Post Feb 1st 2013 - bugs - Single Player

Status
Not open for further replies.
In italics seems to be where we're not picking up any data where we should. A break on the red line indicates there's a null value coming in for pCityNode, which since I don't understand the programming methods in use here fully is where I lose the trail.
In the first run of the for loop that will always be NULL, as the sorted list is still empty. Mind that that is an inefficient way to sort but it should be working.
 
In the first run of the for loop that will always be NULL, as the sorted list is still empty. Mind that that is an inefficient way to sort but it should be working.

That's interesting to note, but I believe I had ruled that out by attempting to loop further in, it still showed no valid value coming through there. I don't have further time this evening to fully confirm that though.

Either way, something in that function is not coming up with a valid city to trade with where it should be.
 
I got the oddest message when trying to open this:


However, pushing continue did take me to the game. I'm currently trying to get through a recalc and I figure it's going to take more time than I have to be awake tonight to get through it.

Still, this is by far the latest game save I've ever seen. Simply amazing! Late stage Galactic era... wow. I'm going to hazard a guess and say that the overloaded integers are having a heyday with your game at this point and that could have a lot to do with the bug as reported. Sorry I can't do much more tonight on this one.

EDIT: Actually, the recalc completed and I was able to diagnose the hang. But the fix is going to be tricky and will take more thought than I can quite work out at this time. Apparently you're national culture level (or rather the next expected necessary amount for your leader to level again) has exceeded even the limit of unsigned long long!!! So basically what this means is that we're going to have to use some more breaking down in the math so that we can enable things to progress further in your game and others that reach this point. I was a bit worried this might be possible. I doubt it can be fully addressed before the release but I'll see what I can do tomorrow on the matter. I have given some thought as to how this can be addressed if it becomes a problem but not enough to implement yet ;)

Thanks for checking this, Thunderbrd!

Some comments:
  • I have never seen this odd message, no idea where it comes from.
  • I am not surprised that this game might reach the culture limits as I have all religions and all religious buildings in my cities and I dominate the world. But then this is only a standard size map, so even more culture could be possible in a game. But generally speaking I think there are way too many buildings producing culture and also producing too much culture each.
  • But maybe the culture issue is not the only reason for this. I have left the frozen game running for about 3 hours now and the memory usage has been steadily going up from 1.2 GB to 1.8 GB now. Maybe there is a kind of loop running there.

Added: I do have trade routes in my game (you asked whether there are trade routes at all).
 
Can someone look at my savegame? The AI turn never ends. I don't want to start new game after investing a week on playing it.
 
I am looking at the two reports of end-turn hangs FYI...

@Zeinul - in regard to your case, is that on he latest SVN or on V28?
 
latest svn

Ok thanks.

Having looked into these saves here are the conclusions:

Snorfu1's save:

First of all - props on getting so far!! I was amazed when I loaded the game and saw your current research tech was 'ascension'!

Your hang is due to leader-experience-related overflows. Basically it overflows in such a way that the game decides that your next required culture for a new leader level (dynamic traits) is negative. It then gets stuck in the following loop forever:
Code:
				do
				{
					doPromoteLeader();
				}
				while (canLeaderPromote());
I should be able to fix this today easily enough.

Zeinul's case:

This turns out to be an extremely subtle pathing bug, which I'm working on...
 
Thanks for checking this, Thunderbrd!

Some comments:
  • I have never seen this odd message, no idea where it comes from.
  • I am not surprised that this game might reach the culture limits as I have all religions and all religious buildings in my cities and I dominate the world. But then this is only a standard size map, so even more culture could be possible in a game. But generally speaking I think there are way too many buildings producing culture and also producing too much culture each.
  • But maybe the culture issue is not the only reason for this. I have left the frozen game running for about 3 hours now and the memory usage has been steadily going up from 1.2 GB to 1.8 GB now. Maybe there is a kind of loop running there.

Added: I do have trade routes in my game (you asked whether there are trade routes at all).

Oh... I should've been more descriptive. There is an infinite loop taking place. Basically what's happening is your leader is leveling over and over because after the overload, the amount of national culture needed to reach the next level drops to 0.

I figure I can simply establish a maximum level and that will avoid the problem. This is just one of so many bugs to fix but that oughta do it. I'll be working on that in a little bit here.

Ah... just now reading koshling's comment. He found the same diagnosis and it appears he might be trying to repair it himself.

EDIT: also... my wife's game is now showing a trade route... sigh. But her newest city doesn't have one yet despite being connected to the trade network. In the game I was analyzing, there were plenty of trade routes in volume but no cities were being found as valid to trade with for whatever reason. (I'll go in now and look at that again with AIAndy's comment above in mind.) But with her's I wonder... is it that we have no initial trade routes by default and entirely rely on buildings/civics/traits to enable some at first?


@Zeinul: Koshling probably has to make some coding changes and update the SVN for ya is all. Just hold out for now as there's probably not much you can do yourself in this case.
 
In the attached game I have a problem with a hunter being invisible. My rogue could see him when 2 spaces away but I could not attack, and when I moved next to him the hunter became invisible to me. When you hover over the square it says that he is there.

In the game it is a couple of spaces to the right of the Mongolian capital and the space north of my rogue.

Thunderbrd said he has no idea about this and to post it in here :)
 

Attachments

Two bugs related to captives. I updated to v. 5106 copied the contents into game mod folder but in game it says (5098). Not sure why the numbers don't match exactly. But just for DH to check the following with captives.

One is when I captured civilian it said: "You have captured %d civilian unit(s) (Captive - Civilian)". See first screenshot.

The other is when the Captive - Civilian is going to join a city it has a text error on the hover tip that says "TXT_KEY_OUTCOME_JOIN_CITY_POPULATION". See second screenshot.

EDIT: Added one more screenshot.
 

Attachments

  • Bug-CaptureCaptiveText.JPG
    Bug-CaptureCaptiveText.JPG
    228.9 KB · Views: 43
  • Bug-CaptiveJoinCityText.JPG
    Bug-CaptiveJoinCityText.JPG
    211 KB · Views: 42
  • Bug-CaptiveJoinCityEventText.JPG
    Bug-CaptiveJoinCityEventText.JPG
    211.8 KB · Views: 41
Both of those should have been fixed by yesterday's update. I'll take another look.

I'll try to update again and see if it works. I'll assume it's fixed unless I update properly and see it again.
 
In the attached game I have a problem with a hunter being invisible. My rogue could see him when 2 spaces away but I could not attack, and when I moved next to him the hunter became invisible to me. When you hover over the square it says that he is there.

In the game it is a couple of spaces to the right of the Mongolian capital and the space north of my rogue.

Thunderbrd said he has no idea about this and to post it in here :)

Actually, she was working on a slightly outdated svn (from yesterday) and apparently we aren't able to replicate the problem now so perhaps its been fixed at this point by one of the previous resolutions.
 
I also have missing units (but they ARE there)???(USA savedgame)

Also for Koshling, (same pic) the enemy units (far bottom right) declared war, but do NOT attack when they have a far superior force??6770 saved game (both same game really, one is just sooner than the other).

SVN: 5111. No VP.
 
OK besides the above post, now i am also getting to the point i cant attack the AI i am at WAR with:mad:

same SVN as listed above


Huron fixed thx.

EDIT: Minidump from different game, but still some units are invisible and cant fight them?? Even in older games, so playing any games is USELESS:mad:
 
OK besides the above post, now i am also getting to the point i cant attack the AI i am at WAR with:mad:

same SVN as listed above

EDIT: Minidump from different game, but still some units are invisible and cant fight them?? Even in older games, so playing any games is USELESS:mad:

This is a pure XML bug:

Code:
		<UnitInfo>
			<Class>UNITCLASS_HURON_MANLET</Class>
			<Type>UNIT_HURON_MANLET</Type>
                           ...
			<iCombat>4</iCombat>
			<iCombatLimit>0</iCombatLimit>
			<iAirCombat>0</iAirCombat>
			<iAirCombatLimit>0</iAirCombatLimit>
                           ...
		</UnitInfo>

iCombatLimit of 0 means it can only damage enemy units by up to 0% of their total health!
 
Status
Not open for further replies.
Back
Top Bottom