Single Player bugs and crashes v38 plus (SVN) - After the 20th of February 2018

got an endless loop,
k I'll look into it soon if my internet comes back. I'm limping along here. If I did debug it I wouldn't be able to update the SVN at the moment.
 
I paid 1000 and something gold in Ancient era to steal the plans in one Neanderthal city. You will never believe what I discovered. These guys must have had cellphones or social networks - I instantly managed to see the whole map, discovered a couple of world natural wonders, and got much better idea of what is around me and how is the world like...

Ok, probably it is not desired effect - please deactivate this function :)
 
I paid 1000 and something gold in Ancient era to steal the plans in one Neanderthal city. You will never believe what I discovered. These guys must have had cellphones or social networks - I instantly managed to see the whole map, discovered a couple of world natural wonders, and got much better idea of what is around me and how is the world like...

Ok, probably it is not desired effect - please deactivate this function :)
I think that originally you couldn't perform that spy mission (or maybe any spy mission iirc) in a barbarian city, so yeah, that needs to be made impossible to be done against any NPC city. I'll take a look at where that's supposed to be filtered out from the function to see if they can do the mission or not and adjust to include all NPCs rather than just barbarians.
 
Something is bit wonky @alberts2

I'm doing Standard/Normal/Emperor AI autoplay and no AI expanded by 200th turn (10% of game).
Everyone were in Prehistoric era.
But soon after 200th turn some AIs started expanding.

I feel like research penalties in civics are bit too harsh here as if those were set for Immortal or even Deity.
That is it seems like on Immortal or Deity tech leader (AI most likely) should reach Ancient era by 12 000 BC (little under 10% of game).

AIs should try to place few cities as soon as they research tribalism.

I'm using Complex traits and Developing traits with WFL/TD, and with all combat mods.

I'll do another test without those options this time.

By the way could be this pushed to higher value?
Code:
    <!-- AI Gold Slider Safety Percent begin -->
    <!-- This is the BASE amount of slider allocation required towards gold that the AI feels comfortable with -->
    <!-- It noteworthily affects how much tolerance the AI has where it feels ready to grow -->
    <!-- Reduce to make the AI more sensitive to gold concerns and increase for more reckless behavior -->
    <!-- This may be rather interesting to convert to an AI personality matter or at least give a personality based modifier eventually -->
    <!-- Original Value was set to 50 - I've adjusted to 40.-->
    <Define>
        <DefineName>SAFE_GOLD_PERCENT</DefineName>
        <iDefineIntVal>40</iDefineIntVal>
    </Define>
Increasing this may increase will to expand in Prehistoric era.
Its in global defines.

I'm using PPIO and sountrack mods by the way.
 
Last edited:
AIs should try to place few cities as soon as they research tribalism.

I'm using Complex traits and Developing traits with WFL/TD, and with all combat mods.

The AI usally starts to settle soon after researching tribalism. It might take longer as a human player would start to do that but they do.

However the options you are using might cause some issues but i haven't used any of those in a long time so i don't know.
 
The AI usally starts to settle soon after researching tribalism. It might take longer as a human player would start to do that but they do.

However the options you are using might cause some issues but i haven't used any of those in a long time so i don't know.
So I guess research penalties on starting civics are bit too strong - enough that Emperor AI tech leader can't reach Ancient era in time.

AIs started expanding after 200th turn at bit faster rate when I disabled those various options.
 
Last edited:
Increasing this may increase will to expand in Prehistoric era.
Its in global defines.
It may help. Honestly, I think the holdup is probably that the AI is taking a long time to train enough units to accompany their settlers. I don't think its a gold thing in this case. Might need to turn down the minimum escort amount, at least on Size Matters.
 
It may help. Honestly, I think the holdup is probably that the AI is taking a long time to train enough units to accompany their settlers. I don't think its a gold thing in this case. Might need to turn down the minimum escort amount, at least on Size Matters.
On Prince and higher handicaps AI have bigger discount on unit cost than on building cost, which is bigger compared to tech cost to discount.
Dwm 2019-06-21 22-06-27-94.png

So effectively on Immortal AI can build more units and buildings compared to Prince for example.
This relative discount seems to be not enough.

Maybe tribes cost too much compared to other units?

Settler/Chieftain/Warlord exist only for testing meaning Emperor is middle of the road handicap.
On emperor tech leader should regularly reach Ancient era at ~9.7% of game (12000 BC) - calendar structure is based on tech tree.
 
On Prince and higher handicaps AI have bigger discount on unit cost than on building cost, which is bigger compared to tech cost to discount.
View attachment 527246
So effectively on Immortal AI can build more units and buildings compared to Prince for example.
This relative discount seems to be not enough.

Maybe tribes cost too much compared to other units?

Settler/Chieftain/Warlord exist only for testing meaning Emperor is middle of the road handicap.
On emperor tech leader should regularly reach Ancient era at ~9.7% of game (12000 BC) - calendar structure is based on tech tree.
What I'm trying to say is that on SM, they will not just want 2 escorts, they will want 6, 3 merged and 3 merged. After Prehistoric it gets worse, then they'll want 18, 2 double merged and 2 double merged. And that's assuming they only want 2 units for escort now - after the recent fix to the call for a number of units to fill a role by their strength total, they could be asking for more than that. I'm not sure how the math is going but they may be building some seriously hefty stacks just to get a settler out. They may be able to train them faster than the player but that's a lot of delay when a lot of other buildings suddenly come good to construct at tribalism as well. They could also be held up by immediate civic changes - tends to get me as a player, and trying to build all the XP generating buildings first.
 
Something is bit wonky @alberts2

I'm doing Standard/Normal/Emperor AI autoplay and no AI expanded by 200th turn (10% of game).
Everyone were in Prehistoric era.
But soon after 200th turn some AIs started expanding.

I feel like research penalties in civics are bit too harsh here as if those were set for Immortal or even Deity.
That is it seems like on Immortal or Deity tech leader (AI most likely) should reach Ancient era by 12 000 BC (little under 10% of game).

AIs should try to place few cities as soon as they research tribalism.

I'm using Complex traits and Developing traits with WFL/TD, and with all combat mods.

I'll do another test without those options this time.

By the way could be this pushed to higher value?
Code:
    <!-- AI Gold Slider Safety Percent begin -->
    <!-- This is the BASE amount of slider allocation required towards gold that the AI feels comfortable with -->
    <!-- It noteworthily affects how much tolerance the AI has where it feels ready to grow -->
    <!-- Reduce to make the AI more sensitive to gold concerns and increase for more reckless behavior -->
    <!-- This may be rather interesting to convert to an AI personality matter or at least give a personality based modifier eventually -->
    <!-- Original Value was set to 50 - I've adjusted to 40.-->
    <Define>
        <DefineName>SAFE_GOLD_PERCENT</DefineName>
        <iDefineIntVal>40</iDefineIntVal>
    </Define>
Increasing this may increase will to expand in Prehistoric era.
Its in global defines.


I'm using PPIO and soundtrack mods by the way.

This is the value that Pit2015 came up with when he was having problems with the AI expanding. The Discussion with T-brd is still in his Scenario thread. The value was higher "Original Value was set to 50 - I've adjusted to 40. " This is T-brd's comment. So your suggestion raxxo, of upping this value, would re-open this particular area of concern and known problem. But I would say go ahead and put it back to 50 or even as high as 80 and see what happens in your tests.

This whole notion the AI is not expanding fast enough is mainly from players playing the Longer GS and Difficulty levels higher than Emperor. This is Not a problem for players using Blitz thru Long or even Epic. The Players using the Longer GS are impatient with the AI and seem to think the AI should have more cities than the player Before Sed Lifestyle. Because, before we had a 20,000 turn Eternity GS, the AI on Deity would out produce the Player (most of the time) even on the 8,000 Eternity GS that was in place then. And this was basically so, even if SM was used or Not used.

I know some will say this is sour grapes from me, but until we return GS to more logical lengths the GS of 12,000 turns or longer will continue to be a source of this type of problem. It was before when Eternity was 14,000 turns.
 
This is the value that Pit2015 came up with when he was having problems with the AI expanding. The Discussion with T-brd is still in his Scenario thread. The value was higher "Original Value was set to 50 - I've adjusted to 40. " This is T-brd's comment. So your suggestion raxxo, of upping this value, would re-open this particular area of concern and known problem. But I would say go ahead and put it back to 50 or even as high as 80 and see what happens in your tests.

This whole notion the AI is not expanding fast enough is mainly from players playing the Longer GS and Difficulty levels higher than Emperor. This is Not a problem for players using Blitz thru Long or even Epic. The Players using the Longer GS are impatient with the AI and seem to think the AI should have more cities than the player Before Sed Lifestyle. Because, before we had a 20,000 turn Eternity GS, the AI on Deity would out produce the Player (most of the time) even on the 8,000 Eternity GS that was in place then. And this was basically so, even if SM was used or Not used.

I know some will say this is sour grapes from me, but until we return GS to more logical lengths the GS of 12,000 turns or longer will continue to be a source of this type of problem. It was before when Eternity was 14,000 turns.
So increasing this value can backfire then.
But on other hand AI builds more buildings now.

I was testing on Normal game speed (2000) turns.
Tech leader should be in Ancient era somewhere by 9.7 (+-1)% of game.
So tech leader would reach Tribalism sooner than that - there is 98 techs in Prehistoric, and you need to research 58 techs to reach Tribalism - that is 60% of Prehistoric era must be researched.

That is tech leader should reach Tribalism by 6% - 7% of game (60 - 70 turn on Blitz, 120 - 140 turn on Normal and so on in typical game).
Probably this condition is met on Immortal or Deity though.
By 9.7% of game passed (~100th/200th turn on Blitz/Normal) typical game on Emperor handicap should have tech leader in Ancient era, and each civ should have few cities.
That is current balance seems to be set for Immortal or Deity handicap.

I think you could reduce research/commerce maluses by tiniest bit on starter civics.

Calendar is based on distribution of techs in eras, so amount of turns spent per tech should be relatively unchanging trough whole game.

TL;DR on calibrated game:
Emperor game on Normal speed.
Tech leader reaches Tribalism by 60th - 70th turn (or even 80th turn if it stalls this tech to research other prehistoric techs).
Tech leader builds few cities after that.
By 100th turn tech leader is in Ancient era, and other civs reach ancient era soon after.

Lower handicaps have it slower while higher ones are faster with milestones.

With that gold setting of 75% one AI founded city before 10% of game (default setting caused AI to expand for first time well after 200th turn on Normal), and a lot of them reached Tribalism at that stage of game on Normal/Emperor/Standard.
That one AI is about to reach Sedentary Lifestyle, so its bit late.
Traits, combat mods, TD/WFL are disabled on this test.

Gold setting at 100 (most reckless) caused 5 AIs to expand (there are 7 civs, but I don't count it on AI autopay) by 200th turn. No one was in Ancient era though.
High gold setting caused problems in past probably because AI was worse, was building less buildings and units before getting to next era.
Now everyone can build all buildable buildings in cities easily.
It was set back in 2014 last time meaning a lot of stuff was changed and fixed since then.

600/630 =~ 5% lower research costs.
Same effect without messing with civic balance.

They all built city or two and had ready tribe to build yet another city by 10% of game, one of AIs was in Ancient era.
 
Last edited:
So increasing this value can backfire then.
But on other hand AI builds more buildings now.

I was testing on Normal game speed (2000) turns.
Tech leader should be in Ancient era somewhere by 9.7 (+-1)% of game.
So tech leader would reach Tribalism sooner than that - there is 98 techs in Prehistoric, and you need to research 58 techs to reach Tribalism - that is 60% of Prehistoric era must be researched.

That is tech leader should reach Tribalism by 6% - 7% of game (60 - 70 turn on Blitz, 120 - 140 turn on Normal and so on in typical game).
Probably this condition is met on Immortal or Deity though.
By 9.7% of game passed (~100th/200th turn on Blitz/Normal) typical game on Emperor handicap should have tech leader in Ancient era, and each civ should have few cities.
That is current balance seems to be set for Immortal or Deity handicap.

I think you could reduce research/commerce maluses by tiniest bit on starter civics.

Calendar is based on distribution of techs in eras, so amount of turns spent per tech should be relatively unchanging trough whole game.

TL;DR on calibrated game:
Emperor game on Normal speed.
Tech leader reaches Tribalism by 60th - 70th turn (or even 80th turn if it stalls this tech to research other prehistoric techs).
Tech leader builds few cities after that.
By 100th turn tech leader is in Ancient era, and other civs reach ancient era soon after.

Lower handicaps have it slower while higher ones are faster with milestones.

With that gold setting of 75% one AI founded city before 10% of game (default setting caused AI to expand for first time well after 200th turn on Normal), and a lot of them reached Tribalism at that stage of game on Normal/Emperor/Standard.
That one AI is about to reach Sedentary Lifestyle, so its bit late.
Traits, combat mods, TD/WFL are disabled on this test.

Gold setting at 100 (most reckless) caused 5 AIs to expand (there are 7 civs, but I don't count it on AI autopay) by 200th turn. No one was in Ancient era though.
High gold setting caused problems in past probably because AI was worse, was building less buildings and units before getting to next era.
Now everyone can build all buildable buildings in cities easily.
It was set back in 2014 last time meaning a lot of stuff was changed and fixed since then.

600/630 =~ 5% lower research costs.
Same effect without messing with civic balance.

They all built city or two and had ready tribe to build yet another city by 10% of game, one of AIs was in Ancient era.
I wouldn't put it past 80%. The AI does very poorly at dealing with stretching their budget too far. If it really is having a significant effect, that's usually an indicator that we now have gold in a pretty challenging situation during play, which player deal with better than AI. What 80% means really is that the AI will still grow even if they must be at 20% research to stay in the positive on gold income. You can see how having this set too high can allow them to grow but kill their economy in the end. Put it at 100% and you're probably sure to mess them up and throw them into strike as soon as they plant that city it allows them to. What may be problematic for them is that they do seem to lately be trying to grow absolutely ginormous unit stacks and could be paying way too much in unit upkeep for their economy. It's going to take a lot of testing to see what's all really afoot I think but for now we can work with this and see how it goes. What did you actually set it to on the SVN?
 
I wouldn't put it past 80%. The AI does very poorly at dealing with stretching their budget too far. If it really is having a significant effect, that's usually an indicator that we now have gold in a pretty challenging situation during play, which player deal with better than AI. What 80% means really is that the AI will still grow even if they must be at 20% research to stay in the positive on gold income. You can see how having this set too high can allow them to grow but kill their economy in the end. Put it at 100% and you're probably sure to mess them up and throw them into strike as soon as they plant that city it allows them to. What may be problematic for them is that they do seem to lately be trying to grow absolutely ginormous unit stacks and could be paying way too much in unit upkeep for their economy. It's going to take a lot of testing to see what's all really afoot I think but for now we can work with this and see how it goes. What did you actually set it to on the SVN?
I set that to 100.

If you press ctrl+z with chipotle on you will reveal whole map, and you can see their gold if you go into their cities.
They all had positive income for 200 turns of testing game on Normal/Standard/Emperor.
Also they had research at >80% most of time.
They had 1000 - 2000 gold by 200th turn.
Looks like city maintenance is 10x lower than building maintenance at that stage of game.
AI seems to build all buildings, that can be built in cities.

Didn't Toffer or someone else reduce maintenance artificially in code, if there is less than 3 cities?

I guess if problems are here, then they happen after AI leaves Prehistoric era.
But then again Trade tech is powerful gold booster in Ancient era.

Got random crash in last test.
Ignore my civ as in autoplay (same thing as if I was alone on island) players civ isn't treated as AI with Noble setting during autoplay unlike actual AIs.

On 300th turn AIs had 3000 - 15000 total gold and positive income. All of them are in Ancient era.
One of them had trouble with disease in single city.

Got another crash. It is repeatable crash.
 

Attachments

  • minudump and save.rar
    1.2 MB · Views: 176
  • minidump and save after 300th turn.rar
    1.6 MB · Views: 161
Last edited:
@raxo2222
I set that to 100.
Why did you do this during the freeze? This is something that should only have been done After the Release. It's stuff like this that totally screws up a Release. Bad form fellow modder, bad form.

EDIT: and My suggestion was for you to test for awhile. Not add it to the SVN after one simple test game! Crickey talk about jumping the gun. :thumbsdown::cringe:
 
@raxo2222
Why did you do this during the freeze? This is something that should only have been done After the Release. It's stuff like this that totally screws up a Release. Bad form fellow modder, bad form.

EDIT: and My suggestion was for you to test for awhile. Not add it to the SVN after one simple test game! Crickey talk about jumping the gun. :thumbsdown::cringe:
Well so others could test that too - it always can be reverted before release.

Its very easy to hoard gold (AIs had a lot of gold and positive income after 100th turn on Normal and they found first city closer to 200th turn), I guess only during war it could be harmful.

Apparently Immortal AI performed worse with that setting - no idea why.
As if something was set backwards or was overflowing.

I guess I'll revert it.
Immortal on Gigantic expanded less than Emperor on Standard, both on Normal speed.
That change could be innocent here though.

Note that Gigantic sized map, that is generated (Scenarios can have any size tag attached to it) is valid size to play, as it was originally Huge size in C2C_World.
 
Last edited:
2 city name TXT errors:
Looks like @Toffer90 messed up something.

Native American names are messed up too.
Spoiler :

Civ4BeyondSword 2019-06-23 16-40-55-81.png


This game seems to be from April - its last time file was changed.
Those two names were added as backward compatibility now.
 
Last edited:
Upload save, I'll look around saves, that I already collected.

Pits scenario should be goldmine for those :p
just did that, that where i spotted all these, but thx.. . next time i get a new map with each civ, i will let u know, b ut THX. . . .Jeff
 
@raxo2222
Why did you do this during the freeze? This is something that should only have been done After the Release. It's stuff like this that totally screws up a Release. Bad form fellow modder, bad form.

EDIT: and My suggestion was for you to test for awhile. Not add it to the SVN after one simple test game! Crickey talk about jumping the gun. :thumbsdown::cringe:
I agree.

If you press ctrl+z with chipotle on you will reveal whole map, and you can see their gold if you go into their cities.
They all had positive income for 200 turns of testing game on Normal/Standard/Emperor.
Also they had research at >80% most of time.
They had 1000 - 2000 gold by 200th turn.
If that's the case, then in the games you're testing you're seeing inconsistent results because it wasn't a factor to begin with. I suspected it wasn't, to be honest. If they are doing fine with gold, that 40 or 50 on that global won't make any difference. There are other reasons for some civs to be more reluctant to expand, local threat assessments, war plans, property control, etc... and as stated, the potential for holdup in training enough escorts seems to be the last known reason, so it may depend on how good the production in their capital is more than anything.

I know Joseph was talking about faster gamespeeds not seeming to have the issue but I think his experience may be more non-SM based rather than based on a faster GS. In fact, it's probably more likely that a faster GS would experience AI expansion delay than a longer one, but both can see the issue. I was responding some time not so far back to some complaints about delays to grow that were coming from a Blitz speed game.
 
Top Bottom