it some more. Sometimes I could go another turn with the save you posted but most times I get the crash.

I thought at first maybe my settlecity code did it but i removed it and got the same error. So after messing wiith it I think the error might be with the DOUGHNUT shape. it might be throwing off the ai settling targets. You might try loadingit with playest 1 but not sure it will work. ANy chance you are willing to try a normal world map (with less players) and a little smaller just to get through a whole game?

[Fromafar]

The assert in SettleMap.cpp is harmless. From what I can see, it just tells you that there were no (good enough) spots to settle. The first real crash is in ArmyData::Bombard. There are a couple of GetNonLethalBombard calls where you are using the defender index while looking in the attacker list (and the other way around).

And don't forget to put the proper map file in your next distribution. The generated crash.txt is useless when the exe and map files do not match.

[Ekmek]

thanks fromafar,
When I tested the code it seemed to work, and the reason I called the defender index while in the attacker list was because on the counterbombard stuff.

the original code looked like this:

    for (i = m_nElements - 1; 0 <= i; i--) { 
        if (m_array[i].GetHP() < 0.999) {
            m_array[i].KillUnit(CAUSE_REMOVE_ARMY_COUNTERBOMBARD, defender.GetOwner());  
        } 
    }
    
    //kill off defending units that were newly damaged
    for (i = defender.Num() - 1; 0 <= i; i--) { 
        if (defender[i].GetHP() < 0.999) {
            defender[i].KillUnit(CAUSE_REMOVE_ARMY_BOMBARD, GetOwner());

I thought I should add an attacker NonLethalBombard check before the killunits to preventthe unit from being killed. which appeared to work. should I take it before the 'for' statement then? I'll check it out.

thanks!

PS for the map files, I think this relates to the ctp2_debug.exe. Are they also in the ctp folder?

[Ekmek]

Fromafar,

I out commented the conflicting marray and defender lines (the nonlethal lines) and figured out the ctp2.map file so I got this crash text:

0x004cbc40 [?GameObj_Get@@YAPBVGAMEOBJ@@PAV1@K@Z + 0x9]
0x004841d4 [?Get@ObjPool@@QBEPBVGAMEOBJ@@K@Z + 0x24]
0x00502073 [?GetArmy@ArmyPool@@QBEPAVArmyData@@K@Z + 0x13]
0x004fe960 [?GetOwner@Unit@@QBEJXZ + 0x18]
0x004fe6ce [?GetDBRec@Unit@@QBEPBVUnitRecord@@XZ + 0x11]
0x00494198 [?Fight@ArmyData@@QAEJAAVCellUnitList@@@Z + 0x6fa]
0x0049d9e2 [?GEVHookCallback@BattleEvent@@EAE?AW4GAME_EVENT_HO OK_DISPOSITION@@W4GAME_EVENT@@PAVGameEventArgList@ @@Z + 0x79]
0x005ba8ef [?Run@GameEventHook@@AAE?AW4GAME_EVENT_ERR@@AAViter ator@?$list@UNode@GameEventHook@@V?$allocator@UNod e@GameEventHook@@@std@@@std@@PAVGameEventArgList@@ AAJ@Z + 0x63]
0x005ba886 [?Activate@GameEventHook@@QAE?AW4GAME_EVENT_ERR@@PA VGameEventArgList@@JAAJ@Z + 0x7e]
0x005bb44a [?ActivateHook@GameEventManager@@QAE?AW4GAME_EVENT_ ERR@@W4GAME_EVENT@@PAVGameEventArgList@@JAAJ@Z + 0x3a]
0x005b91a4 [?Process@GameEvent@@QAE?AW4GAME_EVENT_ERR@@AAH@Z + 0x2f]
0x005bb2a8 [?ProcessHead@GameEventManager@@QAE?AW4GAME_EVENT_E RR@@XZ + 0x33]
0x005bb24b [?Process@GameEventManager@@QAE?AW4GAME_EVENT_ERR@@ XZ + 0x7b]
0x0040f041 [?Process@CivApp@@QAEJXZ + 0x103]
0x00407db0 [?CivMain@@YGHPAUHINSTANCE__@@0PADH@Z + 0x3b6]
0x00407827 [WinMain@16 + 0x74]
0x008a78bc [WinMainCRTStartup + 0x134]
0x7c816d4f [__onexitbegin + 0x7bc71e17]

With debug.exe I got this in the log

c3debug.cpp@320 : Assertion (!targets.empty()) Failed in File:F:\SVN-Code\trunk\ctp2_code\ai\mapanalysis\SettleMap.cpp, Line:228
c3debug.cpp@323 : Stack Trace: ' 0x00457720 [char * __cdecl c3debug_StackTrace(void) + 0x20]
0x00452174 [void __cdecl c3debug_Assert(char *,char *,int) + 0x44]
0x00b5ad6e [public: void __thiscall SettleMap::GetSettleTargets(long const &,bool const &,class std::list > &)const + 0x2be]
0x00b733a0 [private: static void __cdecl CtpAi::AddSettleTargets(long) + 0x150]
0x00b713f5 [public: static void __cdecl CtpAi::BeginTurn(long) + 0xd25]
0x006204d1 [private: virtual enum GAME_EVENT_HOOK_DISPOSITION __thiscall BeginTurnEvent::GEVHookCallback(enum GAME_EVENT,class GameEventArgList *) + 0xc1]
0x006ed6f6 [private: enum GAME_EVENT_ERR __thiscall GameEventHook::Run(class std::list >::iterator &,class GameEventArgList *,long &) + 0xa6]
0x006ed621 [public: enum GAME_EVENT_ERR __thiscall GameEventHook::Activate(class GameEventArgList *,long,long &) + 0x81]
0x006ef6c5 [public: enum GAME_EVENT_ERR __thiscall GameEventManager::ActivateHook(enum GAME_EVENT,class GameEventArgList *,long,long &) + 0x65]
0x006eb1a2 [public: enum GAME_EVENT_ERR __thiscall GameEvent::Process(int &) + 0x32]
0x006ef295 [public: enum GAME_EVENT_ERR __thiscall GameEventManager::ProcessHead(void) + 0x65]
0x006ef1ce [public: enum GAME_EVENT_ERR __thiscall GameEventManager::Process(void) + 0x7e]
0x006ef0c9 [public: enum GAME_EVENT_ERR __thiscall GameEventManager::ArglistAddEvent(enum GAME_EVENT_INSERT,enum GAME_EVENT,class GameEventArgList *) + 0x189]
0x006eeeb6 [public: enum GAME_EVENT_ERR __cdecl GameEventManager::AddEvent(enum GAME_EVENT_INSERT,enum GAME_EVENT,...) + 0x1b6]
0x004b3071 [void __cdecl dh_endTurn(class DQAction *,class Sequence *,enum DHEXECUTE) + 0x41]
0x004a8f44 [public: void __thiscall Director::HandleNextAction(void) + 0x124]
0x004a660a [public: void __thiscall Director::Process(void) + 0x2a]
0x0046d39f [public: long __thiscall CivApp::ProcessUI(unsigned long,unsigned long &) + 0x22f]
0x0046d8de [public: long __thiscall CivApp::Process(void) + 0xce]
0x00461c2e [int __stdcall CivMain(struct HINSTANCE__ *,struct HINSTANCE__ *,char *,int) + 0x3be]
0x00461233 [WinMain@16 + 0x83]
0x00b88e13 [WinMainCRTStartup + 0x1b3]
0x7c816d4f [\177zlib1_NULL_THUNK_DATA + 0x7b46649f]
'

Can anyone help?


In the meantime, Fugi I reverted back to Build538 before I did some changes and posted a build in my signature. The good news is that I tested it on your save file and it works. So you can install the files in this 538 build (it includes the ctp2.map to get a proper crash.txt too) and hopefully you can finish your previous games!

[Fromafar]

Good news: your dumps start making sense now, and the non-empty targets assert is irrelevant. When you press continue/ignore at the pop-up when using the debug version, the program will run just fine. You may also remove the Assert from the code.

The other one is the real cause. From the first 4 lines, you can deduce that you are once more trying to access an invalid object. This time from ArmyData::Fight. Something that looks suspicious is the AllSneakAttack part. You may be accessing m_array[i] with an invalid i. But you should be able to find this yourself with the debugger.

[Ekmek]

Thanks Fromafar, I messed with trying to fix my sneak attack code but keep getting problems with the loop (I described it in my source code attempts thread).

I'm not sure I'll remove the Assert fro the code, I'll leave that to more experienced hands like yourself (dont want to outcomment the wrong thing).

[Martin Gühmann]

Originally posted by E
You might try loadingit with playest 1 but not sure it will work.

E how often did I say that we don't number the playtests like this?!

Once again for you - and for Fugi as well. Playtest versions are identfied by date so if a playtest was released on December the 9th 2003 than it is playtest from 2003-09-12 and nothing else. By the way the playtest I quoted is the only playtest that you can call Playtest 1. Because that's the only first playtest we ever released and we will never release a second first playtest.

These numberings have only to do something with the threads. And to make it clear again a new playtest thread is posted if the old one has reached or has exceeded the limit of 500 posts, like this one that is actually done for 37 posts now.

So continue there in PROJECT: Playtest (the 4th Thread).

-Martin