[MOD] v41 SVN and GithubDesktop Bugs and Crashes Single player

[raxo2222]

It would have been, if it had been there. But the 1st post on the mod's thread says "Compatible across versions. Continue playing your current game of C2C - just save our current game and load it in the new version. You will be asked "Should the modifiers be recalculated?". Just say yes, wait awhile and then continue your game with all the new features."
So no, I wasn't told that before. Thanks for the info anyway.

You are first person to fall for this misleading message. Maybe no one reads it, as its hard to find without effort.
Moddb updates are once or twice a year roughly, and if save is broken, then its mentioned on new moddb release.

 

[Thunderbrd]

C2C has a much improved ability to not beak saves with every new revision made than most mods have. That doesn't mean ALL of the changes can avoid that. We try to limit how often it happens but now and then it does still require it.

 

[Mravac Kid]

"Hard to find without effort"? It's literally the first sentence on what seems to be the mod's main thread in this forum. I'd suggest amending it to be more correct, such as that old saves (before whatever version last did that) may be broken.
Anyways, to keep this from being spam, is there any way to rescue old saves? Will re-saving them in progressively newer versions of the mod enable them to be brought up to date, or is there a version that just makes them incompatible?

 

[Thunderbrd]

Once incompatible, you'd have to stick with the last version it was compatible on for the save to be playable. It's unrepairable after that point and requires a new game to play further if you want to progress with the design of the mod.

 

[Somebody613]

The reason is that new SVN versions (not even release versions) often change in-game MECHANICS significantly, which breaks it immediately.
RECALCULATION only refers to changed NUMBERS, not MECHANICS.
So, yeah, it's a "sad" fact about a constantly updated and improved mod like C2C, but it's unavoidable.
On the other hand, such changes don't happen THAT much often, so maybe don't start a game that you are then planning on dropping for a month or five of non-playing in the first place?
Just pointing out a "solution", that's all, lol.

 

[Toffer90]

Using the latest SVN (11453) but this also occurs in the previous version as well. Load the saved game (I am also using the MegaPack civilization MOD) and notice that the scoreboard in the lower right corner keeps flashing. I just destroyed the Arabian civ but the scoreboard keeps displaying the civ's score and then removing it. This civ was a vassal of Israel which is also on its last city.

Have you played this game over many SVN versions?
Game seems to be corrupt somehow, Saladin of Arabia has a wardog that is impossible to kill, can't be killed in worldbuilder, and game fails to kill that one unit when killing the player, so the player gets revived as it still has units after having been killed, then game tries to kill the player again as one wardog is not enough to keep him alive without "require complete kill", but again the wardog cannot be killed.

Your save might be slightly toast. Edit: Fixed the issue on git.

 

[JosEPh_II]

CTD repeatable. Go into Washington DC and try to upgrade any of the 3 Enforcers to a Town Watchmen. Will throw you out to Desktop every time.


EDIT: Just updated to latest SVN where Toffer fixed a CTD and no longer have this problem.

Good job Toffer

 

[Divizor]

Good afternoon. Should chemical weapons and improved radar be available to police dogs?

 

[Thunderbrd]

Good afternoon. Should chemical weapons and improved radar be available to police dogs?

My apologies that my ability to stay on top of promo access evaluations has not stayed up with our ability to ensure people could get this far into the game. lol @raxo2222 has been pretty good at figuring out how to resolve these kinds of things so thank you for bringing them up.

 

[raxo2222]

My apologies that my ability to stay on top of promo access evaluations has not stayed up with our ability to ensure people could get this far into the game. lol @raxo2222 has been pretty good at figuring out how to resolve these kinds of things so thank you for bringing them up.

Police Dog has unit combat of Gunpowder.
This unit combat makes units eligible for Improved Radar and Chemical promotions.
So maybe there should be exclusions made too.

 

[Thunderbrd]

Police Dog has unit combat of Gunpowder.
This unit combat makes units eligible for Improved Radar and Chemical promotions.
So maybe there should be exclusions made too.

Yeah - they have that because their handlers would be gun armed, and we ARE assuming this is a Dog 'UNIT' rather than JUST the dog (which we might assume actually is the case at Canine Superior).

But perhaps some things wouldn't be used with the animals or in concert with them pretty much ever, and that certainly makes an animal unit combat exclusion make sense.

 

[raxo2222]

Yeah - they have that because their handlers would be gun armed, and we ARE assuming this is a Dog 'UNIT' rather than JUST the dog (which we might assume actually is the case at Canine Superior).

But perhaps some things wouldn't be used with the animals or in concert with them pretty much ever, and that certainly makes an animal unit combat exclusion make sense.

Animals now are excluded from Improved Radar and Chemical promotions on Github.

 

[irishhombre]

Clone Workers also have access to combat promotions.

 

[raxo2222]

Clone Workers also have access to combat promotions.

@Toffer90 probably its equivalent issue as with modern worker

 

[SirJohnEh]

I stumbled onto a rarely-occurring bug that has noticeable impact I started this game at SVN level 11430 (V42.0.BETA.6724), and have not yet updated. I suspect it may have appeared well before this SVN level because it seems to only be triggered in 1% or 2% of newly-generated games that specifically include 1 or 2 randomly-selected AIs. I did not notice this bug until several hundred turns later when using the WB "player" capability to view and compare counts of resources and troops using the WB "player" function's pull-down list of the to select the player or any AI. When I opened this pull-down list I noticed something very strange and wrong. There were 2 entries for the same leader (Wu of China), both flagged with a red background. I then used WB to scan the complete map, and found 2 widely-separated settlement areas both belonging to Wu.

More information and/or guesses that may help locate and fix this bug, plus suggestions for players on how to minimize the chance of this bug occurring in their game
(1) I suspect this bug occurs only at game start-up and map generation, with a very low probability that a randomly-selected AI will be assigned the same civilization name as an already-named player or AI.
(2) I'm guessing the odds of this bug causing an issue in game generation is about 1 or 2 in a hundred (assuming each randomly-generated AI has about a 1% chance of picking the same civilization as the player or any specified AI or any previously-generated randomly-selected AI ( all in the same list of about 100 civilizations).

Suggestions to avoid this bug until the fix is completed:
1) Do not specify generating any randomly-selected AIs (unless you really really want them, and are willing to use the WB fix-ups described above to detect and erase all cities belonging to one of the two same-leader civilizations).
2) If you have just started a game and not yet played more than about 100 turns, I suggest using WB as above to check for duplicate civilizations. If there are, I suggest starting a new game (unless you really really love this map and are prepared to use WB to find and delete one of the duplicate civilizations. There will probably be some negotiation in a multi-player game with players close to or far from the now-available erased AI's settlement area) .
3) If you are several hundred turns into the game already (like me), want to continue this game, and are willing to use WB to find and erase all settled cities of one of the duplicates, then keep playing. An unexpected bonus is that that some players may may move up on the scoreboard because the scoreboard trustingly summed the impacts of the two same-name civilizations (which will stop happening immediately when the fix is made, or you use WB to erase one of them). This will definitely require a lot of negotiation in a multi-player game.

(2) I'm guessing the odds of this bug impacting game play during new game generation is only about 1 in 100, multiplied by the number of randomly-assigned AIs. This guess is based on the odds that each randomly-generated AI has about a 1% chance of picking the same civilization and leader as the specified player and all the specified AI and all of the randomly-selected AIs ( form a list of slightly over 100 civilizations).
(3) Theoretically this could happen in any game, if the game start-up options screen does not ensure that the player and all specified AIs have different civilization names. I don't this can occur.

Suggestions for other players to avoid this bug:
1) Do not generate any randomly-selected AIs (unless you really really want them, and are willing to use the WB fix-ups described above to detect this case and possibly to erase all cities belonging to one of the two same-leader civilizations (will probably require negotiation in a multi-player game).
2) If you recently started a new game and have not yet played more than about 100 turns, I suggest using WB as above to check for duplicate civilizations. If there are, I suggest starting a new game (unless you really really love this map). It is probably faster and easier than using WB to fix it. This approach will almost certainly require negotiation in a multi-player game.
3) If you are several hundred turns into the game already (like me) and want to continue, and are willing to use WB to find and erase all settled cities of one of the duplicates, then keep playing. An unexpected bonus is that that some real players and AIs will move up on the scoreboard because previously the scoreboard added the impacts of the two same-name civilizations - which will stop happening immediately if you use WB to erase one of them, or start a new game after the fix is implemented). This will definitely require a lot of negotiation in a multi-player game.

 

[Somebody613]

...

I dunno.
I definitely remember seen copies of the same civ on the Score Table (when trying something massive) - as separate civs.
Are you sure that this IS "two separate random AI civs are fused together at generation due to having the same name"?
Not some OTHER bug, more settlement-related or something?
Cause, again, I do recall having seen civ copies that don't mind coexisting as separate civs (and were clearly randomly generated at start-up, though not via Barbarians, which may be different).

 

[SirJohnEh]

I dunno.
I definitely remember seen copies of the same civ on the Score Table (when trying something massive) - as separate civs.
Are you sure that this IS "two separate random AI civs are fused together at generation due to having the same name"?
Not some OTHER bug, more settlement-related or something?
Cause, again, I do recall having seen civ copies that don't mind coexisting as separate civs (and were clearly randomly generated at start-up, though not via Barbarians, which may be different).

Please consider that the code base has been updated hundreds or even thousands of times since the game you remember. Some of these code changes may have accidentally or purposely changed the handling of duplicated AI civilizations.

Based on your comments, I'm tempted to run several short tests with the highest possible number of randomly-generated AIs (ideally 25 or more on a very large map). If my theory is correct, I should see this bug appear at least once by running several short games (each having maybe a 25% chance of seeing this bug happen based on starting each game 25 randomly selected AIs. Unfortunately my gutless laptop may not support this many AIs, or not support any map size greater than Large.

 

[Somebody613]

Please consider that the code base has been updated hundreds or even thousands of times since the game you remember. Some of these code changes may have accidentally or purposely changed the handling of duplicated AI civilizations.

Based on your comments, I'm tempted to run several short tests with the highest possible number of randomly-generated AIs (ideally 25 or more on a very large map). If my theory is correct, I should see this bug appear at least once by running several short games (each having maybe a 25% chance of seeing this bug happen based on starting each game 25 randomly selected AIs. Unfortunately my gutless laptop may not support this many AIs, or not support any map size greater than Large.

You can make a much quicker version of it.
It's entirely possible to cram 40 civs onto a DUEL map (but you must choose a script that makes it mostly LAND, so that they have enough space to be placed).
Then you can just use WorldBuilder to check the civs for duplicates.
Takes a couple minutes to produce, and another to check, loool.
Probably can even be done repeatedly without restarting the game (it tends to crash, when I do so for bigger maps, might even be specific for my computer specs).

 

[SirJohnEh]

You can make a much quicker version of it.
It's entirely possible to cram 40 civs onto a DUEL map (but you must choose a script that makes it mostly LAND, so that they have enough space to be placed).
Then you can just use WorldBuilder to check the civs for duplicates.
Takes a couple minutes to produce, and another to check, loool.
Probably can even be done repeatedly without restarting the game (it tends to crash, when I do so for bigger maps, might even be specific for my computer specs).

Thanks, Somebody613!!! Great suggestion!!

 

[Somebody613]

Thanks, Somebody613!!! Great suggestion!!

Tell us the result afterwards, ya know.