Fall Further 050 Bug Report Thread

My prefered "safe" mehod, would be using the block version. It tests for mission validity and pushes the MISSION_SKIP, only if the mission is invalid. In fact, this is the version I decided to use for myself, until the next patch comes out :).
The only thing is that it is slower.

That last fix, is in a block with a gametime check?
My save is in a turn near 400!(>337 IIRC) in a standard time game...

My guess, since the block version fixes the problem, is that, at some point, the AI may be given invalid plot assignment as a mission target. And so it fails since the if (canMoveOrAttackInto(pLoopPlot)) was never called, to verify that the plot is valid for a move or attack. My guess is the latest change does not fix it, and there should be a validity check for the target plot, as was in the commented-out code.

vdex34 said:
VC++ 2005 express edition offline install available here

The offline install is an .iso image though (CD). The 2008 edition is available on the same website, but for some reason M$ made it so 2008 can only be burned to a DVD. (Its a CD size image both versions.)

Thank you! I'll give it a shot at first opportunity!

Here is the save game.
 

Attachments

While messing around with the Hyborem, I tripped across another div by zero crash in CvCityAI.cpp, line 7909. Here's my patch...
Code:
iTempValue /= 3 * (iConsumtionPerPop == 0 ? 1 : iConsumtionPerPop);
 
While messing around with the Hyborem, I tripped across another div by zero crash in CvCityAI.cpp, line 7909. Here's my patch...
Code:
iTempValue /= 3 * (iConsumtionPerPop == 0 ? 1 : iConsumtionPerPop);

Just happened to notice - is it really 'Consumtion' without the 'p' here? Of course, if its a freely defined variable, the spelling wouldn't matter, but it sure looks funny...:)
 
Just happened to notice - is it really 'Consumtion' without the 'p' here? Of course, if its a freely defined variable, the spelling wouldn't matter, but it sure looks funny...:)

It may look funny, but this is a copy/paste. So it should be definately, consumtion.
When you misspell defining a variable, you, usually do not go back correcting the spelling. It is not a grammar test anyway. You would be amazed on how many such 'misspelled' variables you find out when you are looking over code. Even if this is your own code ;).
 
This is a dll, containing the "block" version of the hang_issue(the safe one) + the latest "division by zero" fix by MootPoint. Of course, it includes all previous fixes, also.
 

Attachments

I have a problem with the modmod/patch installer as one needs to be administrator of the computer to use it (while it should not be required). FFH installer does not have this restriction (which is good!). Would it be possible to change FF installer so that this is not required anymore ?
 
Yeah, I saw that too.

I actually liked it a lot until finding out I couldn't assign - :food: to them and + :hammers: to them for the dwarf races.
 
So, I have a new, difficult to track problem...

A friend and I are playing a Sheiam game, he swaps away from his sheaim to play infernals. Predicatable OOS, so we save, restart, no problems for like 80 turns. But now we OOS at the start of every turn. We both have randomlogger active this time, and now it looks like the game is running the AI barbarian checks in different orders for both of us - it's clear in WinMerge we're running the same checks, but in a totally different order.
 

Attachments

I'm thinking it has to be something with Hybie's code because I've never experienced this problem before.
 
Anyone has an idea, why the civilopedia entry of the minster doesn't work when you open the civilopedia via the main menu? It works when you use the civilopedia after starting a game. It has something to do with the buildings requirement of the minster BUILDINGCLASS_ABBEY. If you change that to PAGAN_TEMPLE for example the minster entry in the civilopedia always works. I haven't found a way so far to fix it without changing the PrereqBuildingClass.
 
I dont really have a bug per say, so much as a coding oversight, involving Korinna the Protector. When you save a world in worldbuilder, thus creating a scenario, after Korinna has spawned in the capital, she will spawn again when you play the scenario. Thus, you will have two Korinna's.

I encountered this after circumventing a bug of sorts, where you would end your turn, and then the game would stop with the "waiting for civilizations" message flashing, as if in a multiplayer game, only it is single player. It happens occasionally, is not bypassed by loading a previous save because it will still happen on the turn it happens to happen on, but doesn't occur is you create a scenario out of the world using worldbuilder on the preceding turn.

So yeah, there you go.
 
dont have OOS logs, not sure if i even imported the logger right, but I did run through the turn in Hotseat multiplayer mode, and the game crashed consistently on Kuriotate turn. When i set the Kuriotate scout to auto-explore the game crashes to desktop. Im pretty sure that it isn't a problem with the auto-explore feature, so it is more likely that there is something, some feature on the map, which is causing it to crash. only thing i see that is special on the map when i give the command is something called "Sirona's Beacon" or something like that, but since that feature is not hidden by the black fog, i dont think it is the cause either. By the way, im playing on huge map with 27 participants (all ai's, and all civilizations accounted for besides hell and angels)
 
Here are the OOS logs - bearing in mind it's Iceciro and Stan that are the only two human players - "Andrew" is an AI who has just kept my friend's name when we stopped the game to swap him to Hybie.

Neither of the actual player's assets are desynched because the problem simply lies in us pulling the random numbers out of order, throwing the positions of barbs and AI units out of whack :crazyeye:
 

Attachments

I found it. Whenever Kuriotate's scout got within sight of Dural's borders, when the diplomacy window normally pop's up, the game crashes.

EDIT: Bah nevermind, this is a separate issue. When in hotseat apparently you crash whenever you encounter another person. Back to the problem i had before, where the game stalls when you end your turn, it appears to happen during the barbarian turn; you end your turn and the "waiting for players" message flashes.
 
dont have OOS logs, not sure if i even imported the logger right, but I did run through the turn in Hotseat multiplayer mode, and the game crashed consistently on Kuriotate turn. When i set the Kuriotate scout to auto-explore the game crashes to desktop. Im pretty sure that it isn't a problem with the auto-explore feature, so it is more likely that there is something, some feature on the map, which is causing it to crash. only thing i see that is special on the map when i give the command is something called "Sirona's Beacon" or something like that, but since that feature is not hidden by the black fog, i dont think it is the cause either. By the way, im playing on huge map with 27 participants (all ai's, and all civilizations accounted for besides hell and angels)

27 AIs on a huge map is probably destroying your computer's memory. Are you using 64-bit vista, or... pretty much anything that isn't 64-bit vista... and how much RAM do you have.

EDIT: Just saw the bit about diplomacy - lategame (Turn 300+) on a large map my crashes usually happen when I try to load the diplomacy for some AI that's harassing me. I haven't had them since using the newer DLL files yet, though.
 
Back
Top Bottom