RESEARCH: City Battle auto-resolve factors

[Dol Guldur]

In all auto-resolved battles we now know that the secondary hp for a unit can advantage the auto-resolved battle whilst being ignored in the battle mode. This is great because units previously disadvantaged or heavily advantaged in autoresolve can now be balanced out to create more realistic auto-resolved battles.

However, I am interested in this research thread to try to find if there are any factors taken into account in auto-resolved city battles which are not taken into account in field battles - or which are and can be exploited to strengthen the advantage of the defender. Things can be done on the battlemap to advantage the defender. Can this be mirrored in autoresolve?

Some thoughts are:

Does the autoresolve taken into account the EDB values for fortifications?

Does the autoresolve take into account the load/ticks/health etc?

Does the autoresolve take into account siege weapons?

What is the terrain type for cities (defender) considered to be and can it differ from that of the attacker?

I have opened this up in the hope that those who have noted any such factors might post their findings OR their suspicions, or just their general observations.

I will updated this research when I can - if indeed anything can be found.

Thx.

[Flying Pig]

Shouldn't this be in Ludus Magna?

[Dol Guldur]

It's not just about the vanilla game mechanics, but about what can be modded to alter the outcome.

[Bardo]

The use of the secondary hp value seems to be a powerful tool to adjust the autoresolve battles. I have been testing it a bit in our mod and the effect is clearely noticed. I have used the values suggested in your "complete EDU guide", and it seems it does not only change the autoresolve behavior, but the whole behavior of the AI in campaign. I'm almost sure that when you increase the 2nd hit point of all units to 5, all the ai factions become more agressive and they conquer, lose and reconquer the border cities more often. I'm thinking that the AI uses this value also to evaluate the power of his army, and to decide if his army is big enought to start an attack. And I'm wondering if it could also be affecting the choices of the ai while recruiting units.
However, it is going to need many testing before we can really adapt and use it into the mod.

I'm finding very difficult to get the dessired result when large armies of weak units fight against small units of powerful soldiers (like the heroes of our mod), and I'm finding that changing just 1 point of the secondary hp value produces too big effects on the result of the autoresolve (at least in some cases), so I'm also wondering if the secondary defensive value (stat_sec_armour) could also be used to fine adjust the autoresolve behaivor of the infantry units. Do you have tested this stat_sec_armour value?

About city battles, I have no idea about the possible factors, and I'm sorry that I'll not be able to help you (by now) with the ammount of testing needed to estimate these factors, but I would like to say that the time you can spend with this is not wasted time, because I believe this will be a powerful tool to rebalance not only the autoresolved battles, but at the same time the behavior of the ai in the campaign.

[Aradan]

Gate health and arrow ticks/reload time don't seem to have an effect and neither do siege machines (rams, towers etc) in auto-resolve. I'm fairly sure that EDB fortification values don't change anything either. Gonna check if siege weapons (catapults/ballistae) have an effect soon... Judging from the results so far though, it's doubtful.

[Dol Guldur]

Thx. Aradan, for your time and effort here. There must be something built into that autoresolve other than the siegeattacker/defender values for generals....when you say edb fortification values you are referring to the capabilities? Have the actual levels been tested? (wooden wall vs stone wall for example).

[Aradan]

I was referring to the capabilities; comparison of wooden/stone walls in the to-do list as well. But given that the various levels of walls draw info from descr_walls mainly and editions there have no effect, I don't believe anything exciting will happen there. Though I'd love to be surprised.

[Dol Guldur]

I've had no luck as of yet with any factors to harmonize autoresolve battles involving walls with their battlemap counterparts. This is truly a great disparity in the game, even more so than the battlefield AI.

I tried making the city another climate/ground type but of course that makes the whole tile change and so would affect both defender and attacker.

Does anyone know if command values above and below 10 are ignored by the engine in factoring battles? They do seem to exist if you test them with traits.