Single Player bugs and crashes - After the 29th of March

[Bohb Daishi]

The save game does not seem to correspond to the images. The save is of 6425BC. The images are from a few thousand years later! This **might** be an assets change thing, but regardless they don't match (the mammoth rider isn't north of Basra in the save). Did you attach the wrong file?

*facepalm* Yes, it's the wrong save from a different bug. I messed it up when I did a copy/paste.

I'll keep my eye out and try to get a new save, now that I know the bug CAN be reproduced.

 

[Talin2009]

Can you clarify the state in the save game - is it where a civ will begin researching Tengriism the next turn or where they already are (in which case it's too late to diagnose from)

They already are unfortunately (it is visible in civs tab in right bottom corner).
Current date and date when tengriism was founded also visible on screenshot.
I will pay attention and if i see this again ill try to get save before they start research.

 

[God-Emperor]

Am not sure which combination of promotions is causing this, but attacks into a stack do not use up any movement points. Game is set for attack from spearman into the stack to the SW. 0.5 mp left before attack, 0.5 mp left after the attack, so did not use ANY mp for the attack.

Looking into this. There was a tag that's supposed to enable this and it may be that it was not correctly merged into the system to filter properly.

I just had this happen last night. Oddly it was not with C2C, it was with R2R (which does not have the Combat Mod stuff from post v26 merged in). So this issue may have already existed back in C2C v26, or been added by a non-combat-mod bugfix/change.

I noticed that the combat log shows that the unit I attacked with took no damage, so it might be related to that.

My attacker and the defender were both Atl-Atl units. The defender was a barbarian in a city on a hill, no promotions, 0 city defense, and possibly no fortification bonus as I think it was the next to last defender and two had been moved in the turn before I attacked. The attacker had these promotions: Hunting I, Bamboo Armor, Poison Tips, and City Raider I. I think it may have gotten a +25% strength on the attack from the Great Commander with the attack stack, since I think it had the first two strength giving promotions at that time and I think it was the 4th attacker so the GC would have still had command points.

After the attack the attacker still had its full movement point. It could not attack again. (I'm guessing that if the unit had Blitz it could get an extra when this happens.)

I'm pretty sure I've seen this happen a few times before. I have always attributed it to some unnoticed property of a promotion as it has always seemed to be highly promoted units.

 

[Bohb Daishi]

The save game does not seem to correspond to the images. The save is of 6425BC. The images are from a few thousand years later! This **might** be an assets change thing, but regardless they don't match (the mammoth rider isn't north of Basra in the save). Did you attach the wrong file?

This might be the same bug - the Mammoth Rider goes from 6.6 HP down to 3.2 HP without being attacked. Again, this is when a Bandit Rider comes within visible range.

Edit- this is really weird. I clicked off the unit to go do some other stuff on that turn, then went back to it and now it has 7.8 hp. However, the other (full HP) mammoth rider on the same stack began "ghosting" and is killed instantly when fighting any surrounding units. The "ghost" mammoth doesn't show any combat odds either.

 

[Bohb Daishi]

Here's another save where the above-mentioned "ghost" glitch occurs. Note that the bandit rider kills the worker that's grouped with the upper mammoth (instead of fighting the mammoth like it should), and the following turn BOTH mammoths start "ghosting".

 

[Bohb Daishi]

ALL of this is the problem I was trying to blow the whistle on in our quick debug phase before the release. We were unable to isolate or identify the cause then and it's still elusive because the problem is repaired by a reload. Not all games get this effect so its also not so easy to play under the debugger constantly running in hopes of finding it. It's going to take some strong theorizing as to what the problem may stem from to find it I think. I'm almost thinking it may have something to do with units spontaneously losing some of their gamestate data. I've noticed rogues and invisibles seem to have a link to the issue but I have also been able to rule it out as THE trigger because it can happen under other circumstances. Ugh...

BEHOLD!!! Attached is a save that successfully reproduces the glitch! Notice how the Mammoth Rider 10 goes from 4.1 health to 0.0 HP on the next turn, but does not get attacked. I'm pretty sure it has to do with that bandit rider that shows up.

The save game does not seem to correspond to the images. The save is of 6425BC. The images are from a few thousand years later! This **might** be an assets change thing, but regardless they don't match (the mammoth rider isn't north of Basra in the save). Did you attach the wrong file?

Here's the save for the 0.0 HP glitch, which may or may not be related to the above "ghosting" glitch.

 

[Whisperr]

This isn't exactly a bug, but the peyote resource image shows on the square next to where the resource is.

 

[Thunderbrd]

I just had this happen last night. Oddly it was not with C2C, it was with R2R (which does not have the Combat Mod stuff from post v26 merged in). So this issue may have already existed back in C2C v26, or been added by a non-combat-mod bugfix/change.

I noticed that the combat log shows that the unit I attacked with took no damage, so it might be related to that.

My attacker and the defender were both Atl-Atl units. The defender was a barbarian in a city on a hill, no promotions, 0 city defense, and possibly no fortification bonus as I think it was the next to last defender and two had been moved in the turn before I attacked. The attacker had these promotions: Hunting I, Bamboo Armor, Poison Tips, and City Raider I. I think it may have gotten a +25% strength on the attack from the Great Commander with the attack stack, since I think it had the first two strength giving promotions at that time and I think it was the 4th attacker so the GC would have still had command points.

After the attack the attacker still had its full movement point. It could not attack again. (I'm guessing that if the unit had Blitz it could get an extra when this happens.)

I'm pretty sure I've seen this happen a few times before. I have always attributed it to some unnoticed property of a promotion as it has always seemed to be highly promoted units.

I'm not sure how you would've isolated it out so I think you're probably experiencing the effect of Onslaught (Granted to the units via a Commander's ability derived from an advanced promotion). What doesn't seem right is that they aren't continuing to attack/they can't make another one. I think I must've changed some things about how it works and the way you've split it out has somehow led to this effect.

This is probably different from the effect being experienced in C2C which I have a strong feeling has to do with a partial merge that will soon be a full merge of theLadiesOgre's unit tags. I THINK that will repair the problem being noted here as I'm suspecting the partial merge may not have enabled the code to take full consideration of his coding and something missed there may be allowing this bug.

 

[Dancing Hoskuld]

This isn't exactly a bug, but the peyote resource image shows on the square next to where the resource is.

I have noticed this also, but I don't have the tools or knowledge to fix it.

 

[AIAndy]

I'm not sure how you would've isolated it out so I think you're probably experiencing the effect of Onslaught (Granted to the units via a Commander's ability derived from an advanced promotion). What doesn't seem right is that they aren't continuing to attack/they can't make another one. I think I must've changed some things about how it works and the way you've split it out has somehow led to this effect.

God-Emperor started with the codebase before you added the combat mod so he never had to isolate it.

 

[Thunderbrd]

God-Emperor started with the codebase before you added the combat mod so he never had to isolate it.

Hmm... it must've actually worked before then. It was believed it had no effect. Maybe something happened to it between where he picked up the dll and where I put in the combat mod changes to disable it incidentally. Either way I don't think its working quite as intended in his version - close - but it should enable the unit to attack again thereafter.

 

[Bohb Daishi]

v29 - When viewing the upgrade cost while having multiple different units selected, the single-unit cost works, but the "all units" cost (shown by pressing Alt while hovering over the upgrade icon) is 0. The unit in particular is an archer. The bug does not happen when the archer is individually selected.

 

[DRJ]

just a quick question as I took a break for some months and only returned lately -
was the "flying improvements bug" (the farms and quarries and such seem like to be half over the tile they belong to) that I think is related to viewports solved in the meantime or is it still active? Cause if it was solved it is back

 

[Koshling]

This might be the same bug - the Mammoth Rider goes from 6.6 HP down to 3.2 HP without being attacked. Again, this is when a Bandit Rider comes within visible range.

Edit- this is really weird. I clicked off the unit to go do some other stuff on that turn, then went back to it and now it has 7.8 hp. However, the other (full HP) mammoth rider on the same stack began "ghosting" and is killed instantly when fighting any surrounding units. The "ghost" mammoth doesn't show any combat odds either.

Ok, pinned this down. Testing the fix now, should be pushed later today. Thanks for the saves.

 

[Koshling]

just a quick question as I took a break for some months and only returned lately -
was the "flying improvements bug" (the farms and quarries and such seem like to be half over the tile they belong to) that I think is related to viewports solved in the meantime or is it still active? Cause if it was solved it is back

It was not solved and my conclusion was that it probably cannot ever be with the current executable engine + viewports.

 

[DRJ]

ok what about the placement of the improvements - they have a certain value for height (different for terrains - a mine on a hill has other height as on grassland?) - if playing with vieworts on from beginning of the game (this maybe would be a gameoption "normalized viewport improvements") these values for improvements would be adjusted in a customized xml? /all values -XX height

What I dont know is when the improvemtns start to fly, as I only got the bug some way in the game and reloading sovled temporarily I think

 

[Koshling]

ok what about the placement of the improvements - they have a certain value for height (different for terrains - a mine on a hill has other height as on grassland?) - if playing with vieworts on from beginning of the game (this maybe would be a gameoption "normalized viewport improvements") these values for improvements would be adjusted in a customized xml? /all values -XX height

What I dont know is when the improvemtns start to fly, as I only got the bug some way in the game and reloading sovled temporarily I think

No - the issue is that the game engine seems to store its own values for map heights independently of the info in the DLL-mediated part of the game state. When you move the viewport around it changes the terrain at a given coordinate (as the game engine sees it) but there is no way to update the game engine's idea of what it thinks the height is at that location (which is whatever it was when it first saw something at those (game engine) coordinates I think). The exact mechanism is not known (we have no source code for the game engine), but empirically that's how it behaves, and it seems intractable.

 

[ls612]

It was not solved and my conclusion was that it probably cannot ever be with the current executable engine + viewports.

There was though an update in November that significantly reduced the issue, and I'm not convinced that it is unsolvable. But I understand very little of how the engine works so I certainly am not going to try and solve it.

 

[Debellatio]

1.I got into a problem with viewports after I`ve build enough religious buildings to locate the holy mountain. The "locate" button couldn`t work in that case. Don`t want to sound lazy, but is there a way to highlight it in the military advisor? (currently i play without a personalized map)

2.Sometimes after I accepted a pecuniary trade with a leader, it was his proposal, the relations between us went worse. Was he upset, because I didn`t haggle?

 

[Bohb Daishi]

Ok, pinned this down. Testing the fix now, should be pushed later today. Thanks for the saves.

No problem. Just curious, was it something to do with the bandits/criminals?