Single Player bugs and crashes v40 plus (SVN) - After Oct 2019

Ah, that would definitely be it. I was wondering why horsemen would occasionally defend a stack while I still had better defensive units in it, but if they both had good chances to defend, makes sense that a dedicated pursuit unit would be chosen.

 

So that is normal behavior, but I'm a bit curious about:

@Dancing Hoskuld or @raxo2222 Could one of you resolve this one?

Not really a bug in coding terms, but certainly questionable design. The forts that 'count as a city' inherit the city placement range limit, whereas those that don't do not. That a fort that does not count as a city upgrades into one that does seems to be a bad design and this is what results - shouldn't towers be upgrading into towers guys?

Ah... makes sense to perhaps make it take on a pt in Tourism after perhaps. But yeah, obsolete!

This would be a python issue I think since it's supposed to be working with python to achieve this, though since the action takes place in the code, I don't know how it's supposed to be working in python. Certainly something that may require a new tag and coded effect for it to work properly.

I'll try to look into this one at some point soon.

Ah yeah, old issue I'd forgotten about. I tried to resolve this many many versions ago but my AI coding had a bug I couldn't solve for at the time. I should take a look into that again soon.

hmm... definitely something to resolve. I thought that had been corrected a while back.

Yeah very aware of this complaint. Haven't been able to find the problem yet. Sorry.

Some aspects of the game are not necessarily planned in tandem. I'm not sure if there is a tech that allows that or not but if there ever becomes one or if the rule ever is to change as to not being able to found there, the tag for 'destroyed at pop' is set in case.

 

Towers used to upgrade but it caused problems with other things being introduced at the same time so we removed all but the first. I think they may still be in the XML and art just deactivated, unless someone tidied them out.

It used to be that if two palisades were too close only one would upgrade to a fort. The idea of two forts to make a canal was supposed to be replaced by an actual canal improvement instead.

I thought the limit on how far between cities was removed from forts. There was a bug early on that caused some siege warfare issues. Originally you could claim a plot next to an enemy city and build a fort on it and attack from the fort with all the benefits of being in your own territory. Just like in real life.

By the way has the bug for claiming territory been fixed? You used to be able to claim a plot in the wilderness and get your units to heal up faster. When I reported the bug such plots were reverting to wilderness the next turn leaving you vulnerable to fish attacks (for ships).

 

No worries! I figured some of these would be very much known, but there's quite the number of threads to look through here, heh.

Towers do show up in the 'pedia, which claims that stone watch towers can be made after masonry, but ingame workers can't construct the improvement.

Can confirm the limits have not been removed, tried to make a wall of staffed forts to keep enemy strike teams out of my territory, but found I couldn't finish it after mathematics and was sad.

With fixed borders you can still claim tiles outside of your territory that nobody has claimed, but you are correct it will revert after a single turn.

That actually reminds me of another bug I saw, there there occasionally are tiles in between my cities that flicker back and forth between mine and unowned every turn. I imagine this is some bug with realistic cultural expansion interacting with logic about when tiles get autofilled.

 

Yeah last I saw they were just commented out and I missed out on understanding why this happened... doesn't seem right to me so I'm asking what those problems actually were if anyone remembers.

Nothing checks for validity for the upgrade to keep one from upgrading. You are right, though, that canals are still something we need to get in the game. The tag for a canal is in place but the art and how to angle it properly on the map would need some more advanced python work. I think that's where that got held up.

It can't be if the fort counts as a city because the counts as a city indication includes the city distance placement limits.

I do believe that some adjustments were made that should make it so that if you are claiming a plot it shouldn't be subject to culture removal but I was fixing a different problem to anything dealing with fixed border rules and its possible that it was the problem is wrapped up in some kind of logic collision between the need for plots that aren't gaining a cultural influence to eventually be lost to the nation that no longer is exerting that influence and the need for units to maintain their hold on a claimed plot through fixed border mechanics. To fix one breaks the other sort of thing. Would have to take a day to look at the code with the unit plot claiming in mind and may require a lot more programming to get it to work for that purpose.

EDIT: ok so apparently there is a bug in unit tile claiming that will need extensive deeper engineering in the code to resolve.

As an aside, in fixed borders, I didn't think naval units were supposed to be able to claim a plot?

 

The later towers were supposed to provide larger visibility and aid with interception air missions. This required code. They (stone and above) should provide a boost to detection of hidden units but not as good as dogs. Radar towers should help fighters and anti-missile units intercept targets in range of the tower not just the cities they are stationed in. There was talk of having military airports instead of forts and command bunkers. In fact iirc the talk was about having forts upgrade to command centres, training camps (one naval on coast, and one land), military airbases, missile silos.

I don't think I was using fixed borders, civic or otherwise. Some civics do allow it. Basically it is the just allows you to repair your ship and continue on with your journey. Like in real life before the Industrial Era.

 

It would be possible to improve sight range with an improvement and I can see how one could also program a plot improvement to the invisibility sight range on camo invisibility. One interesting way this could both be achieved now would be to add a new kind of buildup where the promotions in the buildup require an improvement on the plot - the tags for that are there but have not yet been used. You could do a Tower Fortification buildup that gives the +5% defense modifier (perhaps even +6%) at each stage and on the first stage of it also includes +1 camo visibility range and +1 overall visibility range. It could require any one of the tower types (that improvement prereq tag should be an OR prereq). The same method could be used for everything you're talking about in a way, since promos can increase interception ability as well.

I plan to give some units improved fortification options, particularly at the stage of trench warfare where the trench promos will unlock improved trench fortification buildups. Also going to be giving some ability to reduce the enemy's defensive combat modifier (different than an outright bonus to attack in that it can only counter the defensive amount of combat modifier the defender currently has built up.)

I'm working on a land unit review where I will be looking to put much of this into effect, but when I eventually do an air unit review, some more details should be built into interception rules... I MAY end up doing some of that for this review as well - it really should work a touch differently to how it currently does based on some discussions we had on the forum a long time ago - interception vs evasion should be a contested single interception check (interception-evasion = interception chance) rather than how it is now where a successful evasion check trumps a successful interception (each check is made on its own % chance).

One other thing that isn't yet included is invisibility rules and interception. If an air unit is invisible to the intercepting unit, the interception should automatically fail. I'll probably only get into that concern during the air unit review though.

 

This was previously reported right at the end of the v39 cycle, but as far as I can tell, never taken up.

Criminals as espionage agents appears to be broken - instead of adding to your espionage points when a criminal infiltrates a rival's city, it removes the number of espionage points it is supposed to add. Included is my latest save game - I have several thieves in the Hittite capital city, to the southeast of my capital (Washington DC). If I use any of them to infiltrate, it puts my espionage total against the Hittites to -100. (As I suspected what would happen, I saved immediately before testing this.) The SVN version is 11056. Please look into this - if there's an option I've got selected incorrectly, I'd like to get this fixed, as it makes espionage essentially useless at the moment. Thanks in advance.

 

@Thunderbrd can check it, it is easy to replicate it.
You can see that in espionage screen, that you lose 100 espionage points when using that action.

 

I'll try to take a look at this soon.

 

Python error when playing multiplayer hotseat, every time active player changes:

Traceback (most recent call last):
File "BugEventManager", line 307, in _handleDefaultEvent
File "CvStrategyOverlay", line 80, in onSwitchHotSeatPlayer
TypeError: onSwitchHotSeatPlayer() takes exactly 2 arguments (1 given)

Also, camera resets to very close when active player changes.

 

Resolved.

 

Well I put in something that should theoretically fix it, but I don't even know how to start hotseat game let alone test it!

 

Looked at it, the fix is good.

 

Just setup a multiplayer game with the hotseat option. It doesn't require multiple computers for this type of MP game. Quite a fun way to play if you only have one screen actually.

 

First off I'd like to thank ya'll for developing this mod!

Some stuff I've noticed using the following build:
SVN-11056
v40.0.625-alpha - 2019-10-15

1. The great Holmegaard hunt doesn't give the advertised hunter II promotion.
2. AI related: In my last game one civilisation (the Mongolians) recruited mostly healers, storytellers (13!) and enforcers in the prehistoric era. Since I haven't delved too deep I might be unaware of some of the context the AI operates in, but assuming these incur the same costs as the player this must have hampered their development since this was all before they could even found more cities. I noticed they produced wealth with their capital at times, so it does appear to have hampered them, however, until I destroyed them they were doing relatively well. (2nd most advanced)
Unfortunately I don't have AI logs, but I can try to find out if one of my previous saves demonstrates this behaviour.
Other civilisations were farther away, so I wasn't as much aware of their unit roster until later on. As noted elsewhere, I have also noticed they tend to recruit a lot of dogs, but those don't incur extra costs and for the rest their roster is a lot more balanced.

General feedback on balance: (not a bug necessarily)
3. Playing on emperor I find the difficulty a bit tame compared to vanilla. (unless I'd get lucky with the starting position I'd mostly be struggling to keep up in vanilla on that difficulty level)
4. I find that the economy is very forgiving and doesn't force the player to balance tech progress and the speed of expansion. (I have never needed to adjust the 100% tech rate) This could of course be a design choice, but somehow it feels ahistoric to have 30.000 in reserve in the ancient era. In vanilla, I think it's exactly that which prevents players from cruising by the AI opponents. (that and of course the fact that they share most of their research with each other)

 

Please list your Game Set up Options. This behavior sounds like a typical Emperor game with Revolutions On. How Many AI and what sized map used?

Also with 942 Techs spread over 15 Eras comparing Vanilla to C2C game play is just a little skewed. And yeah it is a bit easier because of the content vs vanilla's sparsity.

Finally Thank You for playing C2C!

 

I actually noticed some similar behavior to this, though with revolutions off. Launched on noble difficulty and turn 1 switched to deity through bug options, shortly after composite tools tech I captured Brazil's city partly because they marched their Neanderthal warrior stack off the city (presumably to go attack mine?) right when my macemen got to their city, and also because they had 17 storytellers. I figured it was because on deity the AI has money for days and they don't really care about the upkeep so there's some value in having persistently high education/low cultural revolt chance, but still seems odd. 10 AI, large map.

 

I may be wrong, but - by starting on Noble the AI only gets the Noble settings. Switching to Deity (via WB) on the next turn does not change the AI settings. They are still on Noble. So you have a bigger advantage.

 

And you are playing with the New Complex Traits Option (per your post in another thread). This changes everything. Because Every Leader is no longer the same as the default Trait Options (the base Traits that we have had for several years). And Complex Traits is still basically a WIP from Thunderbrd. You and others are now providing test games through your usage of them. Expect oddities.

I run test games constantly for the base game options while I do Civic rebalancing. I can not use Complex Traits. And as such I do not see the things you are reporting in my test games. I don't have AI with 17 Story Tellers in 1 of their cities. That was a problem a couple of versions back, v37 and v38. But since billw2015 has joined the Team, in base games with minimum Game Set Up Options, they do not exhibit these tendencies.