Todo:
- Strategic diplomacy update (experimental; aiming to finish for next version); continue to make adjustments & improvements until diplo AI is a work of art
- Debug option to always display a leader's true approach
- Advanced debug logging tool for approach calculation (can be enabled by user by modifying an SQL file, logs every step of the approach calculation and the weight changes throughout)
- Examine deal AI logic to get a full understanding, look for opportunities to improve
- Examine league (voting) AI logic to get a full understanding, look for opportunities to improve
- Tweak league AI for sphere of influence resolution
- Backstabbing penalties rework
- Military aggression warning rework
- Full sweep of all English diplomacy dialogue: correct grammar, typos, spacing; get an understanding of how diplomacy dialogue works
- Learn how to make new dialogue messages
- Modify dialogue logic to increase randomness/variety (currently I find the same messages tend to show up a lot in some cases; leaders will tend to always say the same thing if you DoW on them as a warmonger)
- Special dialogue project :)

Looks like I have my work cut out for me, haha.
Unless some kind soul can manage to gift me some plutonium, a new version adorned with your improvements will certainly be the greatest gift for Supreme Leader during this holiday season.
 
Unless some kind soul can manage to gift me some plutonium,

:shifty:
a new version adorned with your improvements will certainly be the greatest gift for Supreme Leader during this holiday season.

The likelihood I will finish everything during the holiday season is pretty slim, but the strategic diplomacy update should be soon. :)

Be very careful with it. WC logic is extremely sensitive.

G

I swear on my life, I will be extraordinarily careful.
Now what value to change...hmm, eeny, meeny, miny, moe...;)
 
- Full sweep of all English diplomacy dialogue: correct grammar, typos, spacing; get an understanding of how diplomacy dialogue works
- Learn how to make new dialogue messages
- Modify dialogue logic to increase randomness/variety (currently I find the same messages tend to show up a lot in some cases; leaders will tend to always say the same thing if you DoW on them as a warmonger)

Diplomacy Responses Extended is pretty good at adding some variety although it's by no means perfect.
 
Diplomacy Responses Extended is pretty good at adding some variety although it's by no means perfect.

The problem with DRE, with no disrespect intended towards the mod author, is that many of the dialogue lines are taken from other mods. While credit is given and there's no problem with it as a modmod, integrating the dialogue from there into VP would be potentially problematic/messy for copyright & fairness reasons.

I do have something in mind for the future, though. :)
 
The problem with DRE, with no disrespect intended towards the mod author, is that many of the dialogue lines are taken from other mods. While credit is given and there's no problem with it as a modmod, integrating the dialogue from there into VP would be potentially problematic/messy for copyright & fairness reasons.

I do have something in mind for the future, though. :)

I thought you forgot about DRE...

G
 
I thought you forgot about DRE...

G

Nonsense! I brought it up before, after all. :)

https://github.com/LoneGazebo/Community-Patch-DLL/issues/5657

Also, at least one issue I see with some of the dialogue wouldn't be solved by DRE; the AI tends to highly prioritize the "DoW'ed by warmonger" message over other messages, and so when you declare war, a lot of the other potential things they could say never appear, which I find boring. I think it'd be better if that was one of the options for them to say and others (including leader-specific messages) are also possible, with some kind of weighted random choice.
 
Last edited:
Nonsense! I brought it up before, after all. :)

https://github.com/LoneGazebo/Community-Patch-DLL/issues/5657

Also, at least one issue I see with some of the dialogue wouldn't be solved by DRE; the AI tends to highly prioritize the "DoW'ed by warmonger" message over other messages, and so when you declare war, a lot of the other potential things they could say never appear, which I find boring. I think it'd be better if that was one of the options for them to say and others (including leader-specific messages) are also possible, with some kind of weighted random choice.

Ahem
 
Finished a one city challenge, Arabia, King, Epic, Planet simulator continents (2+2+4 civ on continents) on 12-1. Authority Carthage and I were on the same continent. Did not discover Netherlands and Mayans until medieval. Discovered the rest in renaissance.

Got one DoF offer from a sanctioned China (pretty late) but she was hated by everyone and no trades anyway, so I refused. China and I went freedom. Took until Venice opting for freedom to get a second DoF. Could be an outlier due to having a small military because of one city.

Global peace accords seems to do nothing in terms of wars being declared despite the description saying greatly increased warmonger scores for BOTH declaring wars and capturing cities. I got a -200 modifier against Carthage for removing 2 of her cities which then decayed to -100. Netherlands, with tradition, statecraft, industry had an impressive navy, was declaring on everyone and snagged 2 Chinese island cities and 1 Mayan city even when he had no rewards for conquest. Spain, China and Inca were at war for most of the game. China vassalised Inca. Spain took a city or two from China. Inca managed to get a city back after Global liberation. Carthage was bleeding gold and even with negative GPT before the war, she declared on Netherlands. Global peace accords does little to impact AIs but can really down a player with the proposal coming online around Industrial when everyone has gold problems.

I was able to trade votes with just about everyone. I had no warmonger diplo penalties with anyone except Carthage and could even trade with her. 10GPT for 1 delegates and 20 for 2 seems is in line with what the AI offers when buying votes.
Spoiler :

Sid Meier's Civilization V (DX11) 12_11_2019 2_25_20 PM.png


Sid Meier's Civilization V (DX11) 12_11_2019 2_22_34 PM.png



Being refused a vote for any proposal felt quite alright. (The leader is feeling self righteous at the moment)

Victory block diplo penalty (Your behaviour infuriates them) was not removed a couple turns after I won.

More dialogue please. Playing one city challenge and AI dialogue be like
Spoiler :

Sid Meier's Civilization V (DX11) 12_11_2019 2_22_46 PM.png

 
Last edited:
Victory block diplo penalty (Your behaviour infuriates them) was not removed a couple turns after I won.

That's upcoming, not in the game yet. :)
 
In the upcoming version I see : "Increased AI war/hostility weight for border disputes in the early game"

I find it already very hard NOT to be at war with your close neighbors. Do you really think it is necessary to increase that weight?
 
In the upcoming version I see : "Increased AI war/hostility weight for border disputes in the early game"

I find it already very hard NOT to be at war with your close neighbors. Do you really think it is necessary to increase that weight?

The increase is not that significant compared to how it is already (except for Montezuma, who is pretty much guaranteed to declare war early and often because that's his thing).

Note that I also added extra weight for friendliness towards distant players during the same time, so you should be able to get some support from those quarters.

Land dispute penalties are based on contested tiles (if they don't want lands you own, they won't care) and warmongering penalties are lesser in the early eras, plus it's advantageous for the AI to take early game land and conquer opponents before they can snowball. This is especially the case now that early game Settler spam is more difficult. I don't see it as poor strategic thinking for the AI.

The (usually correct) reasoning is that in the early game, the AI is not likely to make good friends with their neighbors, so making friends with people far away that they aren't competing with for territory is a smart and mutually beneficial idea.

The AI was, however, adding competitor weight to players who are at war with their neighbors without checking the circumstances, which was probably making it worse; I've edited this check to be more nuanced for the next version.

The AI can be pacified significantly by building up a strong military.

Other than that, if it's too difficult for you to handle, you may need to drop down a difficulty level?

I appreciate the feedback, but I haven't seen a general consensus that the AI is too aggressive to its own detriment in the early game. If anyone has any thoughts on this, please share!
 
Last edited:
Finished a one city challenge, Arabia, King, Epic, Planet simulator continents (2+2+4 civ on continents) on 12-1. Authority Carthage and I were on the same continent. Did not discover Netherlands and Mayans until medieval. Discovered the rest in renaissance.

Got one DoF offer from a sanctioned China (pretty late) but she was hated by everyone and no trades anyway, so I refused. China and I went freedom. Took until Venice opting for freedom to get a second DoF. Could be an outlier due to having a small military because of one city.

Global peace accords seems to do nothing in terms of wars being declared despite the description saying greatly increased warmonger scores for BOTH declaring wars and capturing cities. I got a -200 modifier against Carthage for removing 2 of her cities which then decayed to -100. Netherlands, with tradition, statecraft, industry had an impressive navy, was declaring on everyone and snagged 2 Chinese island cities and 1 Mayan city even when he had no rewards for conquest. Spain, China and Inca were at war for most of the game. China vassalised Inca. Spain took a city or two from China. Inca managed to get a city back after Global liberation. Carthage was bleeding gold and even with negative GPT before the war, she declared on Netherlands. Global peace accords does little to impact AIs but can really down a player with the proposal coming online around Industrial when everyone has gold problems.

I was able to trade votes with just about everyone. I had no warmonger diplo penalties with anyone except Carthage and could even trade with her. 10GPT for 1 delegates and 20 for 2 seems is in line with what the AI offers when buying votes.


Being refused a vote for any proposal felt quite alright. (The leader is feeling self righteous at the moment)

Victory block diplo penalty (Your behaviour infuriates them) was not removed a couple turns after I won.

More dialogue please. Playing one city challenge and AI dialogue be like

The AI is programmed to prefer friends and DPs that are at least of average military strength. If you have only one city friendship is therefore not likely.

Global Peace Accords does reduce the AI's likelihood to go to war, but for AIs that are doing well on Immortal difficulty, it's unlikely to have much of an effect (they get a unit maintenance discount). The AI is intended to be very aggressive on the top two difficulties.

Re: dialogue - I have to say, it's a fitting taunt, but I intend to work on the dialogue stuff to add more variety to it at some point.
 
Global Peace Accords does reduce the AI's likelihood to go to war, but for AIs that are doing well on Immortal difficulty, it's unlikely to have much of an effect (they get a unit maintenance discount)

Definitely explains why only Netherlands was declaring on me repeatedly. He was making 350GPT while the others were losing money, had little income or Venice. I was surprised more with how he was declaring on everyone when he should be friendly (how I thought I would play Netherlands) and trading his 350GPT for luxuries.

Was playing on king not immortal (just to keep in mind). I do intend to get there eventually.
 
We need more of this.

I think you'll be happy if you play against India in the next version, then.

Failure to keep Gandhi away from the Uranium can have...interesting consequences. :)

Definitely explains why only Netherlands was declaring on me repeatedly. He was making 350GPT while the others were losing money, had little income or Venice. I was surprised more with how he was declaring on everyone when he should be friendly (how I thought I would play Netherlands) and trading his 350GPT for luxuries.

Was playing on king not immortal (just to keep in mind). I do intend to get there eventually.

Oh, I misread, my bad. Netherlands are supposed to have strong weight when it comes to trading, I adjusted this for next version. Perhaps the others weren't being very friendly to them, though.
 
Last edited:
Something interesting just happened to me. I demanded that Morocco accept my embassy. The trade value said impossible, but he still accepted. It's probably not a bug, just the "impossible" is misleading.
 
One thing I will still see is that sometimes I offer the AI open borders and the deal value immediately shifts to "Impossible". As far as I know there shouldn't be any reason for that, giving them my open borders is a pure benefit (unless they have checks around archeology to not allow open borders).
 
One thing I will still see is that sometimes I offer the AI open borders and the deal value immediately shifts to "Impossible". As far as I know there shouldn't be any reason for that, giving them my open borders is a pure benefit (unless they have checks around archeology to not allow open borders).
Your open borders increases your tourism bonus, the AI doesn't want that if you're getting too close to CV.
 
Top Bottom