Crash to Desktop on Ancient Era on the same turn

[kilem]

Hello !
I'd appreciate a little feedback on how I can improve the stability on my game:
I'm playing a snail game (6000 turns) and I have a CTD on the same turn that prevents me from going further (still on ancient era).
Here is the savegame and the minidump.
I'm playing on windows 10 (64bits)/ 6 GB RAM/ core I7 4700 HQ 2.4 Ghz If you need any further information I'll be glad to answer your questions.

Any help will be much appreciated

 

[KaTiON_PT]

Got this when running the debug DLL.

Assert Failed

File:  CvCityAI.cpp
Line:  17780
SVN-Rev:  10103
Expression:  AI_buildingValueThresholdOriginal(eBuilding, iFocusFlags, iThreshold, bMaximizeFlaggedValue) == 0

Hard crash too, I can't manage to breakpoint it with visual express... Meaning I can't even see what's the thing that's causing this.

The thread 'Win32 Thread' (0x1654) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x24e4) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x291c) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x2174) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x830) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x1968) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x32c) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x2900) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x1298) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x8c4) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x1e68) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x1490) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x2348) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x984) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x1f9c) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x2048) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x1480) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x1520) has exited with code -1073741819 (0xc0000005).
The program '[11416] Civ4BeyondSword.exe: Native' has exited with code -1073741819 (0xc0000005).

 

[kilem]

Thank you for your reply KaTiON_PT.
Maybe someone else will have an idea on what to do?

 

[KaTiON_PT]

@Thunderbrd is our expert but he's taking a much needed break.

 

[Thunderbrd]

Got this when running the debug DLL.

Assert Failed

File:  CvCityAI.cpp
Line:  17780
SVN-Rev:  10103
Expression:  AI_buildingValueThresholdOriginal(eBuilding, iFocusFlags, iThreshold, bMaximizeFlaggedValue) == 0

Hard crash too, I can't manage to breakpoint it with visual express... Meaning I can't even see what's the thing that's causing this.

The thread 'Win32 Thread' (0x1654) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x24e4) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x291c) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x2174) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x830) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x1968) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x32c) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x2900) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x1298) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x8c4) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x1e68) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x1490) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x2348) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x984) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x1f9c) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x2048) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x1480) has exited with code -1073741819 (0xc0000005).
The thread 'Win32 Thread' (0x1520) has exited with code -1073741819 (0xc0000005).
The program '[11416] Civ4BeyondSword.exe: Native' has exited with code -1073741819 (0xc0000005).

Check what exceptions you are selecting will throw a pause in the processing. I often find you can't run with all of them on (check under the debug tab->exceptions in VS) without causing a bunch of unnecessary stops but if you aren't getting the game to hit a stop point before the crash while you're running with the debug dll attached to the process in VS then check the exceptions and make sure they're all on, or find a good break point just before the crash where you can turn them all on.

@kilem - I'm not in a position to look into this right away but maybe soon - but I'll lose this report if it's not made in the regular bugs and crashes thread and make sure that you're running on the latest SVN version with the posted game and that the mod folder is named caveman2cosmos or I won't be able to look into it.

 

[KaTiON_PT]

Managed to process the turn, interesting...

I did get a couple more different assert failures this time, twice for each:

Assert Failed

File:  CvSelectionGroup.cpp
Line:  6920
SVN-Rev:  10103
Expression:  pPlot != getPath().begin().plot() || pPlot == getPath().lastPlot()

Assert Failed

File:  CvPathGenerator.cpp
Line:  2922
SVN-Rev:  10103
Expression:  !newNode->m_bIsKnownRoute || node->m_iCostTo + iEdgeCost == newNode->m_iCostTo

@Thunderbrd I notice that in debug you can ignore these assert failures and the game processes each turn. But once in the normal dll, as you can't ignore them, it forces the game to crash.

Any precedence to this case?

 

[JosEPh_II]

I'm playing a snail game (6000 turns) and I have a CTD on the same turn that prevents me from going further (still on ancient era).

Snail is not 6000 turns. Is this an old save from a previous C2C version now trying to be played on 10103 svn?

 

[raxo2222]

Snail is not 6000 turns. Is this an old save from a previous C2C version now trying to be played on 10103 svn?

He is playing on Epic speed not Snail - he probably forgot what speed hes actually playing.
Game progressed normally for me.

I guess some bug got fixed in SVN or by Toffer's Interface Overhaul.

Hello !
I'd appreciate a little feedback on how I can improve the stability on my game:
I'm playing a snail game (6000 turns) and I have a CTD on the same turn that prevents me from going further (still on ancient era).
Here is the savegame and the minidump.
I'm playing on windows 10 (64bits)/ 6 GB RAM/ core I7 4700 HQ 2.4 Ghz If you need any further information I'll be glad to answer your questions.

Any help will be much appreciated

You have Revolutions AND Minor Civs on.
AI is less competitive with those options, but I guess Monarch/Prince player could play on Immortal/Deity/Nightmare with it.
AIs then will have much faster research/production speed.

 

[kilem]

Check what exceptions you are selecting will throw a pause in the processing. I often find you can't run with all of them on (check under the debug tab->exceptions in VS) without causing a bunch of unnecessary stops but if you aren't getting the game to hit a stop point before the crash while you're running with the debug dll attached to the process in VS then check the exceptions and make sure they're all on, or find a good break point just before the crash where you can turn them all on.

@kilem - I'm not in a position to look into this right away but maybe soon - but I'll lose this report if it's not made in the regular bugs and crashes thread and make sure that you're running on the latest SVN version with the posted game and that the mod folder is named caveman2cosmos or I won't be able to look into it.

Thank you for the answer
This thread should already be in the regular bugs and crashes thread (if I'm not mistaken) and the mod folder is indeed named caveman2cosmos. As for the SVN if it's necessary I'll update; but once there was an SVN update that changed the game duration (in terms of turns) and it ended up messing up my save (that was something like a year ago) so I tend to wait for "full" releases. But again If necessary I'll do the SVN update (currently it's the v38.5 - latest "full release")

Snail is not 6000 turns. Is this an old save from a previous C2C version now trying to be played on 10103 svn?

Yes I forgot which speed I am on... this is v38.5 (no SVN updates though)

He is playing on Epic speed not Snail - he probably forgot what speed hes actually playing.
Game progressed normally for me.

I guess some bug got fixed in SVN or by Toffer's Interface Overhaul.


You have Revolutions AND Minor Civs on.
AI is less competitive with those options, but I guess Monarch/Prince player could play on Immortal/Deity/Nightmare with it.
AIs then will have much faster research/production speed.

I initially started with Monarch, then used that option to increase the difficulty every 50 turns; changed the turns 'turnaround' to 1 turn, and then deactivated it. I should be on Emperor now.

 

[raxo2222]

Thank you for the answer
This thread should already be in the regular bugs and crashes thread (if I'm not mistake) and the mod folder is indeed named caveman2cosmos. As for the SVN if it's necessary I'll update; but once there was an SVN update that changed the game duration (in terms of turns) and it ended up messing up my save (that was something like a year ago) so I tend to wait for "full" releases. But again If necessary I'll do the SVN update (currently it's the v38.5 - latest "full release")

One of SVN revisions changed turn amount of all game speeds and added two new game speeds and another was purposefully breaking save compatibility due to bugs with AI introduced in V38.1.
That it was before V38.5.

It was real mess

 

[kilem]

So should I do the SVN update? will my savegame stll be playable?

 

[Thunderbrd]

Managed to process the turn, interesting...

I did get a couple more different assert failures this time, twice for each:

Assert Failed

File:  CvSelectionGroup.cpp
Line:  6920
SVN-Rev:  10103
Expression:  pPlot != getPath().begin().plot() || pPlot == getPath().lastPlot()

Assert Failed

File:  CvPathGenerator.cpp
Line:  2922
SVN-Rev:  10103
Expression:  !newNode->m_bIsKnownRoute || node->m_iCostTo + iEdgeCost == newNode->m_iCostTo

@Thunderbrd I notice that in debug you can ignore these assert failures and the game processes each turn. But once in the normal dll, as you can't ignore them, it forces the game to crash.

Any precedence to this case?

Sometimes its not the assert that's pointing to the real issue but is just a hint as to what's being setup to be a problem. On the last assert you get, stop the process and then walk through the process one step at a time until you go as far as you can before the actual crash. That's going to reveal a lot more. You might be surprised how meaningless the last assert is and that it's actually something else entirely down the way, which may be a graphic bug or something happening in the exe.

 

[Thunderbrd]

So should I do the SVN update? will my savegame stll be playable?

Should be playable but the more important thing to know is if it isn't, I can't debug it for you and have you be able to keep playing it anyhow. The debug would be corrected on the latest SVN once patched.

 

[kilem]

Should be playable but the more important thing to know is if it isn't, I can't debug it for you and have you be able to keep playing it anyhow. The debug would be corrected on the latest SVN once patched.

All right I will do the SVN update and see what happens.
I'll keep you posted

 

[kilem]

After the SVN update the turn progressed normally !!
Except for the year change (from BC-13834 to BC-8767) and a couple of new buildings everything seems fine so far.
Thank you everyone !

 

[raxo2222]

After the SVN update the turn progressed normally !!
Except for the year change (from BC-13834 to BC-8767) and a couple of new buildings everything seems fine so far.
Thank you everyone !

Tech tree was revised a little, and tech redundancies in tech/building requirements were removed except for natural wonders (Language + other tech), achievements (Writing + other tech), group wonder universities (Education + other tech), guilds (Guilds + other tech) and corporations (Corporations + other tech)

Calendar was readjusted and tech/building/unit cost were increased, so Prehistoric era should end at 10% of game.
At 8% game (10 000 BC) calendar slows down and goes to 6000 BC at 10% of game.
As for later eras if you have fairly steady tech rate (turns per tech) you should reach checkpoints fairly close.
I was aiming for half of skippable techs being researched before going to next era.
As for other half - they could be gotten for free or traded.
Kation added Upscaled Research cost options - Nightmare players, who trade techs and have TD/WFL on for everyone should be able to keep up with calendar and not overshoot easily (techs are 40% more expensive).
If still they can overshoot, then there is Upscaled Production Cost.

Buildings and resources were revised - now buildings should be buildable, when they are unlocked.
Few buildings got additional or had replaced tech requirements - in some cases requirement removals didn't make sense.

Ramping up production should be easier in new cities in later eras - minimum +/% yield from building is higher in later eras.

 

[kilem]

Awesome thank you very much @raxo2222 !!
Quick question though; after the SVN update one of my cities has a sulfur resource right under it (wasn't the case before as now it seems that sulfur appears with mining which I already have discovered). But when I hover over it, it still says "has 0" (even after a couple of turns).
Would you happen to know if I need to have a specific building to access it? I though that having cities on top of resources "provided" them automatically.

 

[raxo2222]

Awesome thank you very much @raxo2222 !!
Quick question though; after the SVN update one of my cities has a sulfur resource right under it (wasn't the case before as now it seems that sulfur appears with mining which I already have discovered). But when I hover over it, it still says "has 0" (even after a couple of turns).
Would you happen to know if I need to have a specific building to access it? I though that having cities on top of resources "provided" them automatically.

Sulfur is revealed by Mining - buildings, that need this resource in vicinity, can be built now, if other requirements are met.
It is enabled at Ancient Ballistics - then rest of empire can use it and can be traded.

Now all resources are revealed (visible on map and usable by vicinity buildings (raw vicinity for resource without improvement)) before or on tech, that enable them (resource is tradable and usable anywhere).

 

[kilem]

All right thank you very much

 

[KaTiON_PT]

@kilem recalculate modifiers by pressing CTRL+SHIFT+T.