Auto-calc balance

[Aradan]

I'm interested to know: What does everybody think of the balance of the auto-resolve function regarding unit types? Do you think that the strengths of units are on par with their average capabilities on battle-map?

For example there are opinions that horsemen are generally underpowered and take too many casualties, archers also suffer a tad too much and chariots own everything, compared to how well the can do in the hands of the AI or an average player on battle-map. There are also some contradictory opinions on horse-archers, some suggesting that they are overpowered in auto-resolve (esp the heavy ones) while other suggesting the opposite.

These are just opinions so far, so I'd like to know what everyone thinks on the matter...

[GamblerTuba]

I agree with all the points you made and will add a few. The autocalc seems to work on the assumption of an all out charge by the attacker. So archers in an attacking army don't seem to help much. Especially annoying against bandits.

Recently autocalced a battle as ERE. 2 bandit peasants against a family member, 2 legio lanciarii, and 2 eastern archers. I lost 20 men. Obviously, under human control, I would have just let the archers go to town and would have lost none.

There is at least one situation where autocalc can be your friend. Had a situation as Sassanids where A Roman general was under siege basically as the only defending unit behind some large stone walls. I did not like the idea of dealing with all the towers (no artillery in the army) so i autocalc's and lost only a handful of men. Much less costly than my experience of taking the walls even when they are not defended.

I only autocalc when my army is composed of cheap easily retrained units. Normally against bandits.

[Omanes Alexandrapolites]

In addition to what already has been said, the game seems to fail to take walls into account when considering auto-calculated victories. In many cases battles which I could have won with walls in place seem to act as if it were simply a plain open field. Since my garrisons are usually deliberately small due to the difficulty foes have assaulting stone walls, this is especially noticeable in my campaigns.

~:)

[Aradan]

Thx for the replies, each bit of info is of immeasurable help :) Do you believe that horse-archers are overpowered or underpowered in auto-resolve?

[Omanes Alexandrapolites]

Horse Archers are, in my opinion, underpowered in autoresolve compared to their performance overall on the battlemap. As said, it seems that their ability to attack from a distance and keep out of the way of melee weapons is completely ignored and are treated as if they are simply charging head on into the enemy.

~:)

[Punicus]

Thx for the replies, each bit of info is of immeasurable help :) Do you believe that horse-archers are overpowered or underpowered in auto-resolve?Definitely underpowered. I've rarely won an auto-resolve battle in my current Parthian campaign. As said before it definitely seems as if the AI just makes the two sides collide in hand-to-hand, and we all know that Horse Archer hand-to-hand combat is usually not pretty.

On the contrary, I rarely lose auto-resolve battles against Parthia, if I remember correctly (I haven't played as Seleucids or anyone near them in a while).

[mrdun]

In which case the chariots would not do as much damage as they do. Head on they will take a few and get stuck in hand to hand.

[Aradan]

Some context would be nice regarding the last post, I admit I didn't quit get it its exact meaning.

[Hannibalbarc]

Units with 2hp like bullwarriors seem to do pretty well on autoresolve as well.

[Quirinus]

The autocalc seems to work on the assumption of an all out charge by the attacker.
This is something I've noticed as well. An all-out charge, as well as staying in the fight once battle has commenced, meaning cavalry is undervalued, as well as archers. Horse archers suffer the most-- in situations where horse archers should most definitely have won, like, for example, a battle between three horse archer units and two units of hastati, autocalc will give victory to the hastati army.

There is also one unit type that I see no one has mentioned yet: units with the phalanx ability. The autocalc seems to work on the assumption that the units fight with phalanx turned off. So, where three units of militia hoplites should probably have won against three units of hastati, autocalc will tip the favour against the militia hoplites.

I've noticed the chariot imbalance as well, and, in a Seleucid game I played a long time ago, I conquered the whole of Greece with six scythed chariot units, without a general, when the Greek phalanx should have dismantled the chariots without any difficulty.


Units with 2hp like bullwarriors seem to do pretty well on autoresolve as well.
Wait, bull warriors have two hit points? Whoa.

[mrdun]

Some context would be nice regarding the last post, I admit I didn't quit get it its exact meaning.
I quoted the wrong user, no edit button..




Take this comment by gambler tuba which I should have quoted..


I agree with all the points you made and will add a few. The autocalc seems to work on the assumption of an all out charge by the attacker.



In which case the chariots would not do as much damage as they do. Head on they will take a few and get stuck in hand to hand.
With an all out charge, chariots which do so well in auto-calc basically would get stuck an killed pretty quickly.

[Barbarian]

I have stopped auto-resolving completely, I fight all battles by myself, even if it is just a small clash. I only use autoresolve after modding something, to get somewhere quickly and test new features. However, I have noticed few things too:

1) Horse archers are definitely underpowered. I knew few people, who auto-resolve battles against scythia, because they can't handle their horse archers. If a fight between 10 units of infantry and 10 units of horse archers is autoresolved, the results are the same as if 10 units of infantry fought 10 of infantry in melee.

2) Command stars have even greater effect in auto-resolving, than they have on battlefield.

3) One of the main reasons, why I hate auto-resolving, is that it often evens out the casualties among all units, like not an army who lost one 160 men strong unit, but an army, who's 16 units each lost 10 men. That's far from reality, as in real battle, I would keep my valuable and rare archers in safety, my elite infantry as a reserve, and send some basic recruits and mercenaries to attack first and inflict as much damage as they can, and would probably never have to use my valuable reserves.
This makes a huge difference in battles far away from my main cities. I can't afford to loose few dozens of my best men in each battle, because there is no way to retrain them. But Auto-resolving will never understand it.

4) In A-R, faction members almost always survive. With the FOW turned off, you can see AI factions attacking lonely rival faction members some 20 times, until they finally beat them. The leader dies only when he has some 20-30 men left, and he is attacked by half as stack army.
The situation in real-time battle is much different. AI almost always looses it's leaders, and it happens early in the battle. Fighting Egypt with phalanxes is a good example. I would say that there is 50% chance that enemy's leader will be the first one, who charges his chariot in spears. Try to autoresolve the same battle, and you will never kill their leader.

5) Surely Auto-resolving doesn't count the strength of the city walls. I recently change the files a bit - gave Gauls an epic stone wall around Alesia, and added superb tower defences to the walls: ballista bolts could fire at ANY direction. Then tried to take it with the silver chevron praetorian cohorts.
Gauls had some weak skirmishers and warbands there. In real-time battle I lost heavily due to the superb tower defences. When I auto-resolved the same battle, it was counted as a heroic victory with minimum loses, as my army was much stronger.

BTW, what do you think about elephants in Auto-resolve? I suppose they should be quite powerful, as the battle is counted as an all-out charge :idea2:

[Paradox]

Yeah, I stopped auto-resolving and, instead, used the forward button when in short battles. I too learned that command stars have a VERY great effect on auto-resolving.

[El Diablo]

The problem with A/R and elephants is, as has been previously stated, a percentage of casualties are sperad amonst the whole army.

10% losses means you have lost 10% of your precious ele's.

NEVER AUTO RESOLVE WITH ELEPHANTS!!!! This is particularly important if the are Merc elphants or you have no ability to replenish your unit.

(Mind you I feel the same way about my beloved Cretan Archers - I never AR with them!)

[Stoneface]

The most extreme case with A-R I had was in a test campaign. I just moved the troops into their, cities told all cities to build automaticly and just hit end turn a few times. The Gaul attacked my only continental settlement and I autoresolved. If I'd loose the settlement, so what? I just wanted to see what would happened without an aggressive player. I had only a few troops there but several family members. To my great surprise they won battle after battle. Even a full stack of not to shabby Gaulish troops were destroyed by 4 British family members and 3-4 low level units.

[mrdun]

what fun chariots are

[Quirinus]

I have stopped auto-resolving completely, I fight all battles by myself, even if it is just a small clash. I only use autoresolve after modding something, to get somewhere quickly and test new features.
I do autoresolve a lot when I'm playing the Julii, or Carthage in Iberia. This is because the computer never takes into account the very potent 'warcry' ability. Battles that I couldn't win in many attempts produce a clear victory when autocalc'd. Though that means rotating and ferrying troops back to your core cities a lot, I don't mind overmuch. No matter what critics say about the battlefield AI, I find that it handles basic infantry tactics reasonably well, for instance, activating warcry before charging.


Horse archers are definitely underpowered. I knew few people, who auto-resolve battles against scythia, because they can't handle their horse archers.
I'm one of them..... haha. If I see a large army that is more than one third horse archers, I will definitely auto-calc. Horse archers in small doses I can handle.


Command stars have even greater effect in auto-resolving, than they have on battlefield.
Yeah! Though I guess that makes sense..... better command is supposed to reflect a better understanding of battlefield tactics, so I guess it makes sense on auto-calc that they win against the odds more often.


One of the main reasons, why I hate auto-resolving, is that it often evens out the casualties among all units, like not an army who lost one 160 men strong unit, but an army, who's 16 units each lost 10 men. That's far from reality, as in real battle, I would keep my valuable and rare archers in safety, my elite infantry as a reserve, and send some basic recruits and mercenaries to attack first and inflict as much damage as they can, and would probably never have to use my valuable reserves.
Yeah. What I do, though, is to pool the casualties together. Say, for example, I have a bunch of hastati who all lost less than ten men. I will 'merge' the troops-- as in, using units from one troop to replenish the rest, and then taking that unit (or units) back to the nearest city for retraining. That's why, at any one time, only about two-thirds of my actual army in a particular theatre are in the main army itself. I find that this strategy works pretty well, as you don't put all your eggs in a basket-- at any one time, you're only fighting at 2/3 strength. If the going gets really tough, significant reinforcements are but a couple of turns away.


This makes a huge difference in battles far away from my main cities. I can't afford to loose few dozens of my best men in each battle, because there is no way to retrain them. But Auto-resolving will never understand it.
I agree with this. Which is why, if I happen to have elite units, or rare mercs like Cretans or beserkers, I tend to play the battles myself, or remove the troop from the army if I'm confident enough about my odds without them.


In A-R, faction members almost always survive. With the FOW turned off, you can see AI factions attacking lonely rival faction members some 20 times, until they finally beat them. The leader dies only when he has some 20-30 men left, and he is attacked by half as stack army.
The situation in real-time battle is much different. AI almost always looses it's leaders, and it happens early in the battle. Fighting Egypt with phalanxes is a good example. I would say that there is 50% chance that enemy's leader will be the first one, who charges his chariot in spears. Try to autoresolve the same battle, and you will never kill their leader.
Gods! I couldn't agree more. I play the battles myself if I ever deem it particularly important to prune my enemy's family tree.


BTW, what do you think about elephants in Auto-resolve? I suppose they should be quite powerful, as the battle is counted as an all-out charge :idea2:
Actually, I don't find elephants to be battle deciders in autocalc, like chariots seem to be, although technically elephants should be much better than chariots. Anyone else have any thoughts on this?