Diplomacy Developments

[Revenant342]

The thing with terrain bonuses is the AI should know how to use them or it just becomes random chance. Say an infantry army sees a stronger cavalry-heavy army coming towards it and tries to run. To make use of the bonuses, it would need to find some forest nearby to run to, then calculate if it can reach there before the cav army catches it. But what if when it gets there, the calculation tells the cav army that the infantry army is now stronger? Does it engage anyway, or run off?

Right now it doesn't seem like the AI even takes terrain into account for pathfinding, they'll walk straight through a clump of forest instead of going around it, so getting them to intelligently pick terrain for combat might be a much bigger job than people think. Just adding terrain bonuses without overhauling the AI at the same time could still be a short-term fix, but could have unintended consequences.

Plus, with the Khuzait speed advantage, they can still control when and where to fight, so if the calculation told them the enemy army was suddenly stronger than them because it's in a forest, they could choose not to engage just off of numbers even if they don't really understand what a forest is.

Worst-case unintended consequence with the current AI: Infantry army enters forest, becomes stronger than cavalry army. Cavalry army runs. Infantry army sees it is stronger than cavalry army and chases it. Infantry army leaves forest and is suddenly weaker than cavalry army. Cavalry chases infantry. Infantry runs into forest... You get the idea, lol.

 

[froggyluv]

Has it on BL's engine? The only auto calc mod I know of so far for Bannerlord is this one, which just increases the power difference between troop tiers. If you say it was done in a mod for the OG M&B I'll take your word for it because I don't know how that game functioned under the hood.

Its merely spreadsheets of items/terrain type/Unit Type -this really has nothing to do with the Engine and could theoretically be done on any engine. Theres nothing inherent about those types of calculations that would make them engine dependent -they are more of a time sink than anything else. Again I can see your giving them the benefit of the doubt -dont get me wrong i very much admire and appreciate their games but its not uncommon for a Dev team to just never advance a glaring area of their flagship game simply because -they never have before

 

[froggyluv]

The thing with terrain bonuses is the AI should know how to use them or it just becomes random chance. Say an infantry army sees a stronger cavalry-heavy army coming towards it and tries to run. To make use of the bonuses, it would need to find some forest nearby to run to, then calculate if it can reach there before the cav army catches it. But what if when it gets there, the calculation tells the cav army that the infantry army is now stronger? Does it engage anyway, or run off?

IMO any attempt to go in this direction would be a major step forward into an actual interesting Campaign map and strategy game. Giving Generals or Lords even a simplistic "I prefer my men on X terrain over Y" is stunningly different to what we have now. What we have now is like a Pac Man chase scene to Benny Hill soundtrack -there is just no rhyme or reason to their movement except "Run Away!" That shouldnt be acceptable in 2020 gaming

 

[Lord of Banners]

Yeah I guess you could put an artisan in a village. You'd still need to determine how much of each resource to transport back to the village, and some way to get rid of the excess so raw resources don't just stack up in a village's inventory if the rate of production doesn't quite line up.

The towns themselves are what consume the goods using a formula based on prosperity and item prices. Artisans just take raw goods and convert them into manufactured goods.

Is there a way to make cavalry consume double food and test that? I feel like that could make a difference especially with longer siege times and level 3 walls in NE.

 

[froggyluv]

Sure, in many cases it won't matter, but if you want to make these factors powerful enough to make a difference depending on the situation you'll still end up with many edge cases where the aggressor and defender flip in unexpected ways. My example was just one possibility.

Look at historic battles. They are filled with General X ran from General Z until he reached the Marshes (plug favorite terrain here). He then decided to dig in and fight back. He lost. But THAT is a story! We could now have stories whether the Lord makes a good or flawed decision - something happened and for a reason. These chases we have now have ZERO story.

 

[Ananda_The_Destroyer]

Look at historic battles. They are filled with General X ran from General Z until he reached the Marshes (plug favorite terrain here). He then decided to dig in and fight back. He lost. But THAT is a story! We could now have stories whether the Lord makes a good or flawed decision - something happened and for a reason. These chases we have now have ZERO story.

It's banner lord it's like "he ran until it got dark and he lost sight of his pursuers and immediately turned around and went back to raid the village and was again spotted and chased by his enemy."

 

[皮w蛋]

My thoughts on Khuzait:

Plan 1:
I think speed is the biggest factor. But now we can't change the speed of cavalry and delete khuzait's innate skills.

Since Sturgia and Aserai have innate skills equal to zero.

The innate skills of other countries are much weaker than Khuzait in the war.

So can we weaken the innate skills in Khuzait's culture? For example, from a 10% increase in cavalry movement speed to a 5% decrease.

Plan 2:
I heard that the cavalry's combat power is 1.2x when attacking the city. It's the same as field combat.
(please let me know if it has been changed)

This is not in line with human thinking. Changing it may slow Khuzait's occupation of land.

Plan 3:
I heard that in battle calculation, a soldier's combat strength value is determined by the proficiency value of the weapon he uses.This setting is not used by the game
(if not, please let me know. I can't read the code. I heard it from a friend who read the code.)

Khuzait's T2 cavalry is so strong that it has gained an advantage in movement speed and combat computing. Can we change his proficiency, for example, to bring his proficiency closer to that of a T1 recruit.

Plan 4:
A friend of mine put forward the view.

Khuzait has gained too many advantages in the wild. He thought that Khuzait is a nomadic people. Maybe Khuzait's AI can reduce the speed of building camp and building siege tower by 20% ~ 33% in the siege, highlighting the advantages of field operations and giving them the disadvantage of attacking the castle. It can also slow down their occupation of land.

 

[mexxico]

It seems nobody is sleeping today

I am thinking all these stuff for a while too. Instead of nerfing Khuzait bonuses it seems real solution to all these problems can be alliances. Let leave game start unbalanced and if one faction start to be a problem for others (capture their settlements) they (at least two factions) should move together to stop this enemy and after they succeed this they can end their alliance and continue game individual (normal).

However problem is this solution has really work load and need good ai stuff even in map and deciding alliances. Different faction armies and parties will think together and join their map events / sieges / raids they will take war / peace decleration decisions together. However this will be a masterclass solution to all problems and if implemented good game can be even 90+ rating and this will be the real diplomacy game needs.

I do not know what will be decided for next and I will be away for a time (I know this is not a right way of development we should already need to know this) However everybody in company should read this post and all ideas and found results here before deciding next step.

Anyway good night.

 

[Blood Gryphon]

Sorry been a bit busy. Here is a table that compares the 3 patches worth of data mexxico provided.

Some quick take aways from the upcoming hotfix:

• Khuzait still a heavy aggressor declaring war 16-4, Aserai still didn't declare war on Khuzait. That might be something to look at why they wont fight, essentially having an alliance.
• I think this is the sweet spot we are looking for as far as war declarations, seems to be reduced by about around 30% based on my estimate
• The war length has been significantly improved, over 50% of wars longer than a season compared to 37%! Also nice to see Month long wars significantly reduced from 21% to 9% of wars.

Patch	1.5.0	1.5.1 beta	1.5.1 upcoming hotfix
Ending Year	1093​	1091​	1093​
War/Peace Declarations	139​	234 (~292 estimate for 1093)	213​
Avg. Declaration per year	14​	29​	21​
Total Wars	75​	119 (~150 estimate for 1093)	109​
Avg Wars per year	7​	15​	10.9​
Avg Active Wars	7.4​	4.0​	4.4​
Avg. War Length (days)	86​	23​	34​
Min War Length (days)	0​	0​	0​
Max War Length (days)	341​	77​	189​
War Negotiations (0-1 days)	1​	13​	9​
Wars Shorter Than a Month (2-7 days)	2​	25​	10​
Wars Shorter Than a Season (8-21 days)	7​	37​	35​
Wars between a Season and a Year (22 - 84 days)	39​	44​	48​
Wars longer than a Year (84+ days)	26​	0​	7​
% Wars Negotiations	1%​	11%​	8%​
% Wars shorter than a Month	3%​	21%​	9%​
% Wars shorter than a Season	9%​	31%​	32%​
% Wars between a Season and a Year	52%​	37%​	44%​
% Wars longer than a Year	35%​	0%​	6%​
Avg. Daily Tribute	$ 251	$ 1,755	$ 1,228
Min Daily	$ -	$ -	$ -
Max Daily	$ 565	$ 7,910	$ 4,720

Finally updated, i'll have some graphs for 1.5.1 hotfix eventually as well and bring in the 1.5.1 beta graphs from earlier for easy comparison.

 

[Blood Gryphon]

It seems nobody is sleeping today

I am thinking all these stuff for a while too. Instead of nerfing Khuzait bonuses it seems real solution to all these problems can be alliances. Let leave game start unbalanced and if one faction start to be a problem for others (capture their settlements) they should move together to stop this enemy after they succeed this they can end their alliance and continue normal.

However problem is this solution has really work load and need good ai stuff even in map and deciding alliances step. Different faction armies and parties will think together and join their map events / sieges / raids. However this will be a masterclass solution to all problems and if implemented good game can be even 90+ rating and this will be the real diplomacy.

I do not know what will be decided for next and I will be away for a time (I know this is not a right way of development we should already need to know this) However everybody in company should read this post and all ideas and found results here before deciding next step.

I 100% support unbalanced factions if it can be dealt with through alliances. I've often wondered even if the empire is in a civil war, wouldn't they fight together in the mean time for survival if necessary? Also Khuzait and Aserai already have an odd relationship similar to an alliance ?

 

[mexxico]

Thanks for analysing latest data and placing all three tests 1.5.0-1.5.1-1.5.1 and hotfix into one table @Blood Gryphon I will examine all results you found.

 

[Bannerman Man]

It seems nobody is sleeping today

Sleep is for the weak mexxico, you know that! All sounds good. Enjoy your time away!

Is there a way to make cavalry consume double food and test that? I feel like that could make a difference especially with longer siege times and level 3 walls in NE.

Yeah, that probably wouldn't be too difficult. Might make for an interesting test. You'd want to look in the CalculateDailyFoodConsumptionf method as a starting point (that's where the amount of food a party consumes is calculated).

-this really has nothing to do with the Engine and could theoretically be done on any engine.

Sorry, I just wasn't sure what you meant by 'in this engine,' and took it to mean in Bannerlord.

He then decided to dig in and fight back. He lost. But THAT is a story!

It's the deciding that's the tricky part though. In my opinion it would just lead to more erratic behavior rather than emergent gameplay.

 

[Blood Gryphon]

Thanks for analysing latest data @Blood Gryphon I will examine all results you found.

Enjoy your break, like I said before it's much deserved!

 

[Apocal]

Plan 3:
I heard that in battle calculation, a soldier's combat strength value is determined by the proficiency value of the weapon he uses.
(if not, please let me know. I can't read the code. I heard it from a friend who read the code.)

Khuzait's T2 cavalry is so strong that it has gained an advantage in movement speed and combat computing. Can we change his proficiency, for example, to bring his proficiency closer to that of a T1 recruit.

This part is not true.

The autocalc only uses the simplified GetPower method, which considers troop tier with a possible cavalry bonus of 20%. There is a more detailed GetSimulationAttackPower but the autocalc does not use this method. On a quick skim, it doesn't look like anything does...?

 

[皮w蛋]

Thanks for analysing latest data and placing all three tests 1.5.0-1.5.1-1.5.1 and hotfix into one table @Blood Gryphon I will examine all results you found.

I hope you're already sleeping. Physical sleep can affect mental stress, as we all know. Have a good sleep.

This part is not true.

The autocalc only uses the simplified GetPower method, which considers troop tier with a possible cavalry bonus of 20%. There is a more detailed GetSimulationAttackPower but the autocalc does not use this method. On a quick skim, it doesn't look like anything does...?

Thank you

Are combat calculations now based on simple soldier ranks?

In short, a soldier of level 6 has a combat effectiveness of 6; a soldier of level 21 has a combat effectiveness higher than level 16, and a soldier of level 26 is higher than level 21.

If I make minor adjustments, such as raising a soldier's rank from level 21 to level 22, will his automatic combat calculation also increase slightly?

Do you have any comments on some of the other parts of the khuzait intensity that I have raised?

Recently, I was planning a balance modification of the types of soldiers. I wanted to make the strength of soldiers in different countries and different types more reasonable by modifying the proficiency, equipment and level of the arms. Thank you for your information.

 

[Apocal]

Thank you

Are combat calculations now based on simple soldier ranks?

In short, a soldier of level 6 has a combat effectiveness of 6; a soldier of level 21 has a combat effectiveness higher than level 16, and a soldier of level 26 is higher than level 21.

If I make minor adjustments, such as raising a soldier's rank from level 21 to level 22, will his automatic combat calculation also increase slightly?

The GetPower formula is actually very simple and you can trust this because even I understand it. (Mostly)

Shifting by level is how it works for named characters ("HeroObject"). It takes their level, divides by four and adds one. That's their tier, for purposes of autocalc. Unfortunately for your idea, tier can only be an integer, so it any fraction gets rounded away. So it won't have any effect in some cases.

edit:

Do you have any comments on some of the other parts of the khuzait intensity that I have raised?

Plan 1, I think a modest nerf to the Khuzait culture bonus would work but I disagree that the Aserai bonus is zero. I actually think it is incredibly powerful but most people don't know how to play with trade and caravans so it comes across as weaker than it actually is.

Plan 2, I doubt matters particularly much. The AI only besieges and assaults when it has an overabundance of advantage, which is why the attackers virtually never lose when facing the garrison + militia alone.

 

[Revenant342]

I just thought of something: If spearmen are given an innate bonus against cavalry in autocalc, we might not need to nerf Khuzait at all.

All factions field a certain number of spearmen already which will help, but they could prefer making spearmen if they border Khuzait and Khuzait has a higher power level then them.

Even without that, just the bonus alone would make it much harder to take settlements because militia infantry are *all* spearmen by default. Khuzait would still be strong in the field but struggle in sieges.

 

[froggyluv]

It's the deciding that's the tricky part though. In my opinion it would just lead to more erratic behavior rather than emergent gameplay.

Look at the new Battlefield feature in which Allied AI commander now yell out distinct orders ie "Cav Falnk Left" -a very cool effect imo. Is it always the smartest order? Would just Charging have been as if not more effective? Possibly. But thats not the point. The point is they are attempting to elevate both the narrative and tactical portion of the battlefield and give it illustrious cohesion. Sometimes it may feel emergent, other times erratic -but its a stepping stone towards something better and away from the bland mindlessness.

Now apply that to the Strategic map.

 

[Wolfmensch007]

Thanks for analysing latest data and placing all three tests 1.5.0-1.5.1-1.5.1 and hotfix into one table @Blood Gryphon I will examine all results you found.

Wow you are actually listening to Features suggested by community ans try to implement them wich makes me actually like you and rethink about some threads i made.

 

[Apocal]

Wow you are actually listening to Features suggested by community ans try to implement them wich makes me actually like you and rethink about some threads i made.

They've been doing this since release.