Specific Bug Reports

[Iustus]

not sure if this one been reported yet:
Assert Failed

File: CvPlot.cpp
Line: XXXX - not same as your lines
Expression: getCulture(eIndex) >= 0
Message:

----------------------------------------------------------

You are trying to set the culture value on a plot to a negative value. This feels like an XML error where you have some building or something which has a negative culture value in error.

This one would be easy to track down with the debugger, but otherwise, you may have to add a bunch of logging to find it. If you can get it with a regular BetterAI build, I would love to see the save.

-Iustus

 

[Topher]

Uh, I posted the wrong save (75 BC instead of 75 AD)

This is the right save, I downloaded it and checked to be sure:

View attachment Topher AD-0075.Civ4SavedGame

Sorry about the mix up.

 

[Iustus]

Build 1/30

File: .\CvCityAI.cpp
Line: 5409
Expression: false
Message: infinite loop(!!!!!)

Just hit enter. I suspect it has to do with the Greek AI choosing what to build in Machu Picchu (sp), which it just took from me (in a move it never would have made in vanilla - nice work!)

http://forums.civfanatics.com/uploa...0700_CiyAI.cpp_Assert_failure.CivWarlordsSave

Edit: And the same thing after ending turn in this game, although without the same city conditions (that the player can see, anyway) - could be something different altogether:
http://forums.civfanatics.com/uploads/68411/AutoSave_AD-0425.CivWarlordsSave

As reported before, this bug should be harmless, just click ignore always. It will make your turns go a tiny bit slower.

This is caused by the city governor not knowing which plot to work (a few being almost equally good). It does this 'juggle' function, switching one, and then checking again, well it gets into a loop, cycling between 3 plots. After about 20 tries, it gives up, reporting this error.

It will be fixed in the next build, but for now, ignore it. Since it is caused by the city governor (which the AI uses), many events can trigger it.

-Iustus

 

[gnome]

Assert Failed

File: .\CvArtFileMgr.cpp
Line: 174
Expression: false
Message: get##name##ArtInfo: OVERLAY_ACTION_FORTIFY was not found

----------------------------------------------------------

http://forums.civfanatics.com/uploads/23357/Montezuma_AD-1065.CivWarlordsSave

Action: end of turn, appears when barracks built and prompting for next build. Hit "Examine City"

If ignored, city view does not appear correctly.

Mods: Better AI, Balancer for Better AI, Look and Feel, Bad Ronald's flags.

Note--only started today, and happening frequently. I may have messed up something.

This error is getting more and more perplexing. I am getting this error every time I try to examine a city now. I even uninstalled and reinstalled all of civ and warlords, and reinstalled the same mods I was using yesterday, but I'm still getting these overlay errors. I'm wondering what I've done to my setup to cause this.

 

[Iustus]

Assert Failed

File: .\CvPlayer.cpp
Line: 7365
Expression: getNumOutsideUnits() >= 0
Message:

----------------------------------------------------------

Build 07-01-30. Save attached. Call up Washington, ask for Capitulation and Facism. Freaks at that point.

Do you happen to have a save a few turns earlier? Washington has a worker over in Ragnar's territory, but his outside unit count is zero. I would like to know how this occured.

This is starting to look to me like a firaxis bug. (The only effect of which is that the unit costs you pay for units outside your territory might be sometimes off a bit).

-Iustus

 

[Quagga]

^^^ Sorry, no. That happened three days ago. I save only 10 autosaves. I know I manage more than 3.3 turns per day!

 

[Iustus]

I did some more testing with Mulitplayer and the OOS errors using Better AI 1/30. I attached the autosave from the turn before along with all of the logs. I finally got a MPLog file after turning on Sync logging in the INI file. An excerpt from it is below. Maybe it will be of use to someone.

Scout(Bolt Jenkins)[980]: AI_explorePlotValue (68, 16) using cache-> 0
Scout(Bolt Jenkins)[980]: AI_explorePlotValue (69, 15) using cache-> 0
Scout(Bolt Jenkins)[980]: AI_explorePlotValue (69, 14) using cache-> 0
Scout(Bolt Jenkins)[980]: AI_exploreRange MISSION TO-> (68, 16)

***** OUT OF SYNC! MY SYNC SEED = -653456736 : MY OPTIONS SEED = 26760276 *****

Rand = -1441652362 on 982 (Religion Spread)
Rand = -953261193 on 982 (Religion Spread)
Rand = 370723812 on 982 (Religion Spread)
Rand = -1928955827 on 982 (Religion Spread)

If I had to venture a guess, the bug is with some of the fogbusting code changes that were made about 3 weeks ago.

Thank you for your post.

I made a post with instructions on what needs to be done to look into these sync issues.

We need to look at the MPLogs from both computers. I do not have sync log enabled on my machine, so I do not believe that is needed. The problem is that if you quit the game, the MPLog file may be deleted. So you need to have everyone alt tab out and make a copy of the file right away. Then send them all to one person to do a comparison.

There were definitely some fogbusting errors, but we fixed all of those we could find. There is still quite a bit of extra logging from the explore code, but this does not mean that is the cause for the sync errors. The explore code does use random more than most things, so is more likely to 'discover' a problem that was caused by something else.

That said, I suspect the problem is something else, based on my own testing. If we can find a difference in the MPLogs, we can track it down. (If the logs are identical, then we may need to add some more logging of other things, until we can track down what it is that is different). We do have a way to do a build that generates an enormous logfile, with almost everything tracked, if someone wishes to test with it. (The file will get to be hundreds of megs if you play for a while).

-Iustus

 

[gnome]

Update on Overlay error: score another one for the methodical approach.

I tested a game after loading each mod component... and found that the Look and Feel mod is incompatible with the Title Screen Mod that gives you back the Earth and "Baba Yetu" song--at least if you combine them into a single mod.

Brings up the question of "why?" but makes it a much more low priority issue... whew.

 

[Iustus]

My bug report (version from 1/31 with handicaps)

FAssert(isHuman() || isBarbarian() || ((getGold() + iGoldChange) >= 0));

As I said previously, this is caused when an AI goes into strike (which should almost never happen). In this case, it turned out to be a real problem that I could fix.

Wang Kon was fine financially. However, he had just spent some money I suspect (perhaps upgrading to longbows), and was down to 4 gold. Well, he decided to change civics, which is fine, but that put him into anarchy, changing his +12 gold per turn into -5 gold per turn. (Trade deals are still active during anarchy, and he had a 5 gold per turn going out in trade deals). Since he only had 4 gold, the -5/turn during the 1 turn of anarchy put him into strike.

This is not good. The fix is that AIs will wait a turn before changing civics or religion, if it will cause anarchy, if they do not have enough gold in the bank to cover their trade deals.

I would really like to see Firaxis change the behavior under anarchy so that you do not strike. I could do this, but it would be a gameplay change, so I am not going to do so.

-Iustus

 

[Ralgar]

This one turns out to just be a bit too liberal on asserting here. Since this assert was added by Firaxis, I am reluctant to remove it. You should only get it when someone is bribed into a war they did not wish to start. I will add a fix in the next build, but this one should be totally harmless.

Thanks for the feedback on every bug report. I found it rather strange how very long the query in my assert was.

 

[Iustus]

CivIV Warlords with 2.08

Build 1/30

Assert Failed

File: .\CvUnit.cpp
Line: 8566
Expression: getCombatUnit() != NULL
Message: getCombatUnit() is not expected to be equal with NULL

----------------------------------------------------------

I was playing on Chieftain and I tried to unload troops from a Galleon onto a land square with a barbarian worker. I received the msg that I had captured the worker, following by the msg my Macemen killed the worker, and then the game locked up with the troops showing but still on the ship.

Do you have a save game from this? I have seen this before, but I was unable to duplicate it.

-Iustus

 

[KHSOLO]

I have a bug to report and plainy replicable, have saved game also.

The only thing i dont have is a place to upload the saved game cause the uploading screen says files smaller than 500 Kb when my save is almost 900 Kb.

Pls someone point me to a place where i can upload this file, thanxs.

 

[Iustus]

I have a bug to report and plainy replicable, have saved game also.

The only thing i dont have is a place to upload the saved game cause the uploading screen says files smaller than 500 Kb when my save is almost 900 Kb.

Pls someone point me to a place where i can upload this file, thanxs.

You want to click the red Easy Upload System button when making a post:

That should let you upload a file that is larger (you will want to make a zip of it).

-Iustus

 

[Iustus]

^^^ Sorry, no. That happened three days ago. I save only 10 autosaves. I know I manage more than 3.3 turns per day!

Next time, when you report a bug, make a copy on your own hard drive of the whole autosave folder (just copy and paste it). Then, if we need an earlier save, you have that folder. You can give the folder a date if you wish, so you can then delete it in a couple weeks if we dont need it.

-Iustus

 

[Iustus]

Uh, I posted the wrong save (75 BC instead of 75 AD)

This is the right save, I downloaded it and checked to be sure:

View attachment 147671

Sorry about the mix up.

Thank you for following up. This was a good bug! Both problem with how stacks were attacking and also a crash bug.

(both fixed in next build)

-Iustus

 

[Iustus]

Shooting for a build tomorrow, which addresses all known issues. If you are sitting on a bug report, try to post it asap so we can try to get a fix in by tomorrow.

-Iustus

 

[Mexico]

Shooting for a build tomorrow, which addresses all known issues. If you are sitting on a bug report, try to post it asap so we can try to get a fix in by tomorrow.

-Iustus

as reported earlier in http://forums.civfanatics.com/showpost.php?p=5060449&postcount=142 and checked in rev.350, this bug is still present
when AI want move unti to city for upgrade, but path is blocked, game crashed

 

[shadyforce]

I found a bug (assert) last night but I never got around to reporting it. Basically, I have 1/30 installed in CustomAssets. Then I started a game of the Chinese Unification Scenario. When the game was loading, it froze as an assert popped up in the background. It also popped up the next time I loaded the scenario to continue the save. Oh and if I hit ignore once, the assert appears again instantly. I just hit ignore always after that.

I don't have the text or the save as I'm at work. but I just wanted to mention it now (before the new build) as the assert seems easily recreatable.

 

[jray]

Build 1/30

File: .\CvGlobals.cpp
Line: 1081
Expression: eLeaderHeadNum > -1
Message:

This one only happens when using my jrayUGH mod (v208b). Hit End Turn to recreate. I don't know if it's within your scope to troubleshoot mod-specific asserts, but if you happen to find a problem with my mod, I'd appreciate it if you let me know. Thanks .

 

[Wodan]

Assert Failed

File: .\CvPlot.cpp
Line: 3906
Expression: eRiverDir == CARDINALDIRECTION_WEST || eRiverDir == CARDINALDIRECTION_EAST || eRiverDir == NO_CARDINALDIRECTION
Message: invalid parameter

----------------------------------------------------------

got this as soon as I started the Earth scenario

Wodan