Bug Reports

[phungus420]

Battle effects are a poorly implemented feature. I don't know who leaves them on. I'd be just as happy if you removed them.

Also pretty major bug found in RevDCM:
If you play a scenario, all civs will have contact with other civs. There is no way to set contacts in the WBS for a civ that exists on the map...

 

[frekk]

Battle effects are a poorly implemented feature. I don't know who leaves them on. I'd be just as happy if you removed them.

Seconded. They're a fun gimmick the first couple of times, but once they've been seen, they tend to get old and I turn them off, and I figure everyone else does too. The smoke is quite annoying.

 

[BobTheTerrible]

Battle effects are a poorly implemented feature. I don't know who leaves them on. I'd be just as happy if you removed them.

Also pretty major bug found in RevDCM:
If you play a scenario, all civs will have contact with other civs. There is no way to set contacts in the WBS for a civ that exists on the map...

I believe this can be fixed by disallowing minor civs when you're setting up the scenario, and then re-allowing them once ingame or something like that.

 

[phungus420]

jdog were you able to reproduce and isolate the hang from the 0.8.2 build I posted?

 

[jdog5000]

No, sorry. The surprise 3.19 patch got in the way ...

The good news is that RevolutionDCM is now working in 3.19, we've got a few more things to tweak/check but it shouldn't be too long till it's out.

There are a couple of people with reports of hangs in modded versions of BBAI 0.76, hopefully the issues are all related. They largely seem to be happening in the later eras just like what you described.

 

[phungus420]

Cool beans on getting a 3.19 working build. Here's too it's speedy and uneventful release

You still might want to take a look at that reported hang though, as it's clearly there, and may help you isolate the issue, and as you say it may be related to other reported hangs.

 

[phungus420]

Assert Failed

File:  CvPlayer.cpp
Line:  14230
Expression:  isHuman() || isBarbarian() || ((getGold() + iGoldChange) >= 0) || isAnarchy()
Message:

Probably harmless and caused by AIautoplay. But just reporting it. Clicked ignore and game continued with no problems.

 

[glider1]

Thanks. Out of interest, you must be adding extra code to cvplayer, because that line appears at 14198 in the stock code and yours is at 14230. As another interest, are you using the "doGold()" function anywhere in your python? This may cause such an assert. I checked the Revolutions code and it does not use this function.
Cheers.

 

[phungus420]

The code I've added to CvPlayer is connected to two XML tags for the Legends of Revolution mod. The first is a Trait ability which allows culture from non state religions, and the second is a BuildingsInfos tag which puts the Leonardo's Workshop functionality in the SDK, instead of using the upgrade price override ability of python (just seemed much cleaner and in the longrun better to do it through the SDK). Both of these are stable and I can't see how they could cause any problems, the code is really pretty simple. If you want I can post it, or upload the couple source files that are altered from standard RevDCM, but there are no references to doGold() in any of my altered Source. And as far as python is concerned, Since jdog incorporated the taxes and city distance change, the only altered python I have is for the civilopedia to display the new trait, and legends concept I've added.

Anyway based on this message:
Expression: isHuman() || isBarbarian() || ((getGold() + iGoldChange) >= 0) || isAnarchy()
I'm fairly certain this has something to do either with being in AiAutoplay or the new Barbarian not being subject to financial trouble code recently added to BBAI. Perhaps there's an unintended consequence of the BBAI barbarian code that allows the barbs to go into negative gold It happened for about 3 turns then went away. Unfortunately I didn't save it because I was in AIautoplay when it happened and you can't escape from that. That'd be a nice ability though, to be able to escape from AI autoplay

 

[Ninja2]

If it's the autoplay from Better AI, then you can CTRL+SHIFT+X, and it stops.

 

[phungus420]

Another strange Assert, never seen it before:

Assert Failed

File: c:\program files\firaxis games\sid meier's civilization 4\beyond the sword\mods\bst\src\debugworking\cvgamecoredll\CvGameCoreUtils.h
Line: 51
Expression: iHigh >= iLow
Message: High should be higher than low

Happened when clicking on a particular custom model in the game. Not sure what to make of this one. Model seems fine, clicked on ignore, but it kept popping up. So did always ignore. Model caused no problems and everything seems to work fine, but figure I'd report the assert.

 

[jdog5000]

Assert Failed

File:  CvPlayer.cpp
Line:  14230
Expression:  isHuman() || isBarbarian() || ((getGold() + iGoldChange) >= 0) || isAnarchy()
Message:

Probably harmless and caused by AIautoplay. But just reporting it. Clicked ignore and game continued with no problems.

This means that an AI player has basically completely failed economically ... it's completely out of money. It happens very rarely, maybe 1 in 200 AI players or less. Won't cause a crash or anything, but it does mean that AI player is totally screwed.

 

[phungus420]

I got this python exception while running AI autoplay in the background. I'm not sure, but I think it popped up when it stopped and Boudica came to make a deal. At least AI autoplay had only been on for 3 turns, and when I checked on it's status I saw a diplo screen where boudica was proposing a tech trade with the exception box up.

Not at all sure when this one happened.

 

[phungus420]

Another note, it seems pretty much all the revolution text is broken. When a weak civ vassalized and then was absorbed it said % has given up their independence and joined %. And also when you hover over a vassal the text saying to have them prepair for war sais TXT_REV_ALT_SHIFT or something like that. So really need to look back at how the text is being handled in Rev, cause something has messed up.

 

[glider1]

As for the first error, I think an import is missing from RevEvents? Try this:
import Popup as PyPopup as one of the imports for RevEvents.
Hmmm.

As for the second error, that's an old BUG error that's been around a lot and occurs when you do something in the interface in conjunction with autoplay or switching player, can't remember. Doesn't normally occur in straight games and seems only to point to an interface issue, not a gameplay issue.

Cheers.

 

[Duneflower]

Glider, check back a ways - I believe that's the exact same problem I hotfixed a while ago, with exactly the fix jdog suggested. Doublecheck me tho.

Now I have a new one; see attached image. I believe this was a city attempting to rebel due to my culture overwhelming their motherland's in the city-square; they'd already had a couple protests.

 

[phungus420]

LoR Bug Report

Remember this bug from the RevDCM 2.00 version. Thought it had been fixed, but aparently it is still broken, and an AI can still be completely crippled from the Slave Revolt event.

 

[phungus420]

Assert Failed

File: CvPlot.cpp
Line: 6960
Expression: getBlockadedCount(eTeam) >= 0
Message:

No changes in CvPlot, so the line should match with RevDCM. Assert had to be always ignored.

 

[phungus420]

In regard to the above assert, I have a strong reason to suspect this is causing severe increased lag. Any feedback on it would be nice. I know this isn't introduced by LoR, I think it's a flaw in the blockade code fix.

 

[jdog5000]

No changes in CvPlot, so the line should match with RevDCM. Assert had to be always ignored.

This assert will fire sometimes with BBAI and the fixes to the range of blockades reaching across isthmuses and things like that. The < 0 part is fixed immediately after the assert, I'll comment out the assert in the future (it was helpful to let me know when this was happening while testing).

There is not any noticeable increase in computation time associated with the blockade change.